FINAL DRAFT

ANALYSIS OF THE INCREMENTAL COST OF PROPOSED REVISIONS
                             TO THE
    NPDES PERMIT AND WATER QUALITY STANDARDS RULES
                            July 30,1999
                            Prepared for:

                   U.S. Environmental Protection Agency
                    Office of Wastewater Management
                          401 M Street, S.W.
                        Washington, D.C.  20460
                            Prepared by:

                          Environomics, Inc.
                    4405 East-West Highway, Suite 307
                         Bethesda, MD20814

                          Under contract to:

                           TetraTech, Inc.
                      EPA Contract No. 68-C7-012

-------
                              TABLE OF CONTENTS

EXECUTIVE SUMMARY .......... ..................................................  1
       Introduction [[[  1
       Summary of Proposed Provisions .................................................  1
       Summary of Estimated Costs  [[[  2
       Organization of This Report ............... ......................................  2

I.  THE COST OF PROPOSED OFFSET REQUIREMENTS ..................................  4
       Proposed Provision for New Dischargers and Significant Expansions ....................  4
       Overview of Cost Implications ......................... < .........................  5
              Costs to Small Entities [[[  5
              Costs to Large Entities [[[  6
              Costs to Permitting Authorities .......................... • ..................  6
              Focus of This Analysis ...... . ...........................................  7
       The Nation's Experience With Trading and Offsets  ..................................  7
              Water Trading [[[  7
              Transactions Costs ........ ..............................................  8
       The Potential Cost of Offsets for Storm Water Dischargers ...........................  10
              The Construction Industry ...............................................  11
              Other Storm Water Dischares ...........................................  24

       The Potential Cost of Offsets for Process Dischares ................................  30

D. PROPOSED SILVICULTURAL AUTHORITIES ...................................... .45
       Proposed Provisions Allowing Some Silvicultural Activities to be Designated as Point Sources
              Subject to NPDES ............................................ .- ........ 45
       Overview of Cost Implications .................................................. 45
       Summary of Types of Costs and Procedures for Estimating Them ...................... 46
       Key Assumptions in Estimating Costs ............................................ 47
       Estimated Costs of the Proposed Silviculture Provisions .............................. 49
       Summary of Impacts of Proposed Silviculture Provisions ............................. 81

III.  PROPOSED AUTHORITIES REGARDING ANIMAL FEEDING OPERATIONS .......... . 83
       Proposed Provisions Allowing EPA to Designate Some Animal Feeding Operations (AFOs) as
              Concentrated Animal Feeding Operations (CAFOs), Thus Making Them Subject to
              NPDES [[[ 83
       Overview of Cost Implications .................................................. 84
       Summary of Types of Costs [[[ 86
       Overview of Procedures for Estimating Costs ...................................... 87
       Key Assumptions in Estimating Costs ............................................ 88
       Estimated Costs of the Proposed AFO Designation Provision .......................... 90
       Summary of Impacts of Proposed AFO Designation Provision  ........................  130


-------
      Does the Proposed Regulation Impose Incremental Costs?	  133*
      Summary of Types of Costs	  133
      Brief Background on Aquatic animal production facilities in the U.S	  134
      Overview of Procedures for Estimating Costs	  137
      Key Assumptions in Estimating Costs	  138
      Estimated Costs of the Proposed AAPF Designation Provision 	  139
      Summary of Impacts of Proposed AAPF Designation Provision	  154

ATTACHMENT 1: SURVEY OF OFFSET ACTIVITIES 	  156

ATTACHMENT 2: CHARACTERISTICS OF OTHER STORM WATER FACILITIES ..../...  157
                                                                       X

ATTACHMENT 3: CHARACTERISTICS OF NEW PROCESS WATER DISCHARGERS FROM
      1995-98	  158,

ATTACHMENT 4: UNIT COST / POUND OF OFFSETS FOR PROCESS WATER DISCHARGERS
      AND CONCENTRATION OF POLLUTANTS AT END-OF-PIPE
        	V...."	:	  159

-------
                                EXECUTIVE SUMMARY

INTRODUCTION

       This report estimates the incremental costs of EPA's proposed revisions to its National Pollutant
Discharge Elimination System (NPDES) permit and water quality standards regulations addressing
several categories of sources. For the purpose of estimating incremental costs, the proposed regulatory
provisions are grouped into four categories:

       I.      offset requirements for new sources and significantly expanded existing sources,

       II.      authority to designate some silvicultural activities as point sources subject'to NPDES
               requirements,

       III.     authority for EPA to designate some animal feeding operations as subject to NPDES
               requirements, and

       IV.     authority for EPA to designate some aquatic animal production facilities as subject to
               NPDES  requirements.

Each of these provisions is summarized below, along with a brief description of the analysis provided in
this report. Following this summary,  we provide an overview of total annual costs associated with these
revisions and further describe the organization of this report.

SUMMARY OF PROPOSED PROVISIONS

I.  Offset requirements for new sources and significantly expanded existing sources.

       This provision applies to new sources and to significantly expanding existing sources that 1)
discharge directly into a  water body that does not meet water quality standards, 2) contribute to the
impairment of the water  body by discharging the pollutants causing the impairment, and 3) are large
entities. New or significantly expanding existing sources that meet all three conditions must provide a
1.5:1 offset for the pollutants of concern.

       The analysis estimates the potential cost to storm water discharges by the construction industry,
to  storm water discharges by other industries, and to dischargers of process waste water.  The analysis
also includes the administrative burden and costs to the private sector as well as to permit authorities.
There are no impacts to small businesses because the offset requirements only apply to large businesses.

II. Authority to designate some silvicultural activities as point sources subject to NPDES
requirements.

       This provision removes the existing exemption for silvicultural activities if EPA determines that
such designation is necessary in connection with the development of a TMDL, or if a State determines
that these activities contribute to violation of WQS or significantly contribute pollutants, independent of
whether or not a TMDL  is involved.

-------
                   «                                f
        The analysis estimates the potential cost of compliance for those silvicultural activities that are
designated, as well as the administrative burden and costs to the private sector and to permit authorities.
The analysis also estimates the impact to small businesses.

m. Authority for EPA to designate some animal feeding operations (AFOs) as subject to NPDES
requirements.

        Under existing NPDES regulations, States (and U.S. EPA in non-authorized States) already have
the authority to designate some AFOs as concentrated AFOs (CAFOs), thus making them subject to
NPDES requirements. This provision gives designation authority to EPA also in these authorized States.

        The analysis estimates the potential cost of compliance for those animal feeding operations
activities that will be designated by EPA under the proposed provision, as well as the administrative
burden and costs to the private sector and to permit authorities. The analysis also estimates the impact to
small businesses.

IV. Authority for EPA to designate some aquatic animal production facilities as subject to NPDES
requirements.

        Under existing NPDES regulations, States (and U.S. EPA in non-authorized states) already have
the authority to designate some aquatic animal production facilities as "concentrated," thus making them
subject to NPDES requirements.  This provision gives designation authority to EPA also in these
authorized States.

        The analysis estimates the potential cost of compliance for those aquacultural activities that will
be designated by EPA under the proposed provision, as well as the administrative burden and costs to the
private sector and to permit authorities. The analysis also estimates the impact to small businesses.

SUMMARY OF ESTIMATED COSTS

        The following table summarizes the results of the analysis. Costs to the private sector include
the capital and operating costs for incremental control measures, as well as the additional administrative
costs associated with new permit-related requirements.  Costs to State and Federal permit authorities
include  the additional permitting and evaluation burdens associated with the proposed revisions.  The
annualized costs shown in the table reflect all  costs projected to be incurred from 2000 onward. The
total annualized cost associated with the four provisions is estimated to be $17.2 - $65.2 million.

ORGANIZATION OF THIS REPORT

        This report is organized according to the four provisions described earlier in this summary.
Section  1. analyzes the offset provisions, Section 2 analyzes the designation provisions for silviculture.
Section  3 analyzes the designation provisions  for animal feeding operations, and Section 4 analyzes the
designation provisions for aquatic animal production facilities.

-------
                                               Exhibit ES-1
                        Summary of Incremental Annual Impacts For All Provisions
                                           (March, 1999 Dollars)
                          Proposed Provision
Annualized Cost
   (S Million)
  # Entities
  Affected
  Annually
Offset Requirements  (does not apply to small businesses)

Cost of storm water offsets:          for the construction industry

                                  for other storm water dischargers

Cost of offsets for new & significantly expanded existing process dischargers

Administrative cost to Federal and State governments

                       • •    -      •                           Subtotal
   8.12-2226  ,

   1.50- 8.64

   1.71 - 10.86

   0.21 - 0.52

  11.54-42.28
  292 - 877

  90 - 405

   4- 18


 386 - 1,300
Designating Silvicultural Operations Under NPDES

Cost for the silviculture industry

Administrative cost to Federal and State governments
                                                             Subtotal
   3.45 - 12.93

   027- 0.28

  3.72 - 13.22
 613 - 1,225
Designating Animal Feeding Operations Under NPDES

Cost for AFOs

Administrative cost to Federal and State governments
                                                             Subtotal
   0.97 - 4.79

  0.034- 0.157

   1.00 - 4.94
   15-71
Designating Aquatic animal production facilities Under NPDES

Cost for the aquaculture industry

Administrative cost to Federal and State governments
                                                             Subtotal
   0.95 - 4.79

  0.001 - 0.007

   0.95 - 4.80
    1-2
                                               Total Annualized Cost
   17.2-65.2
1,015 - 2,598-

-------
             I.  THE COST OF PROPOSED OFFSET REQUIREMENTS

PROPOSED PROVISION FOR NEW DISCHARGERS AND SIGNIFICANT EXPANSIONS

       The proposed regulation (40 CFR § 122.4) prohibits the issuance of a permit to any new or
significantly expanded1 discharger unless they comply with the antidegradation requirements of State
water quality standards applicable to such waters, including the antidegradation provisions adopted
pursuant to 40 CFR §131.12(aX2). 40 CFR §131.12(a)(2) requires dischargers that are not a small
business (as defined in 5 U.S.C. § 601(6)), to provide reasonable further progress toward attainment of
water quality standards by offsetting 1.5 times the amount of any increase in mass  loadings of the
pollutant(s) (however, the Director may, with the concurrence of EPA, determine an offset ratio of less
than 1.5:1,  but more than 1:1, is sufficient to achieve reasonable further progress).  The following
requirements for the offset must be met:

               (a)     the pollutant load reductions must be achieved from a source(s) discharging the
                      pollutant(s) for which the water body is impaired and that  the new or existing
                      discharger undergoing a significant expansion is required to offset;

               (b)     the pollutant load reductions must be achieved from a source(s) located on the
                      same waterbody as the discharge from the new discharger or existing discharger
                      undergoing a significant expansion;

               (c)     the pollutant load reductions are the result of pollutant control measures
                      implemented by, or secured and assured by, the new discharger or existing
                      discharger undergoing a significant expansion;

               (d)     the pollutant load reductions must be achieved on or before the date the
                      discharge commences and remain in place until

                      (i)     a TMDL for the water body is approved or established by EPA, and all
                             dischargers in the water body have effluent limits  or conditions
                             consistent with their waste  load allocations (WLAs)  in the TMDL and
                             other pollutant controls necessary to achieve the load allocations in the
                             TMDL have been implemented, or

                      (ii)     until the discharger ceases to discharge  the pollutant(s) causing the
                             impairment,

               (e)     where a discharger obtains pollutant load reductions from  an existing point
                      source(s), the NPDES permit(s) for the existing point source(s) must be
                      modified to reflect those reductions on or before the date the permit is issued to
                      the discharger: and
        1 Significant expansion means a 20% or greater increase in loadings above the discharger's current permit
limits. While the entire discharge would be subject to 122.4(i), only the expanded portion would be subject to the
offset requirements of 131.12(a)(2).

-------
               (f)     where a discharger obtains pollutant load reductions from an existing nonpoint
                      source(s), the discharger's permit must include any conditions the Director
                      believes are necessary, including the offset requirements and any accompanying
                      monitoring and reporting requirements, to ensure continued achievement of the
                      offset from the nonpoint source(s).

In addition, dischargers that must obtain offsets may need an individual NPDES permit where normally
they have been covered under an NPDES General Permit (such as for storm water).

OVERVIEW OF COST IMPLICATIONS

        New or significantly expanding dischargers will be affected by this'proposed provision if:
1) they wish to discharge a pollutant into a water body that does not meet water quality standards for
that pollutant and 2) there is not a TMDL that provides an allowance for future growth that
accommodates the discharger. Dischargers that are not small entities will be required to obtain 1.5:1
offsets.2 This will result in increased administrative costs to both the dischargers affected and to the
permit authorities, and increased compliance costs for the dischargers.

        The applicability of this requirement is expected to decline over time. As discussed earlier,
offsets must be maintained only until the TMDL is in place the water body, unless the source ceases
discharging first. Nearly all TMDLs for water bodies that are currently impaired are scheduled to be
developed by 2015. While new water bodies requiring TMDLs will likely be listed  in future years, it is
anticipated that the overall  number of impaired water bodies without TMDLs will continue  to decline.
Thus, the applicability of this requirement will likely be greatest in the near term, and decline steadily
over the next 16 years as TMDLs are developed for impaired water bodies.

        The types of costs that may be incurred by small entities, large entities and permitting authorities
as a result of this provision are described below.

Costs to Small Entities

        Small entities are not required to obtain offsets. However, they will need to certify that they are
in fact small entities, as defined in 5 U.S.C. § 601(6).   It is anticipated that the cost  associated with this
certification will be minor since:

        1)     many small businesses are aware of whether they are considered to  be small entities (for
               example, to qualify for small business programs or exemptions of various types) and

        2)     determining whether a business is considered small is only a matter of knowing the
               applicable  SIC code and the applicable SBA revenue or employee size cut-off.

Therefore, to the extent such costs are incurred, they will be minor, and would occur only once.
        2 As noted earlier, the Director may determine that an offset of as low as 1:1 may be appropriate. In this
analysis, we conservatively assume that all offsets will be 1.5:1.

-------
Costs to Large Entities

        Large entities that are new or that are significantly expanding will bear additional costs
associated with securing offsets as follows:

        •      Compliance Costs comprised of the cost of securing and maintaining the offset,
               including transactions costs.  Offsets need only be maintained until the earlier of the
               TMDL being approved and WLAs developed, or until the discharger ceases to discharge.

               To the extent that meeting the offset requirement or obtaining an appropriate permit
               results in delays, dischargers would also incur additional costs associated with the delay.
               The extend to which there may be delays and the cost  of delay has not been estimated in
               this report  It is important to note that some industrial sectors, such as the construction
               industry, are likely to be very sensitive to delays. As  discussed later, the current storm
               water requirements .and permitting processes do not currently have any administrative
               delays associated with them. Delays could result due to the need to secure offsets, the
               need to incorporate them into the permit prior to beginning construction, and the need to
               obtain an individual storm water NPDES permit rather than just submit a Notice of
               Intent (NOI) under a storm water NPDES General Permit. However, delays are not
               inevitable - for example, a construction company might obtain a long-term offset in a
               watershed that could be used over and over again without delay for successive
               construction projects, or may establish a continuing relationship with another company
               that typically manages sites that can readily provide offsets.

        •      Administrative Cost will be incurred by dischargers when they determine whether or not
               they are discharging problem pollutants into an impaired water body.  It is anticipated
               that this information will generally be readily available, and therefore it is anticipated
               that this will normally be a minor cost.

               In addition, large entities that are normally covered under a general permit, but that are
               now required to provide offsets, must be covered under an NPDES permit and would
               incur the additional cost associated with NPDES permits.

Costs to Permitting Authorities

        Permitting authorities will bear additional administrative cost as follows:

               1)     approving proposed offsets and incorporating them into permits, and

               2)     issuing NPDES permits where general permits would normally suffice.

The first type of costs (approving offsets and incorporating offsets) occur only when offsets are required
(i.e.. when the source is a large entity).  The  second type of cost (issuing NPDES permits) only occurs as
an additional cost in cases where an offset is required and the discharger would have otherwise been
covered under a general permit.

-------
Focus of This Analysis

        The impact of this provision is minimal for small businesses. Therefore, this analysis focuses on
the potential administrative and compliance costs associated with offsets as described above for large
businesses. These dischargers potentially needing offsets are generally grouped in this report as follows:

        •      Dischargers that must obtain a permit and offsets due only to wet weather flows, and

        •      Dischargers that must obtain a permit and offsets due to process discharges.

The analysis also includes the additional administrative costs to permitting authorities.
                                                                         X
        The procedure for analyzing the costs of this provision generally includes six basic steps:

        1.      estimate the number, and type of new dischargers annually,

        2.      estimate the proportion of new dischargers which might discharge pollutants of concern
               into impaired water bodies,

        3.      estimate the proportion of these which are large  businesses, and therefore subject to the
               offset requirement,

        4.      estimate the number and magnitude of the offsets required,

        5.      estimate the unit cost (for example, cost/per pound) of obtaining offsets, and

        6.      estimate the total administrative and compliance costs of the offsets.

Each of these  steps is discussed in the following subsections of this section for each of the types of
regulated entities that are affected by the provision — the specific application and order of these steps
varies depending on the industry groups being analyzed. To provide perspective on the offset
requirement itself and the status of this type of activity to date, we begin with an overview of the nation's
experience with water trading and offsets.

THE NATION'S EXPERIENCE WITH TRADING AND OFFSETS

        The following first describes the nation's experience with water trading, and then discusses
experience regarding transactions costs associated with  trading in the air program. Trading provides
firms the opportunity to lower the compliance cost associated with meeting their discharge limits. If the
trades involve reductions in loadings that allow new facilities to locate in areas that could not otherwise
accommodate additional pollutant loadings, these trades are called offsets.

Water Trading

        EPA is in the process of completing a survey of the status of water trading based on an extensive
canvass of EPA regional, state and local staff, as well as national experts on water trading.  Thus far. we
have completed an initial review of 36 trading activities, which likely represents most of the trading

-------
activity in the country.  Five of the'3 6 trading activities involve new dischargers Jthat are required by the
states to obtain offsets, and four involve existing dischargers that are expanding capacity. Twelve of the
36 activities have actually resulted in trades. In cases where trades have occurred, they generally
involved a single trade but ranged up to a handful of trades. Generally, these trades posed substantial
administrative difficulties and several years of preparation preceded them. A summary of those trading
activities  involving offsets is provided in Attachment 1.

        From this trading experience, it is clear that one of the common difficulties has been finding and
concluding a deal with a potential trading partner. Often, potential partners may not want to participate:
in at least one case, potential partners  refused to consider providing offsets because they were concerned
about their ability to meet potential future obligations once a TMDL is developed; in several other cases,
unregulated non-point sources feared being involved in a trade because £iey. feared this would eventually
subject them to regulation. In addition to finding willing partners, another difficulty has  been developing
acceptable legal arrangements for the  trade.  In 6 cases (3 implemented and 3 under development,
including all of the cases in which multiple trades have occurred), the local authority established a
trading bank in which dischargers that need offsets pay the bank, and the bank obtains the offsets, and
most importantly, the control authority assures and vouches for the validity and appropriateness of the
offset. In other cases, an independent entity such as a nonprofit organization is responsible for finding
appropriate offset candidates. The difficulties of finding and securing trades (primarily reflected in high
transactions costs, including delays) sometimes result in companies deciding to pass on trading and
forego the savings in compliance cost that potentially could be realized.

        It is also clear that there is very limited experience across the country regarding water trading.
The 36 trading activities are located in 20 states (including DC), and actual trades to date have  occurred
in about half of these states — most of the trading activities are still in the planning and development
stage.  While state experience is limited, it is clear that industry's experience is even more limited
regarding trading and offsets. It is anticipated that interest in trading and trading activity will increase as
the states  begin to accelerate the development of TMDLs.

        Therefore, it is likely to be difficult, at least in the near term, for businesses to identify  and
secure offsets. It is likely that small businesses, in particular, would find it especially difficult  and
relatively costly to be subject to the offset requirements of this provision. As specified earlier,  small
businesses are exempt from this requirement.

Transactions Costs

        In addition to the costs associated with the added treatment or BMP's needed to achieve the
offset and the added permitting-related costs that may be associated with the proposed  offset (such as
monitoring requirements, obtaining approval for the offset, ensuring compliance over time, etc), there are
also transactions costs associated with obtaining  offsets including:

        •      search costs such as finding appropriate and interested partners, including the cost of
                information to accomplish this;

        •       internal decision-making costs for the potential partner to decide whether to participate:

        •      offset quantification costs, which may generally be minimal for point sources but which

-------
               may be more significant for non-point sources;

        •      negotiation/brokerage costs for all activities related to negotiating agreements between
               participants (including meeting time, legal costs for both sides, etc.); and

        •      profit for the potential partner, over and above covering all the partner's direct costs
               associated with providing the offset, including compensation for becoming subject to
               additional regulatory and administrative requirements (for those who are not already
               subject to the NPDES point source program).

Typically, transactions costs would be much higher,during the initial years of a trading or offset program,
and would tend to decline with experience.  Some of these costs are likely to generally be fixed in nature
(such as the cost of preparing somewhat standard contracts) while other costs may be more related to a
percentage of the total offset cost (such as brokerage costs or profit). For relatively smaller offsets,
percentage-based estimates of transactions cost may not be appropriate because of fixed costs.

        Unfortunately, there is no information available regarding the transactions costs associated with
the limited number of water trades and offsets that have occurred to date.  However, relevant experience
from the air program that can provide insight regarding the magnitude of transactions costs. Emission
Reduction Credit (ERC) trading under the Emissions Trading Program (ETP) was established under the
1977 Clean Air Act as part of the New Source Review process of permitting new  air pollution sources in
nonattainment regions.  Under the ETP, new or expanding sources may obtain ERCs from other facilities
or from within the facility. Information regarding transactions costs associated with the Emissions
Trading Program, especially during the first years of the Program, provide useful  perspective:

        •      The South Coast Air Quality and Management District (SCAQMD) of California which
               implemented its own ETP estimated in 1993 that transactions costs for new source
               review  trading (including fees, commissions, obtaining information, and resources for
               bargaining but not profit) amounted to 35% of the contract price for ERCs.3

        •      Another researcher reported in 1995 that brokerage fees in the ERC market historically
               were  near 15% of sales price, ranging from 7.5% to 49%  with the higher rate reflecting
               all search, negotiation and contracting costs.4

        •      Further, the relatively high  transactions costs for ETP were reflected in the high
               proportion of trades  that occurred within firms rather than between firms because such
               trades avoided much of the transactions costs  -  98% of bubbles,  90% of offsets, and
        • South Coast Air Quality Management District (SCAQMD), Socioeconomic and Environmental
Assessment. California, Volume III. October 1993.

        4 Daniel J. Dudek and Jonathan Wiener. OECD. Environmental Policy Committee, Group on Economic
and Environmental Policy Implementation.  Joint Implementation and Transactions Costs Under the Climate
Change Convention.  May 1996.

-------
               about 80% of banking actions were of the internal type during the period 1977-1986.5

        For additional perspective, it is useful to review the brokerage fees associated with transactions
for the SCAQMD's RECLAIM program, in which sources may buy and sell ERCs for only two
pollutants (NOx and SOx). The brokerage costs associated with a centralized, computerized auction
system in which there are no face-to-face negotiations (nor any search costs, etc.) is 3% of the
transaction price charged to both buyer and seller plus a one-time fee of $100.6 In addition to these fees,
legal fees for drawing up a standard contract would also be incurred — such contracts might, for example,
cost about $2,000.

        Thus, transactions costs can be a significant component of the cost of offsets,  and tend to be ,
feigher in the first years of the program.  Transactions costs for smaller offsets tend to  be relatively higher
than for larger offsets due  to fixed costs.

THE POTENTIAL COST OF OFFSETS FOR STORM WATER DISCHARGERS

        Currently, storm water dischargers are covered under the National Pollutant Discharge
Elimination System (NPDES) through general  permits that specify the requirements that the discharger
must follow.  A permittee  need only follow the requirements of the general permit: prepare and
implement a pollution prevention plan (comprised of implementing Best Management Practices and
other requirements), and submit a Notice of Intent (NOI). For construction, whether for large or small
sites, the activity can begin within 2 days of the postmark of the NOI.

        EPA published regulations in 1990 (Phase I) for storm water discharges associated with certain
industrial activities, medium and large municipal storm sewer systems (MS4s), and construction sites
larger than 5 acres. EPA is in the process of finalizing additional regulations (Phase II), subjecting
discharges from small MS4s and construction sites as small as 1 acre to regulation as well. The Phase II
rule does not alter the Phase I requirements for industrial establishments or expand the universe of
industrial establishments covered.

        Under the provisions of the proposed rule being examined in this report, a new storm water
 discharger that is not a small entity will be required to obtain a 1.5:1 offset if the discharge contributes
to the impairment of a water body. In addition, the discharger will need to obtain an NPDES permit,
rather than being covered under a general permit. The offset requirement does not apply to categories of
facilities that are not covered under the Phase I or II storm water rule - e.g., construction sites that
disturb less than 1 acre will not need an offset, regardless of the size of the business entity. The following
analysis first addresses storm water discharges from the construction industry, and then  addresses storm
water discharges from other sources.
        5 Hahn, Robert W. and Gordon L. Hester. "Where Did All the Markets Go? An Analysis of EPA's
Emissions Trading Program."  Yale Journal on Regulation 6 (1989): 109-153.

        6 Gangadharan, Lata. Transactions Costs in Tradeable Emissions Markets: An Empirical Study of the
Regional Clean A ir Incentives Market in Los Angeles. University of Melbourne Department of Economics.
Research Paper 591. October,  1997.

10

-------
The Construction Industry

        To estimate the potential cost of offsets to the construction industry, it is necessary to estimate:

                1.     the number of large firms in the industry (since small firms are exempt from the
                      offset requirements),

                2.     the number of construction sites that are managed by large firms (since total
                      offset costs will depend on how many construction sites will require them),

                3.     the proportion of construction sites that affect impaired water bodies (since
                      offsets are only required for construction sites that discharge pollutants of
                      concern to impaired water bodies),

                4.     the number of offsets that might be needed,

                5. -    the size of these sites (since the size is related to the discharge, and the resulting
                      offset requirement),

                6.     approaches that might be used for obtaining offsets,

                7.     the unit cost of an offset for the construction industry,

                8.     other alternatives for obtaining offsets,

                9.     the resulting cost of the offset and related permitting costs to industry, and

                10.    the resulting administrative cost to permit authorities.

Much of the data needed to make these estimates is not available. However, progress can be made in
assessing the potential magnitude of these factors. Each of these areas is discussed below.

1. The construction industry is overwhelmingly comprised of small businesses.

        Under SBA's definition of small business, building construction firms with revenues under $ 17
million would be considered small. EPA's Economic Analysis of the Final Phase II Storm Water Rule
used Dun & Bradstreet data to estimate that in  1997 there were about 189,453  building construction
firms, of which  187,610 are small businesses (EA Exhibit 8-1).  Therefore, 99% of the industry would be
exempt from the offset requirement. Consequently, there are approximately 1.843 firms that would be
considered large using the SBA's definitions, and therefore subject to the offset requirement.

2. Larger firms are more likely to manage higher-value construction sites and more sites per firm.

        Since small firms are exempt from offset requirements, it is necessary to estimate the number of
sites that large firms manage annually. Unfortunately, there are no data available that directly indicate
the number of construction sites or the size of the construction sites that are managed by size of the  firm
managing the construction.  However, some inferences can be made from available data. Note that

-------
"sites" in this analysis does not refer to "construction starts" or to construction permits. Rather, "sites"
refers to developments that require a single discharge permit, but which might be comprised of numerous
"starts" and might have numerous building permits. This consideration was incorporated into the
Economic Analysis of the Final Phase II Storm Water Rule and is reflected in the following estimates.

        Although.the 1,843 large construction firms represent about 1% of the firms in the industry, they
represent about 38% of industry-wide construction revenue and about 26% of industry-wide
employment, according to the SB A' most recent compilation of census data (1995). The revenues for
large construction firms ranges from SI 7 million to over $100 million per firm, with an average of nearly
$57 million and an average of about 172 employees per firm. In contrast, the average revenue reported
by the SBA for all small construction firms was $926,000. However, this estimate is understated because
a number of the small construction firms are engaged in remodeling only. Of the 187,610 small entities,
40,339 have revenues under $100,000 annually and tend to be part-time (since they statistically have an
average of less than one employee per firm), so it is likely that these firms are engaged in remodeling.
The number of small entities with revenues over $100,000 is 147,271 firms with an average revenue of
about $ 1.2 million and an average of about 5 employees per firm. Therefore, large firms must be 1)
handling more sites than small firms, 2) the sites they develop must be of higher value, or 3) both.

        The Economic Analysis of the Final Phase II Storm Water Rule estimated that there were
approximately 522,000 construction starts of all sizes in 1998 (EA Exhibit 3-3), and that the 1998 mean
home sales price was $181,300 (EA page 8-11).  As shown later in Exhibit 1-2, smaller projects tend to
have durations of less than 6 months, so that the typical small construction firm would need 2-3 projects
over the course of a year.  If small construction firms with revenues between $100,000 and $500,000
typically undertake projects with a mean value of $181,300 and if small construction firms with revenues
over $500,000 typically undertake projects with a mean value of $500,000, then small construction firms
would manage a total of 492,779 sites annually. Therefore, the large entities might be managing 29,221
sites annually, for an average of about 16 sites per firm at an average value of about $3.6 million per site.
A value of about $3.6 million might represent typical residential or commercial construction7 on about a
3 acre site, which represents the median size site (EA Exhibits 3-3 and B-2-1). If instead, large entities
managed 20,000 sites, there would be about 11 sites per firm at an average value of about $5.2 million
per site; and if instead large entities managed 40,000 sites, there would be about 22 sites per firm at an
average value of about $2.6 million per site. Therefore, it seems likely that 40,000 sites would represent
a conservative upper bound on the number of sites managed  by large entities, and that the number  of
sites could easily be  below 20,000 sites.

        Given the above considerations, we conservatively assume for this analysis that the large firms
might handle 20,000 to 40,000 (30,000 ± 33%) new construction sites annually.  To the extent that they
manage more  or less sites, the incidence of the need for offsets will increase or decrease in proportion.
        "The Economic Analysis of the Final Phase 11 Storm Water Rule estimated (Exhibit 8-9) that: a 1 acre
residential site represents about SI.5 million in revenues, a 5-acre site represents S4.1 million in revenues, and a 7-
acre site represents $6.3 million in revenues. The EA also estimated (Exhibit 8-10) that: a 1-acre commercial site
represents SI million in revenues, a 3-acre commercial site represents S3 million in revenues, a 5-acre commercial
site represents S5.1 million in revenues, and a 7-acre commercial site represents S7.2 million in revenues. Note that
high-rise development results in much higher revenues per acre - for example, the EA Exhibit 8-9 includes  a 3-acre
site with high-rise development having revenues greater than for a 7-acre site.

12

-------
 3. 'Perhaps 2-3% of construction activity contributes to impairment of the nation's waters.

        To estimate the amount of new construction activity that might be subject to the offset
requirement, it is necessary to estimate the amount of construction activity that contributes to impairment
of the nation's waters.

        The National Water Quality Inventory Report to Congress (U.S. EPA, 1998) reports in Tables A-
5, B-5, and C-1,5 and 9 that construction contributes to the impairment of the nation's waters as follows:

                                           Exhibit 1-1
      Percent of the Nations 305(b) Waters Where Construction Contributes To Impairment
Water body Type
Rivers (miles)
Lakes (acres) -
Estuaries (sq. mi.)
Shoreline (miles)
%
Surveyed
19%
72%
40%
6%
Impaired Miles, Acres or Sq. Mi. as a Percent Of:
Total Waters
<1%
2%
3%
<1%
Surveyed Waters
2%
4%
4%
1%
Impaired Waters
6%
11%
11%
8%
As shown in the exhibit, varying percentages of the nation's waters have been surveyed, ranging from
6% for shorelines to 72% for lakes. Across these different types of water bodies, states have identified
construction as contributing to impairments for 6% to 11% of the impaired water bodies, 1 to 4% of the
surveyed water bodies, and <1% to 3% of the nation's total waters.  This is consistent with the causes
reported for the 303(d) listed water bodies, in which about 9% of the listed waterbodies are impaired for
sediment in locations where construction activity occurs.8

        It is clear that a great deal of construction occurs in areas that are not near waterbodies that are
currently listed as impaired by sediment. For example, 5 states account for about 17% of the housing
starts in the nation, but have  not listed any waters that are impaired by sediment. Another 10 states,
representing about 11% of housing starts, report less than 10 waterbodies that are impaired by sediment,
and it seems likely that much of the construction in those states is not near these waterbodies.

        The percentage of construction  activity that might occur near water bodies that are impaired by
sediment-related causes can be bound by the following considerations:

        •      If all construction activity occurred uniformly on surveyed waters, then it might be
               appropriate to estimate that 1 to 4% of construction  might contribute to impairments.
        8For 303(d) listed waters, 30% are reported as impaired for either siltation, sediment, suspended solids,
total suspended solids, turbidity, or erosion.  In about 70% of these cases, agriculture, silviculture or resource
extraction activities were reported as the cause, and it is unlikely that much construction activity is occurring in
these areas. Therefore, in about 30% of the cases where listed waters are reported as impaired by sediment-related
problems, construction activity may be occurring.  This amounts to about 9% of the listed waters (30% x 30%).
                                                                                                i:

-------
        •      If as much as 1/3 of the nation's construction activity were concentrated on 303(d) listed
                waters, and if construction activity is present for 9% of the 303(d) listed waters (where
                construction  offsets might be needed), then perhaps 1/3 of 9%, or 3% of national
                construction  activity might need offsets.

 However, an additional consideration is that offsets are only required if the activity actually discharges
 into the impaired water body. As shown later in this chapter, only l/50th of the process discharges that
 might contribute to an impairment actually discharge directly into the impaired water body and therefore
 would need offsets.  Therefore, only a fraction of the construction sites near waterbodies that are
 impaired by sediment-related causes would be expected to need offsets.
                                       /
        Therefore, it seems reasonable to assume that perhaps 2 to 3% of construction activity might be
 contributing to impairments.  Using this assumption, then perhaps 10,440-15,660 sites of the 522,000 new
 construction sites annually may be contributing to impairments.  Note that this would be an under-
• estimate or over-estimate to  the extent that construction activity is actually more or less concentrated in
 those locations that would actually result in the requirement to-provide offsets. For the remainder of this
 analysis, we assume that-10,000-16,000 sites may be contribution to impairments.

 4.  Perhaps 400-1,200 of the  10.000-16.000 construction sites that mav contribute to impairments are
 managed bv large businesses .

        As discussed earlier, we have assumed that large construction businesses develop about 20,000
 to 40,000 sites per year. If 2-3% of the 20,000 to 40,000 sites managed by large construction firms
 contribute to the impairment of a water body (as discussed above), then this would represent 400-1,200
 sites that may be required to  provide offsets — a portion of these sites, however, are smaller than 1 acre
 and would not be subject to the offset requirement.

        The applicability of the offset requirement and the cost associated with the offsets for these sites
 would depend importantly upon the size of these construction sites.

 5. Perhaps 292-877 of the 400-1.200 sites managed bv large businesses are larger than 1 acre.

        The Economic Analysis of the Final Phase II Storm Water Rule, provides a detailed distribution
 of construction  site sizes for  the 522,000 new construction sites for the sites under  10 acres in size (EA
 Exhibit B-2-1),  amounting to 80% of all sites.  The size distribution for the 20% of sites that are over 10
 acres may be  inferred from the Notice of Intent data base that EPA maintains for the state storm water
 permit programs that EPA manages. Review of 1996-98 NOIs submitted for the construction sector
 indicate that roughly 42% of the construction sites over 10 acres are between 10-20 acres  in size, roughly
 42% of sites are between 20-50 acres is size, roughly 10% of the sites are between  50-100 acres, roughly
 6% are between 100-500 acres and about 0.1% are more than  500 acres in size.  In addition, the
 construction activity typically takes under Vz a year for sites under 5 acres, and tends to steadily increase
 for larger sites;  however, the ranges are broad, and even the smallest sites may take several years while
 some larger sites may take only several months.9  Combining these  observations from EPA's 12-state
 NOI database for the sites over 10 acres with the size distribution of sites provided in the  Storm Water
         9The EA of the Phase II Storm \Vater Rule assumed the life of a construction site to be 6 months (Ex 4-10).

 14

-------
EA provides the rough illustrative profile shown in Exhibit 1-2 below.

        There is no reason to expect that the size distribution of construction sites that are located on
impaired water bodies would be different than shown in Exhibit 1-2.  Therefore, the same size
distribution would be expected to apply to the 10,000 - 16,000 sites that might contribute to impairments,
as shown in Exhibit 1-3.
                                           Exhibit 1-2
                                Construction Site Characteristics
Size
less than 1 acre
1-5 acres
5-10 acres
10-20 acres
20-50 acres
50- 100 acres
100-500 acres
over 500 acres
Total
Approximate
Number
141,000
1 98,000 '
78,500
44,000
43,000
11,000
6,000
500
522,000
Typical Max
Duration
<6mo
<6mo
<12mo
<12mo
<12mo
<12mo
<1 8 mo
<18mo

                                           Exhibit 1-3
          Potential Characteristics of All Construction Sites Contributing to Impairments
Size
less than 1 acre
1-5 acres
5-10 acres
10-20 acres
20-50 acres
50- 100 acres
100-500 acres
over 500 acres
Total
Approximate
Number
2,700-4,300
3,800-6,100
1,500-2,400
850-1.370
800-1.300
220-330
120-185
10-15
10,000-16,000
Typical
Max Duration
< 6 mo
< 6 mo
<12mo
<12mo
<12mo
<12mo
<18 rno
<18 mo


-------
        As describer earlier, none of the sites under l.acre in size are subject to the requirements of the
proposed provision.  Further, as estimated earlier, perhaps 400 - 1,200 of the construction sites in Exhibit
1-3 (i.e., of the 10,000 - 16,000 sites that may contribute to impairments) will be managed by large
businesses and therefore will potentially be required to provide offsets.  However, even for these sites,
those that are under 1 acre in size are not required to provide offsets.

        While the size distribution of the construction sites managed by large businesses is not known,
there is no reason to expect the size distribution to be different from the national average. While it is
clear that larger businesses manage sites that have greater value, it is unclear how value relates to the size
of the area disturbed: high construction value sites could be smaller in size (for example, sites for high-
rise office, commercial, or industrial uses) or larger in size (for example large,  upscale residential
developments or mails); similarly low  construction value sites may be smaller  in size (a single residence
or smaller commercial uses) or larger in size (low-cost industrial).  Therefore, there isn't a good basis for
assuming that large firms have a distribution of site size that is different from the overall distribution.
The resulting size distribution for the 400-1,200 construction sites that may contribute to impairments
and that are managed by large firms is  shown in Exhibit 1-4 below.

                                           Exhibit 1-4
          Potential Characteristics of Construction Sites That Contribute to Impairments
                             and  That Are Managed by Large  Firms
Size
< 1 acre
1- 5 acres
5-10 acres
10-20 acres
20-50 acres
50- 100 acres
100-500 acres
over 500 acres
Exempt (< lacre)
Total Offsets Needed
Approx Number
[108-323]
152-458
60- 180
34- 103
32-98
9-24
4- 13
1 - 1
108-323
292 - 877
Median Acres/Site
0.5
2.5
7.5
15.0
35.0
75.0
300.0
750.0
0.5
5.0
Approx Total Acres
[54-162]
380- 1,145
450- 1,350
510- 1,545
1,120-3,430
675 - 1,800
1,200-3,900
750-750
54- 162
5,085-13,920
        As shown in Exhibit 1-4, about 27% of these sites will not require offsets because they are less
than one acre in size, and therefore would be exempt from the offset requirement. Therefore, of the 400
1,200 construction sites that both contribute to nonattainment of water quality standards and are owned
16

-------
by large businesses, perhaps 292 - 877 sites will be required to obtain offsets annually.10  These sites
represent about 5,085 - 13,920 acres of disturbed area.  Since the size of the offset needed (and therefore
the cost of the offset) is likely to be directly related to the size of the site, it is clear from  Exhibit 1-4 that
the assumption regarding the size distribution for sites managed by large entities is an important factor
affecting the overall cost of obtaining offsets. Note that the estimate of 292-877 sites representing 5,085
- 13,920 acres annually potentially needing offsets is based on  1998 construction levels.  As discussed
later, growth in construction activity through 2015 must be taken into consideration when estimating the
potential cost of offsets.

6. Obtaining offsets will likely require a range of initiatives bv industry and local governments

        There are several possibilities for how industry and state/local governments may eventually
respond to this requirement, although it is unclear how likely each is or how long it would take for these
approaches to develop:

        •      It seems likely that, where possible, the construction industry may seek offsets within the
               construction industry itself.  For example, offsets might be obtained from construction
               sites under 1 acre in size. In this way,  offsets can be obtained from firms that are known
               to one another.  Some firms may even  be able  to generate offsets internally by instituting
               controls at their small sites (< 1 acre) to generate offsets for use at their larger sites.

               However,  this would not be possible in the 10  States and Territories that already require
               sites of all sizes to meet storm water requirements (according to the Storm Water Phase
               IIEA, Connecticut, Delaware, DC, Georgia, Maryland, Michigan, New Jersey, North
               Carolina, South Carolina, and Puerto Rico require all construction sites to meet storm
               water requirements).

               In addition, this approach may only be practical for relatively small construction sites
               that require offsets. Larger construction sites, such as those over 25 or 50 acres may find
               it impractical to secure offsets from the large number of small (under 1 acre) sites that
               would be needed. In addition, there are fixed costs associated with each small offset,
               and the total cost of obtaining many of these small offsets may become prohibitive
               compared to other options.

        •      The construction industry might seek partners for offsets outside of the construction
               industry, including agriculture  and silviculture. Perhaps offset brokers will assist in this
               process. These offsets are likely to be "permanent" in the sense that they are  likely to be
               far more long-lasting than the construction activity requiring the offset,  and may be
               effective until the TMDL is developed (requiring only periodic maintenance and
               inspection once the BMPs are in place).  Thus, a construction firm can obtain a single
               offset that could serve as the offset for a series of successive construction activities.  For
               example, an offset that is obtained for a 10-acre construction site could  be reused for
               future construction activities amounting to 10 acres per year. This could significantly
        '"Since construction sites tend to have durations of less than a year, it is assumed in this analysis that each
of the 292 -877 new construction sites each year will need to obtain offsets.

                                                                                                 r

-------
               lower the overall annualized cost of offsets through the period until TMDLs are
               developed and the offsets or no longer required.

        •      Private parties, perhaps the including construction firms themselves, might establish
               offset banks to "lease" offsets to construction firms that need for the duration of
               individual construction projects. This would provide more flexibility to construction
               firms that undertake projects of a variety of sizes (i.e., land disturbed) and duration.

        •      It may be easier and cheaper in some cases (especially in the 10 States and Territories
               that already require all construction sites to meet storm water requirements or for the
               larger construction sites that would need to obtain agreements from many small sites to
               obtain the required offset) to avoid the offset requirement by going to the equivalent of
               zero discharge.  However, the conditions and requirements for achieving zero discharge
               for the purposes of this provision have not been defined.

        •      State or local authorities may institute offset banks.

Some of these alternatives (in particular, those involving brokers or banks) might take time to develop,
and would probably not be available to those sites needing offsets in the near term.

7. The unit cost of an offset obtained from small construction sites.

        The most likely scenario in  most states for the smaller construction sites than need offsets might
be for the construction industry to find offsets within the construction industry itself.  Such offsets might
be available from construction sites under 1 acre in those states that don't already require these sites to
apply storm water BMPs. In ideal cases, the offsets may even be able to occur within a company, rather
than between companies.

        For example, consider a construction site of 10 acres. If BMPs are 80% effective", an offset
(1.5:1) is needed that is equivalent to 30% of the original discharge. The required offset might be
considered equivalent, then, to completely eliminating the uncontrolled discharge from 3 acres of
construction. If BMPs are 80% effective, controls for nearly 4 acres of construction (i.e., 3.75 acres)
would provide the needed offset. This would suggest that, as a general rule (given these assumptions), a
large construction site (i.e., over 1 acre in size) can obtain the needed offset by controlling the discharge
from a combination of small construction sites that are equivalent to about 40% of its size. Note that if
BMPs are less effective than assumed here, both the amount of offsets needed and their cost would be
higher; similarly, if they are more effective than assumed here, the amount of offsets needed and their
cost would be lower.'2 This clearly provides an incentive for construction site to maximize the efficiency
        11 As discussed in the Economic Analysis of the Final Phase II Storm Water Rule, Section 4.4.2, detailed
modeling taking into account three site sizes, three soil credibility levels, three slopes, the BMP combinations from
EPA's 27 model sites, and 15 climactic regions resulted in a weighted average reduction of 93%, with none lower
than 92%. EPA believes that effectiveness in excess of 80% can generally be expected.

        12The construction acreage that would need to be secured as a percentage of the size of the site that must
obtain offsets is = 1.5 x (1 - BMP effectiveness) / (BMP effectiveness). Thus, for example, for 90% effectiveness,
the additional acres that must be controlled would be 16.7% of the original site; for 70% effectiveness, the

-------
of construction site BMPs.

        We assume that each of the sites that provide offsets would at a minimum need to meet all of the
requirements of a storm water General Permit. Presumably, offsets would be sought from the largest
sites available just under 1 acre, and we estimate the cost of meeting the storm water General Permit
requirements to be $2,50013 for each site providing an offset. Thus, the cost for 4 such offsets would be at
least $10,000 since this estimate doesn't include transactions costs (e.g., the cost of identifying and
securing the offset). Perhaps this estimate of $1 0,000 would reflect the cost to firms that are able to
generate their own offsets internally — however, as shown in Exhibit 1-4, the acreage associated with
small sites developed by large firms represents only about 1% of the acreage associated with sites larger
than 1 acre.  Therefore, it more far more likely that the full transactions costs will be incurred.

        Unlike other situations where offsets or trading have occurred, some large construction firms
will routinely and frequently face the requirement to obtain offsets. These firms will likely face the
requirement to obtain offset several times a year or more. It  is likely that in this circumstance, these
firms will develop on-going relationships with other firms that develop smaller sites (i.e, < 1  acre per
year), thereby  reducing transactions costs; alternatively, it seems reasonable to expect that in this case
brokers could readily serve as intermediaries for a nominal fee. In this circumstance, it seems reasonable
to assume that search and negotiation costs will be nominal once procedures and relationships are
established, and that the primary transactions cost will be associated with legal fees, profit and possibly
brokerage fees. Legal fees, profit and brokerage fees might average $2,000 - $3,000 per trade. In some
cases, it might be necessary to pay additional premiums to secure offsets in a timely manner to avoid
construction delays. For this analysis, we assume that transactions costs will average $2,500 per trade.

        Thus, the cost associated with obtaining an offset from a 1-acre site might average $5,000
(BMP/administrative costs of $2,500 and transactions costs of $2,500). The resulting cost for securing
the total offset needed for a 10-acre construction site would be $20,000 (four 1-acre sites at $5,000 each).
The cost of the offsets for a 10-acre site might be roughly comparable to the cost of that site complying
with its own storm water requirements.

        From the above example, the cost of obtaining a 1 .5: 1 offset from small construction sites
amounts to about $2,000 per acre of the construction site seeking the offset (i.e., $20,000 for the cost of
the offsets for a 10-acre site divided by 10 acres).  Thus, using this factor, the total cost of securing
offsets for a 10-acre site would be $20,000 ($2,000 times 10  acres) and the total cost for securing offsets
additional acres that must be controlled would be 64.3% of the original site.

        lj The Economic Analysis of the Final Phase II Storm Water Rule estimated (Exhibit 4-9) that the average
cost of BMPs for a 1-acre site would be S1.206 (1998 dollars), ranging from S317 to 51,799 depending on soil
credibility and slope.  The cost of the Storm Water Prevention Plan for a small site (1-acre) was estimated (Exhibit
4-10) to be S608 (1998 dollars), assuming a 6-month duration (bi-weekly inspections account for over half of this
cost). The remaining  administrative requirements (Notice Of Intent, municipal notification, records retention.
Notice Of Termination) were estimated (Exhibit 4-11) to be SI 10 (1998 dollars). The total cost for a small site
would be $1,924.  However, since the site is serving as an offset, we assume that there will be additional costs
(perhaps because a more detailed SWPPP might be needed, or perhaps because sites selected to serve as offsets
might have a lonaer duration) amounting to an added 25%. bringing the total to 52,500 (in March. 1999 dollars).
This is equivalent to the average cost estimated in the EA for sites of all sizes from 1-5 acres.
                                                                                                 19

-------
for a 50-acre site would be $100,000 (52,000 times 50 acres).
                                            •

        Larger construction sites may find it impractical to employ this approach for obtaining offsets.
For example a 50 acre site would need to obtain offsets from 25 small construction sites, and a 100 acre
site would need to obtain offsets from 50 small construction sites. In addition, the fixed permitting and
transactions costs associated with obtaining offsets from each of the small construction sites add up
quickly, and maymake this approach too expensive for sites that require the larger offsets (perhaps
starting with sites over 10 acres in size).

        Further, it is important to note that the above discussion includes two key assumptions about the
construction sites requiring offsets and the sites providing offsets:
                                   /           •            •         •
        •      Site Duration. The analysis assumes that the duration of the construction for sites
               providing the offset is the same as the duration of the construction for the large site
               seeking the offset.  However, if the duration of the construction site needing offsets is
               longer than the duration of the sites providing offsets, then more sites providing offsets
               will be needed. For example, if offsets are needed for a year, and the sites providing
               offsets have durations of 6 months, then the total cost of the offsets could double.
               However, it should be noted that a portion of the construction sites under 1  acre have
               durations of more than one year, and it is likely that these longer duration sites would be
               sought out for purposes of obtaining offsets. In addition, this is more likely to be an
               issue for the larger sites than the smaller sites, and there is another important
               compensating factor, discussed below.

        •      Portion Disturbed.  Larger sites needing offsets might not disturb the entire area at once.
               To the extent that work at a site proceeds in stages, then shorter duration offsets may be
               more than adequate.  For example, if a 50-acre site has a year-long duration, but only
               half of the site is disturbed at any one time, then the offsets needed for the entire year
               would be equivalent to two consecutive 6-month offsets for 25-acre sites. In addition,
               staging may also reduce the amount of offsets needed.  For example, if a 50-acre site
               progresses in consecutive 5-acre stages over the course of a year, then two consecutive
               6-month offsets for 5 acres would be adequate for the 50-acre site (which would be  1/5
               of the cost assumed in this analysis for the 50-acre site).

Overall, we believe that the assumptions made  in this analysis tend to be conservative in estimating  the
number and magnitude of offsets needed.

8. Alternatives to offsets from small construction sites.

        Offsets would need to be sought outside of the construction industry in states that already require
construction sites under 1  acre to meet storm water requirements, or where the cost of offsets from small
sites is high compared to other alternatives. In  addition, even if it were an available option  and  cost-
effective, there may not be enough small sites available to provide the required offsets.  To  provide
offsets for 5,085-13,920 acres of construction using small construction sites, it would be necessary to
engage 2,034- 5,568 one-acre sites. As shown in Exhibit 1-3, there may be 2,700 to 4,300 small  sites that
may be  contributing to impairments, and their median size may be about  .5 acres. Therefore, other
alternatives will likely need to be utilized.

20

-------
        At this point it is unclear whether the costs for construction sites for finding, securing and
maintaining offsets outside of the construction industry would be less expensive than the cost of
obtaining them within the industry. This is because it could be more difficult to identify offset partners
outside of the construction industry and the premiums that are required by the seller of the offset might
also be higher (especially from sources who are not currently subject to NPDES requirements). On the
other hand, since the construction industry is only seeking temporary offsets (i.e., offsets that are  in
effect until either construction ends or the TMDL is developed), it is possible that sources that believe
that they might be required to undertake control measures once the TMDL is developed would be very
receptive to have such measures partially or fully paid  for by selling an offset to the construction
industry. Moreover, such offsets would only have to be secured once, and could serve as the offset for
many successive construction projects until the time that the TMDL is developed and implemented.
Further, if offset banks are available, the cost of offsets could conceivably be dramatically lower.

        A potentially attractive option for construction sites that need offsets might be to,avoid the offset
requirement itself by substantially reducing the remaining discharge, perhaps even achieving the
equivalent of zero discharge. However, this depends on the requirements that these sites would need to
meet in order to achieve this, and this has not been specified to date.

        Over time, it is likely that the cost of finding and securing offsets would decline as the practice
becomes routine or as appropriate institutions become  established for facilitating offsets. In addition,
over time, more impaired water bodies will have  TMDLs developed for them, reducing the fraction of
construction sites that will need to obtain offsets.

9. The total annual cost of offsets for construction sites.

        Compliance Cost

        As estimated above, in 1998 there may have been 292-877 construction sites representing 5,085
to 13,920 acres of construction that might have required offsets. To estimate the cost of the offsets, we
assume that this cost can be represented by the cost associated with obtaining offsets from small
construction sites. Since this cost includes a relatively large transactions cost (on a percentage basis, due
to fixed costs), this is more likely to overestimate the cost associated with other means of obtaining
offsets than to underestimate it. In addition, the acreage potentially requiring offsets must be increased
to reflect growth in construction from 1998 through 2015 — as analyzed in the Economic Analysis of the.
Final Phase II Storm Water Rule (Section 4.2.2), the annual growth rate for construction from 1980-1994
was 1.3% and we assume this growth rate applies through 2015. Finally, since states will be making
progress each year in developing TMDLs, the number of construction sites contributing to impaired
water bodies will decline - we assume in this analysis  that TMDLs will be developed uniformly over the
16 years from 2000 through 2015, so that the number of water bodies requiring offsets will decline
steadily at a rate of about 6% per year (assuming that few new water bodies are listed during this  period).

        The cost of securing the required offsets, assuming  that the offsets are obtained from small
construction sites, is estimated to be about $2.000 per acre of the construction site needing the offset.
Using the OMB-approved discount rate of 7%, the present value of the cost of offsets (taking both
construction growth and TMDL development into account)  is S64.4-S176.4 million from 2000 through
2015. and the annualized cost over this period is  S6.82-S18.67 million. However, it seems likely  that
larger construction sites will be able to find less expensive alternatives than obtaining offsets from small

-------
• construction sites.  While we have been able to speculate about these alternatives at this point, we do not
 know whether such alternatives are in fact feasible or acceptable or what they might cost.

        The most important factors affecting the cost estimate are the assumptions that: 2-3% of
 construction activity contributes to impaired water bodies; 20,000-40,000 construction sites are managed
 by large entities; the size distribution of sites requiring offsets; the assumption that the overall cost of
 offsets can be represented by the cost associated with obtaining offsets within the industry from small
 construction sites; the 80% effectiveness of BMPs, that transactions costs for obtaining offsets from
 small construction sites average $2,500 per transaction; and that the future growth in construction is
 1.3% annually.

        Administrative Cost

        The 292 to 877 construction sites annually that will be required to provide offsets will also be
• required to have individual storm water NPDES permits instead of only complying with General Permits.
 It is assumed in this analysis that the current costs for these sites associated with complying with General
 Permits and providing an NOI would still be incurred, since nearly all of the administrative cost for the
 NOI is associated with the preparation of the storm water prevention plan which must still be prepared.
 Thus 292-877 individual NPDES permits annually would be required for the construction sites that must
 obtain offsets.

        In addition, we assume that similar information must be provided for each of the sites that are
 providing the offsets.  The cost of obtaining this information would be included as an additional cost
 associated with the offset transaction incurred by the sites seeking offsets. If each construction site
 needing an offset is able to obtain its offset from single non-point sources, then the additional
 information would be needed  from 292-877 sources. However, if small construction sites are used to
 obtain offsets, as assumed in this analysis, information from all of these sites would be needed.  In this
 case, the number of small construction sites affected would be at least 2,034 to 5,568 sites (the number of
 1-acre sites needed to offset the discharge from 5,085 - 13,920 acres).u

        Based on the December 3, 1998 Information Collection Request for the National Pollution
 Discharge Elimination System Discharge Permit and the Sewage Sludge Management Permit, the
 applicant burden for preparing an NPDES application to discharge storm water (Form 2F) is 28.6 hours
 at an average rate of $36.12/hour for the private sector. However, 5.6 hours of this estimate is associated
 with information that either not applicable to construction or duplicates the NOI.  Therefore, the burden
 per form would be 23  hours. The total administrative cost associated with NPDES permits for offsets
 has a wide range depending on the requirements for the sources that are providing the offsets:

        •       For the 292-877 sites only that need offsets:            S  242,582  -  S   728,577
                Form 2F (292-877 sites x 23 hours x $36.12/hr)

        •       For 2,034-5,568 sites providing offsets:                51,689,766-    $4,625,672
                Form 2F (2,034- 5,568 sites x 23 hours x S36.12/hr)
         14 Note that the administrative cost for the small construction sites associated with meeting the storm water
 General Permit requirements have already been included in the compliance cost, as pan of the cost of the offset

-------
Thus, the total annual administrative cost would range from $1.9 million to $5.4 million in the initial
years.  Taking into account the growth rate of construction activity and the rate of development of
TMDLs results in a present value of S12.2-S33.9 million, representing an annualized cost through 2015
of $1.3-53.59 million.

        Total Cost To Construction Firms

        The total compliance and administrative cost associated with the offset requirement is estimated
to range from $8.12 to $22.26 million.

10. Costs to Permit Authorities
                                       X
        The corresponding additional cost to permitting authorities for processing individual NPDES
permits and offsets is based on the burden of 2.2 hours as estimated in the above ICR for evaluating
Form 2F at an average rate of $31.68/hour for federal or state employees.  In addition, it is anticipated
that the NOI information for the sites providing offsets would also be reviewed by the permit authorities
when evaluating the offset proposal, and this added burden would be 1.5 hours per site, based on the
April,  1999 ICR for Phase II of the NPDES Storm Water Program:
               For the 292-877 sites only that need offsets:
               Form 2F (292-877 sites x 2.2 hours K $31.68/hr)

               For 2,034-5,568 sites providing offsets:
               Form 2F (2,034-5,568 sites x 2.2 hours \ $31.68/hr)
               NOI     (2,034-5,568 sites x 1.5 hours x $31.68/hr)
$  20,351  -    $  61,123
$238,417-     $652,659
Thus, assuming that each construction site needing offsets and as many as 3,566-8,216 sites providing
offsets require Form 2F, and that permit authorities also review NOI information from the sites providing
offsets, then the total administrative cost to permit authorities is estimated to range from $258,768 to
$713,782 in the initial years.  Taking into account the growth rate of construction activity and the rate-of
development of TMDLs results in a present value of $1.6-54.5 million, representing an annualized cost
through 2015 of $173,527-$478,653.

11. Summary

        The annualized costs associated with the proposed offset requirements for storm water
discharges from construction sites is as follows:
Type of Cost
Cost To Private Sector
Cost To Permit Authorities
Total Cost
Annualized Cost
(S Millions)
8.12-22.26
0.17-0.48
8.29 - 22.74

-------
Other Storm Water Discharges

       To estimate the potential cost of offsets for other storm water discharges, it is necessary to
estimate:

       1.      the industries and the number of new establishments annually that might be affected by
               this requirement,

       2.      the portion of these establishments that might be owned by large businesses,

       3.      the portion of the establishments owned by large businesses that might affect impaired
               water bodies,   x                             •

       4.      the number of establishments that might be required to provide offsets, and

       5.      the cost of offsets for these establishments.

The analysis for other storm water discharges is especially difficult because of the diversity of the
industries covered (EPA's General Permit identifies 29 industry sectors), the diversity of the
constituents of the storm water discharges and the BMPs used to address them (in addition to the 29
sectors, EPA further divided some of the sectors into subsectors), and the limited information available
about them. However, as in the case for construction, some progress can be made in most of these areas.

1. There is a wide diversity of industry covered  by this requirement.

       As for construction, for the storm water permit programs that EPA manages, EPA maintains a
data base of the Notices of Intent submitted pursuant to the storm water general  permit requirements.
Unfortunately, the data available in the NOI data base for other storm water discharges does not reflect
the number of new facilities, but is instead more a reflection of the number of existing facilities that are
coming into compliance with the storm water requirements. To a large extent, much of the NOI activity
in recent years is attributable to the expiration of the EPA Storm Water Baseline Industrial General
Permit and its replacement by the Multi-Sector General Permit,  resulting in the re-submission of NOIs
by existing facilities - this also means that from 1996-98,  some facilities would be double-counted. In
addition, the number of NOIs in any given year also largely reflects enforcement efforts to obtain
compliance from key sectors that are subject to  these requirements.  Thus, the number of NOIs in the
data base for a given year is not a good indication of the number of new facilities.  However, the
distribution of facilities by SIC  in the NOI data base should provide an indication of the importance of
both new and existing facilities  from a storm water standpoint.

       Limited additional information is available regarding the number of new establishments by SIC.
The Small Business Administration contracted with the U.S. Bureau of the Census to develop special
tables to provide statistics regarding new establishments.  The most directly relevant SBA table  provides
statistics on the number of "births" of establishments by 3-digit  SIC. Unfortunately, these statistics are
too aggregate for this analysis, which is at the 4-digit SIC level. The next most relevant SBA table
provides the total number of establishments by 4-digit SIC for 1995 and  1996. The difference in the
number of establishments  in a 4-digit SIC between 1995 and 1996 reflects the difference between
"births" and closures of establishments for 1996.  While this difference would be an underestimate of the

-------
number of new establishments and is only available for one year (1996), it nevertheless should generally
provide a reasonable proxy for the number of new establishments annually. In cases where the
difference is negative, this indicates that there were more closures than births, and we generally would
interpret this to mean that it is not likely that there were many births.

        Attachment 2 lists the number of facilities by all of the 383 SICs (mostly 4 digit SICs, some 2 or
3 digit SICs) represented in the NOI data base for 199613 as well as the net change in facilities nationally
for these SICs from the SBA data base - the SICs are shown in order of the number of facilities in the
NOI data base. Exhibit 1-6 provides the results for the top 39 SICs shown in Attachment 2, representing
about 75% of the facilities in the 1996 NOI data base. The top 39 SICs represent at least 21,800 new
establishments in 199616, with over 75% of these establishments occurring in four SICs: SIC 4215
Courier Services, SIC 4212 Local Trucking Without Storage, SIC 4213 Trucking Except Local, and SIC
4225 General Warehousing and Storage. It is noteworthy that the growth in new establishments for these
five  industry groups is significantly higher than average: the overall growth in establishments from 1995
to 1996 was 1.9% while the growth in these five SICs was 7.6% - whether this growth rate will continue
is unknown. All four of these SICs are in Sector P, Subsector 3 (Land Transportation: Motor Freight and
Warehousing) as defined in EPA's general permit for storm water.

        Note that of the remaining 344 SICs  comprising the  remaining facilities  in the NOI data base,
only 45 of the SICs have 10-26 facilities, 61  of the SICs have 5-9 facilities, 72 of the SICs have 3-4
facilities, 58 of the SICs have 2 facilities, and 111 of the SICs have 1 facility. It seems likely that
generally, the facilities in these SICs typically do not pose storm water problems, and as shown in
Attachment 2, they are predominantly small businesses. For example, there are only 3 eating
establishments in the  1996 NOI data base, and 98% of existing eating establishments are owned by small
businesses.  Even though there are many new eating establishments annually (a net gain of over 27,000
in 1996) it would not seem reasonable to view this sector as requiring offsets. Therefore, we focus the
analysis on SICs that represent 75% of the facilities in the NOI data base.

        To estimate the total number of new facilities that could be subject to industrial storm water
requirements, we assume that the 21,800 new facilities for the SICs that represent 75% of the facilities,
also  represent 75% of all of the facilities subject to industrial storm water requirements. Therefore, the
total number of new establishments potentially subject to industrial storm water requirements is
estimated to be about 29,000.

2. The vast majority of the facilities in these  SICs are small businesses

        Exhibit 1-6 (and Attachment 2) also shows the percentage of the businesses within each of the 4-
digit SICs that are small businesses.  In nearly all of the SICs, nearly all of the businesses are small
according to the SBA's definitions.  If the percentage of small businesses in an SIC can be used as an
indicator of the percentage of new establishments for that SIC, then perhaps 600 of the 29,000 new
        15 The 1996 NOI data base contains 6,717 facilities with 463 entries as SIC codes. However, 531 facilities
did not have recognizable SIC codes, resulting in auseable data set of 6,186 facilities in 383 SIC codes.

        16Note that several 2-digit SICs are represented in the table. The difference in establishments between
1995 and 1996 fora 2-digit SIC may not be representative of the difference for a 4-digit SIC.

-------
                                             Exhibit 1-6
                   SICs representing 75% of the Facilities in the 1996 NOI Database n
FacIBtiesin
NOI
...
SIC SIC definition
4581 Airports, Flying Fields, and Airport Terminal Services
5015 Motor Vehicle Parts, Used
3273 Ready Mixed Concrete
5093 Scrap and Waste Materials
4512 Air Transportation. Scheduled
2951 Asphalt Paving Mixtures and Blocks
4215 Courier Services, Except by Air
4513 Air Courier Services
3732 Boat Building and Repairing
1442 Construction Sand and Gravel
4212 Local Trucking Without Storage Except Garbage and Refuse Collection, Without Disposal
4213 Trucking. Except Local
4011 Railroads, Line-Haul Operating
4311 United States Postal Service
4522 Air Transportation, Nonscheduted PT, Offshore Marine Air Transportation Services
4231 Terminal and Joint Terminal Maintenance Facilities for Motor Freight Transportation
3271 Concrete Block and Brick
2421 Sawmills and Planing MiUs, General
^389 Oil and Gas Field Services, N.E.C.
^653 Corrugated and Sofid Fiber Boxes
3272 Concrete Products. Except Block and Brick
4952 Sewerage Systems
1422 Crushed and Broken Limestone
45 MAJOR GROUP 45 - TRANSPORTATION BY AIR
2048 Prepared Feeds and Feed Ingredients for Animals and Fowls. Except Dogs and Cats
4225 General Warehousing and Storage
3731 Shipbuilding and Repair of Nudear Propelled Ships except Shipbuilding of Nonnudear
Propelled Ships and Nonptopelted Ships
5171 Petroleum Bulk Stations and Terminals
4151 School Buses
4493 Marinas
4200 MAJOR GROUP 42 - MOTOR FREIGHT TRANSPORTATION AND WAREHOUSING
4953 Refuse Systems
2851 Paints, Varnishes, Lacquers. Enamels, and Allied Products
1429 Crushed and Broken Stone, N.E.C.
3479 Coating, Engraving, and Allied Services, N.E.C.
2499 Wood Products. N.E.C.
•1013 Railroad Switching and Terminal Establishments
2491 Wood Preserving
3441 Fabricated Structural Metal

SBA
definition
$5.0 Mil
100 ppl
500 ppl
100 ppl
X1. 500 ppl
500 ppl
$18.5 Mil
1,500 ppl
500 ppl
500 ppr
$18.5 Mil
$18.5 Mil '
1,500 ppl
-
1,500 ppl
$5.0 MB
500 ppl
500 ppl
$5.0 Mil
500 ppl
500 ppl
$5.0 Mil
500 ppl
-
500 ppl
51 8.5 Mil
1,000 ppl

100 ppl
$5.0 Mil
S5.0 Mil
-
S6.0
500 ppl
500 ppl
500 ppl
500 Dpi
500 o?l
500 ppl
500 ppl

Expected*
Difference in Owned by
Database, Establishments,
% fikety small
92%
100%
98%
97%
Approximately 94%
92%
Between 98% and 99%
Approximately 99%
99%
97%
Between 98% and 99%
Between 94% and 98%
N/A
N/A
Approximately 98%
84%
95%
98%
97%
95%
98%
90%
92%
-
94%
Between 95% and 98%
Approximately 97%

92%
97%
98%
-
Between 79% and 84%
95%
92%
97%
S8%
N/A
96%
QR%
Total:
1996
715
576
397
273
263
213
202
199
148
136
125
121
99
82
79
76
70
63
63
61
61
54
52
44
42
40
40

33
32
32
31
31
31
30
29
27
26
26
?6
4.653
1995-1996
433
192
220
415
233
49
1046
136
349
145
7912
2593
no data available
no data available
236
10
negative
negative
318
64
negative
166
0
no data available
negative
1325
68

negative
94
450
4835
negative
109
39
42
necaCve
no data available
23
7RP.
21.800
Large
Entities
35
0
4
12
14
4
16
*
-
'
11!
10-
N//
N/,


NJ.
N/
•^

N,
*

N.
N



N


r-
r-



'
'




        ''Facilities without recognizable SICs (under 10% of database) were excluded.
26

-------
establishments are owned by large businesses, representing 2% of all new establishments. While new
establishments might also be owned by small businesses in a similar proportion, it is also possible large
businesses are responsible for a larger share of new establishments. To reflect the possibility that larger
businesses have a greater representation among new establishments than among existing establishments,
we assume large businesses own 2%-4% of the new establishments that might be  subject to industrial
storm water requirements, which represents 600 - 1,200 establishments.

3. While these facilities discharge a variety of pollutants, the predominant pollutants are oil and heaw
metals, and perhaps 10-15% of these facilities might contribute to impairments.

        The predominant pollutants for the bulk of the new businesses identified ir> Exhibit 1-6 are
oil/fuel and heavy metals, but might include other pollutants such as chlorinated solvents and organics.
The National Water Quality Inventory Report to Congress (U.S. EPA,  1998) reports provide perspective
on the sources and pollutants that are impairing the nation's waters.  Exhibit 1-7 provides perspective on
the extent to which urban runoff contributes to impairment, as well as the pollutants of oil/fuel and
metals. However, as described earlier, the waters that are reported in the 305(b) report represent an over-
estimate of the waters that are actually impaired to the point that a TMDL is needed. Of all the water
bodies that are actually scheduled for development of TMDLs (i.e., those on the 1998 303(d) list), about
10% identify metals as a cause of the impairment and less than 1% identify oil and grease as a cause.

        If all of the new establishments discharged to listed water bodies, then perhaps as much as  15%
of the facilities might be discharging pollutants that contributed to impairment. This would likely be an
overestimate to the extent that these establishments are distributed nationally more in proportion to
population and population growth rather than concentrated in areas where there are impaired waters.  On
the other hand, these facilities overall discharge a larger variety of pollutants than metals and oil, and
some of these may contribute to impairments. Therefore, for this analysis, we assume that 15% of the
facilities might be located in a manner as to contribute to the impairment of a water body. For the 600-
1,200 establishments owned by large businesses, this represents 90-180 establishments.

4. Perhaps 45-90 facilities annually mav be required to provide offsets.

        If 90-180 facilities owned by large businesses discharge pollutants of concern to impaired water
bodies, then they may be required to obtain offsets. However, EPA's Economic Analysis of the Final
Phase II Storm Water Rule estimates that on average 44% of the facilities (EA Exhibit 9-2) potentially
subject to industrial storm water requirements are expected to qualify for a storm water exemption on the
grounds that they have "no exposure" of their industrial activities to storm water. Further, in the category
of vehicle maintenance, representing over half of the  new facilities, 80% of the facilities (EA Exhibit 9-3)
are expected to qualify for this exemption. Therefore, if conservatively only half of the  90-180 facilities
qualify for the no exposure exemption, then perhaps 45 - 90 facilities might be required  to obtain offsets.

5. Offsets mav be available from non-industrial storm water sources.

        Compliance Cost

        The storm water requirements only apply to industrial sources, and not other sources such as
commercial establishments. Therefore, for example, it may be possible for industrial vehicle
maintenance facilities to obtain offsets from similar commercial vehicle maintenance facilities. To

-------
                                          Exhibit 1-7
 % of the Nations Waters Where Urban Run-off, Oil/Fuel and Metals Contribute to Impairments
Source/
Pollutant
Urban Runoff/
Storm Sewers
/
Oil
& Grease
Metals
Water body Type
Rivers (miles)
Lakes (acres)
Estuaries (sq. mi.)
Shoreline (miles)
Rivers (miles)
Lakes (acres)
Estuaries (sq. mi.)
Shoreline (miles)
- Rivers (miles)
Lakes (acres)
Estuaries (sq. mi.)
Shoreline (miles)
Impaired Miles, Acres or Sq. Mi. as a Percent Of:
Total Waters
<1%
3%
13%
1%
. <1%
<1%
5%
0%
1%
8%
4%
0%
Surveyed Waters
5%
8%
18%
7%
1%
1% .
8%
1%
6%
20%
5%
1%
Impaired Waters
13%
21%
46%
55%
2%
3%
20%
11%
16%
51%
13%
10%
 estimate the cost of offsets for the 45 - 90 facilities that may be required to obtain them, we assume that
it is sufficient for these facilities to secure BMPs for similar commercial sites, and that the average cost
of such offsets can be represented by the costs for BMPs for industrial storm water.   As shown in the
 Economic Analysis of the Final Phase II Storm Water Rule (EA Exhibit 9-6), the cost of storm water
BMPs across a wide range of industrial subsectors can be represented by a broad range of costs from
$1,023-548,949 annually.  Since the facilities requiring offsets will likely seek lower-cost offsets, it
seems unlikely that offsets in excess of $10,000-$30,000 would be obtained (March, 1999 dollars,
including transactions costs of 45%). Therefore, the average cost of obtaining offsets for the 45-90
facilities might range from about $450.000 to $2.7 million annually.

       If each year, an additional 45 - 90 facilities incur annual costs of $450,000 to $2.7 million then
the annual cost would increase to $900,000 - $5.2 million in the second year, and so on. However, each
year states would make progress in developing TMDLs.  If TMDLs are developed uniformly for all the
current impaired water bodies from 2000 through 2015, then the number of water bodies requiring offsets
would be declining steadily at a rate of about 6% per year (assuming that there are few new water bodies
listed during this period). The result is that the annual cost of offsets steadily increases for about 8 years
to SI - S6.0 million, but then declines until it is eliminated  in 2015.  Using the OMB-recommended
discount rate of 7% annually, the present value of these offsets would be S13.3-S80.0 million,  and the
annualized cost through 2015 would be S1.41-S8.47 million.

       Administrative Cost

       The 45 - 90 new facilities annually that will be required to provide offsets will also be required to
have an individual NPDES permit. It is assumed in this analysis that the  current costs associated with
complying with general permits will still be  incurred. Dischargers of storm water associated with an

-------
industrial activity that wish to obtain an individual permit must submit Form 2F. Based on the December
3, 1998 ICRfor the National Pollution Discharge Elimination System Discharge Permit and the Sewage  ]
Sludge Management Permit, the applicant burden for preparing an NPDES application to discharge storm
water is 28.6 hours at an average rate of $36.12yhour for the private sector. Assuming that the 45 - 90
facilities each obtain their offsets from a single source, and that the sources providing the offsets must
also provide the same information in order for the permit authority to fully evaluate the adequacy of the
proposed offsets, tfien the work associated with 90-180 individual NPDES permits would be needed.
This amounts to $92,973  - $ 185,946 per year.

        We assume in this analysis that  these permits will need to be renewed every five years.
However, as discussed above, as TMDLs are developed, the need for offsets will also steadily decline.
Taking both of these factors into account results in a present value of $0.8 -$1.6 million, and an
annualized cost through 2015 of $85,242 - $170,484.

        Total Cost

        The total compliance and administrative  costs are estimated to range from $1.5-$8.64 million.

6. Cost to Permit Authorities

        The corresponding additional cost to permitting authorities for processing 45-90 individual
NPDES permits and 45-90 offsets is based on the burden of 2.2 hours as estimated in the above ICR for
evaluating Form 2F at an average rate of $31.68/hour for federal or state employees, and totals $6,273-
$12,545. Taking into account a 5-year renewal cycle and the  development of TMDLs through 2015
results in a present value  of $54,331-$ 108,654, with an annualized cost of $5,751-$11,502.

        In addition, the permit authorities would review the NOI information from the sites providing the
offsets. As  specified in the April, 1999  ICRfor Phase 11 of the NPDES Storm Water Program, the
burden to evaluate NOI information is 1.5 hours at an average rate of $31.68/hour for federal or state
employees.  For the 45-90 facilities providing offsets, the cost would be $2,138 - $4,277. Taking into
account a 5-year renewal cycle and the development of TMDLs through 2015 results in a present value of
$ 18,521 -$3 7,042, with an annualized cost of $ 1,961 -$3,921.

        Thus, the overall annualized cost to permit authorities from 2000 to 2015 be $7,712-$15,423.

7.  Summary

        The annualized costs associated with the proposed  offset requirements for storm water
discharges from industrial sites (other than construction) is as follows:
Type of Cost
Cost To Private Sector
Cost To Permit Authorities
Total Cost
Annualized Cost
(S Millions)
1.50-8.64
0.008-0.015
1.51 - 8.66

-------
THE POTENTIAL COST OF OFFSETS FOR PROCESS DISCHARGES

       A new source or a significantly expanded source that discharges pollutants of concern (i.e. the
pollutants for which a water body is impaired) directly into that water body" must obtain a 1.5:1 offset
for the discharge of those pollutants. The analysis of offset requirements for new or significantly
expanded existing dischargers, generally follows the basic steps outlined earlier in this section:

        1.      estimate the number and type of new sources and significantly expanded existing
               sources,

       2.      estimate the proportions of these sources which might discharge into impaired water
               bodies,

       3.      estimate the proportion of the sources discharging into impaired water bodies that are
               discharging the pollutants of concern,

       4.      estimate the magnitude of the offsets that might be required,

       5.      estimate the proportion of the dischargers which are large businesses, and therefore
               subject to the offset requirement,

       6.      estimate the annualized unit cost for offsets for different pollutants, and

       7.      estimate the administrative and compliance costs of the offsets.

       Because of the diversity of the regulated community (representing hundreds of 4-digit SICs) in
combination with the diversity of the pollutants discharged that might require offsets, we use a
retrospective analysis to provide context for making these projections. Specifically, we begin the
analysis by answering the following question: if the proposed offset requirement were in effect during
1995-98, how many new or significantly expanding sources would have been required to obtain offsets,
and what would their characteristics have been? Using these results as context, we estimate a range for
the number of future offsets that might be required annually and their potential cost.

1.  Retrospective Analysis For New and Significantly Expanding Sources. 1995-98

       Overview of Approach

       For this analysis, we assume that recent experience over the past few years regarding new and
expanded sources would be indicative of the  number and type of new and expanded sources that might  be
expected in the future. Using EPA's Permit Compliance System (PCS) data base for  1995-98, we
identified the new and expanded sources that were pen-fritted in each year and extracted key information
about them including their 4-digit SIC, location, design flow and monitoring data.19 Given  that the four
        18 "Directly into" means that the facility is actually situated on or within the extent of the impairment.

        ''Permits issued with new permit numbers were assumed to be for new facilities - this tends to overstate
the number of new facilities because sometimes new permit numbers are assigned when issuing revised permits to

30

-------
years including 1995 through 1998 were generally years of robust economic growth in the U.S., the
number of new facilities for this period may be an over-estimate of long-term trends regarding the
number of new sources.  This analysis does not consider industry-specific factors, such as a specific
industry's business cycle.

        For the facilities with adequate location information we determined whether they discharged into
impaired water bodies.  To determine whether a particular facility discharges directly into an impaired
water body it is necessary to know the location of the facility (i.e., the latitude and longitude coordinates
of the discharge) and the location of any nearby impaired water bodies (i.e. the latitude and longitude for
the upstream and downstream ends of impaired water bodies).  The PCS data base includes latitude and
longitude information for about half of the new sources and nearly all of the expanded existing sources;
ana iocatioif information is available for all of the 303(d) listed water bodies-except those located in EPA
Region 10.20 Therefore, for all of the facilities with latitude and longitude information that are located in
Regions 1-9, we were able to determine whether they actually discharge into impaired water bodies.

        For new sources that discharge into impaired water bodies, we obtained more detailed facility-
specific information from the appropriate state permitting authorities to determine whether they were
really new process .dischargers and whether they were discharging pollutants of concern.  The result was
extrapolated to the other half of the facilities we could not analyze (due to incomplete location data) to
obtain an overall estimate of the number and type of facilities that would likely have been required to
obtain offsets from 1995-1998 had the proposed regulation been in effect at that time.  For significantly
expanding facilities, we did not identify any which discharge into impaired water bodies.

        The following discussion first provides the results of the analysis for new sources and then for
substantially expanded existing sources.

        New Sources

        Over the 4-year period, we identified 4,345 new dischargers, of which 65 were classified as
major dischargers. The results for the top 10 SICs comprising about 1/3 of the facilities is summarized in
Exhibit 1-8 on the  next page and shown  in greater detail in Attachment 3 Characteristics of New Process
Water Dischargers for 1995-98.

        This represents an average of 1,086 new sources per year.  These dischargers represented over
400 SIC groups —  in more than half the cases,  there were only one or two permittees in a SIC over the 4-
year period. About 20% of these dischargers did not have an SIC listed for them. As shown in Exhibit I-
8, The greatest number of new dischargers (that included SIC information in their PCS data) were
sewerage systems, representing about 13% of the new permittees.  Water supply, gasoline service
stations, construction sand and gravel, and bituminous coal and lignite surface mining represented the
next tier of permittees at about 3-4 % of the total each.
existing facilities, or when issuing individual NPDES storm water permits to existing facilities. We used change in
flow as a proxy to identify significant expansions — modifications of permits for flow change were identified and
evaluated to determine whether the change represented an increase of either 50% or more, or 20% or more.

        •° As of
-------
                                             Exhibit 1-8
                    Top 10 SICs Among The 4,345 New Sources for 1995-1998 21
SIC
4952
4941
5541
1442
1221
1422
4953
5171
5172
4911
SIC Definition
Sewerage Systems
Water Supply
Gasoline Service Stations
Construction Sand and Gravel
Bituminous Coal and Lignite Surface Mining
Crushed and Broken Limestone
Refuse Systems '
Petroleum Bulk Stations and Terminals
Petroleum and Petroleum Products Wholesalers,
Except Bulk Stations and Terminals
Electric Services
Numbers
% of 4,345
556 / 12.8%
1667 3.8%
1557 3.6%
147/5.4%
140 7 3.2%
79 7 1.8%
69 7 1.6%
60 7 1.4%
53 / 1.2%
53 7 1.2%
SBA Definition
$5.0 Million or serving 50,000
$5.0 Million
$6.5 Million
500 employees
500 employees
500 employees
$6.0 Million
100 employees
100 employees
4 Million MW hours
% Likely
to be small
90%
97%
92 - 94%
97%
92%
92%
79 - 84%
92%
92%
82%
        Adequate location data (i.e. for both the facility and the impaired water bodies) were available
for 2,404 of the 4,345 new facilities. Using the location data, we determined that only 17 of the 2,404
facilities discharge into impaired water bodies.22  All of these facilities are minor dischargers. State
permitting authorities were contacted to obtain additional facility-specific information about these 17
facilities as well as to confirm the PCS data.  As of the date for this analysis, supplementary information
was obtained for 15 of the facilities. Using the supplemental information for the 15 facilities, we
determined that four were not actually new process water dischargers.23  Therefore, at most 13 of the  17
facilities are new process water dischargers into impaired water bodies.

        Thus, perhaps 13 of the 2,404 facilities we could analyze discharge  into impaired water bodies.
        21 This tally excludes 904 facilities (21% of total) that did not provide SIC information, 83 elementary' and
secondary schools (SIC 8211, 1.9% of total), and 68 non-classifiable establishments (SIC 9999, 1.6% of total).

        - Note that the new facilities dated from 1995-1998, while the impaired water bodies used for this analysis
is based on the most recent 1998 303(d) list. Therefore, this analysis tends to overstate the number of new sources
discharging into impaired water bodies, to the extent that new sources would have avoided such locations.

        23 Two of the permits were for storm water discharges (and one of these permits was for a facility that was
not new).  The third facility, an open water aquatic animal production "net pen" with no treatment requirements
should not have been permitted with an NPDES permit, according to the State permit writer.  The fourth facility was
an existing facility that had been pretreating, but decided to become a direct discharger (we assume in this analysis
that the discharger did not increase pollutant loadings by 20% or more when switching, and therefore would not
have been subject to offsets due to a significant expansion).

-------
Scaling to the rest of the water bodies for which adequate location information was not available,
in an estimate of 23.5 2t new facilities over 4 years or an average of 5 to 6 new sources per year that
started discharging to impaired water bodies during 1995-98 .  If the proposed offset requirements had
been in effect during this period, these facilities might have been required to obtain offsets if 1) they
discharged the pollutants which are impairing these water bodies, and 2) if they were not small
businesses.  These additional factors are discussed below.

        Of the 13 facilities identified above we were able to obtain additional facility-specific
information for 11 facilities. For these 11 facilities, 5 do not discharge pollutants that contribute to the
impairments, 5 of the facilities contribute to  1 cause, and 1 facility contributes to 2 causes. For the two
facilities for which more  detailed information could not be obtained, we assume thaVl of these facilities
contributes to 1 cause.25  Therefore, of the 13 facilities, we estimate that 7 facilities are contributing to 8
causes. Note that none of these facilities is a major discharger. If we extrapolate from this result to the
4,345 new facilities from 1995-98, then about 3 facilities per year would be required to obtain, a total of
3-4 offsets26, if all of these facilities were large businesses.  However, some portion of these facilities is
likely to be owned by small businesses and will not be required to  obtain offsets.

        To determine the size of the businesses that owned the 17 facilities that discharge to impaired
water bodies, we used Dun & Bradstreet data where possible.  Using this data, we could make a size
determination for 11 of the facilities, finding that perhaps 5 of the  11 facilities (nearly half) are owned by
small businesses.27  For the 7 facilities that were found to be contributing to impairments, we determined
that 3 of the facilities are large entities, but could not determine the size of the remaining 4 facilities
(which by itself may indicate that these are small businesses). If we assume that about half of the 7
facilities are owned by small businesses, then perhaps 4 of the 7 facilities that contribute to impairments
would be required to obtain 5 offsets. If we extrapolate this result to the 4,345 new facilities from 1995-
98, then perhaps 2 facilities per year would have been required to obtain a total of 2-3 offsets.

        Thus, had the proposed offset requirement been in effect for the four years from 1995-98,
perhaps 2 new facilities per year that might have been required to obtain a total of 2-3 offsets. All of
these facilities are minor dischargers. To the extent this experience is indicative of the incidence of the
number of facilities in the future which might require offsets,  it does not appear as though there will be
many per year.  Moreover, the 1995-98 analysis was based on the full list 1998 303(d) list of impaired
        24 13 x (4,345/2,404)

        25 In PCS, neither of the facilities had their SICs listed or had showed any monitoring requirements.
However, one of the facilities is a YMCA. and it is unlikely that it is contributing to the listed pathogen problem in
the receiving water body. We are unable to determine the type of the second facility, however, to be conservative,
we assume that it is contributing to the metals problem in the receiving water body (but not contributing to the PCB
and chlordane problems which were also listed for the water body).

        ~6 1 x (4,345/2,404)/4 = 3.16 facilities per year & 8 x (4,345/2,404)/4 = 3.61 offsets per year

        27This is generally a substantially larger fraction than might be expected based upon the representation of
small businesses across the majority of SICs as shown in Attachment 1, which generally shows that small businesses
account for more than 90% of the existing facilities in most SICs. This may indicate that newer facilities may be
more likely to be owned by larger entities.

-------
water bodies. All of these impaired water bodies are scheduled to have TMDLs developed for them by
2015. Once TMDLs are developed, the TMDL requirements for these water bodies will supersede the
offset requirements associated with this proposal. Therefore, it is likely that the number of additional
new facilities annually that would be required to obtain offsets or those that are required to maintain
offsets as a result of the proposed rule would be continuously declining through 2015.28

       Expanded Sources

       Over the four-year period 1995-98, we  identified 27 facilities that increased their flow by 50% or
more, and an additional 2 facilities that increased their flow by 20-50%. These 29 dischargers represent
11 SIC groups as shown in the Exhibit 1-9. Adequate location data were available for 28 of the 29
facilities. None of the 28 facilities were found to be discharging directly into any impaired water bodies.
                                           Exhibit 1-9
                    SICs for the 29 Expansions of 20% or More for 1995-1998
SIC
4952
4941
9199
8999
7033
6199
4785
3674
3011
2911
2865
SIC Definition
Sewerage Systems
Water Supply
General Government, NEC
Services, NEC
Recreation Vehicle Parks and Campsites
Lessors of Real Property
Fixed Facilities and Inspection and Weighing Services
for Motor Vehicle Transportation
Semiconductors and Related Devices
Tires and Inner Tubes
Petroleum Refining
Cyclic Organic Crudes and Intermediates, and Organic
Dyes and Pigments
Number &
%of29
18/62%
21 7%
M 3%
M 3%
1/ 3%
1/ 3%
1/ 3%
1/ 3%
1/ 3%
1/ 3%
1/ 3%
SBA Definition
$5.0 Million or 50,000 served
$5.0 Million
N/A
$5.0 Million
$5.0 Million
$5.0 Million
$5.0 Million
500 employees
1,000 employees
1,500 employees
750 employees
% Likely to
be small*
90%
97%
N/A
97%
99%
95%
90%
89%
92%
91%
86%
Source: Two sets of data prepared by the Bureau of the Census for the Small Business Administration: data for
1995 provides the number of firms within ranges of annual receipts; and data for 1996 does the same for
employment.                                                          /
        28 While there may be new waters listed during this period, it is generally anticipated that more water
bodies will be removed from the list than added to it.
34

-------
        Conclusions

        It does not appear that there will be a significant number of new or significantly expanded
facilities annually that will be required to obtain offsets. For the 1994-98 experience, there were fewer
than 1,100 new facilities annually and fewer than 8 significant expansions annually.  For these facilities,
there were fewer than 6 new facilities that discharged into impaired water bodies and none of the
significant expansions discharged into impaired water bodies. For the 6 new facilities discharging into
impaired water bodies, only 2 of the facilities might have been required to obtain a total of 2-3 offsets
because they contributed to the impairments and were owned by large businesses.  Further, nearly all of
the new facilities were  minor dischargers (99.5%), and all of the facilities that discharged into impaired
water bodies were minor dischargers. These results strongly indicate that there will not be many facilities
annually that will be required to obtain offsets29, and these are likely to be minor dischargers.

        We cannot reliably draw additional conclusions from this analysis. The number of facilities that
discharge to impaired water bodies is too small to provide a reliable basis for extrapolating the specific
characteristics of these facilities (e.g., 4-digit SIC, pollutants discharged, etc.) to future new facilities.

2. Approach for Estimating the National Annual Cost of Offsets.

        Based on the 1995-98 experience, it does not appear likely that the number of new or
significantly expanding facilities requiring offsets will exceed 2 facilities for a total of 2-3 offsets
annually. However, there is no good basis for anticipating the specific SICs that these facilities will
represent, the specific pollutants that these facilities will be discharging or the offsets that will be
required due to the specific impairments of the water bodies that they will be discharging into.
Fortunately, it is not necessary to make such projections in order to develop an overall estimate of the
national average cost of the proposed offset requirement. Regardless of the specific characteristics of the
facilities that will be required to obtain offsets, we know that they will discharge pollutants of concern to
impaired water bodies.  Therefore, we know three important sets of facts about these facilities and the
impaired water bodies that they will be discharging  into:

        1)      We know that these facilities will be achieving limitations equivalent to water quality
                criteria at end of pipe. Therefore, if we can estimate the flow of the facilities, we can use
                the concentrations allowed at the end of pipe to determine the load that these facilities
                must offset.  We can use representative flows across all SICs to develop estimates of the
                loads that these facilities are discharging. Since we  know that they must be discharging
                pollutants of concern (or they wouldn't need offsets), we need only estimate the loads for
                these pollutants, as discussed below.

        2)      The likelihood that facilities will be located on  water bodies that are impaired by specific
                pollutants can be estimated by using the national distribution  of the  causes of
                impairments. Further, the causes that are likely to require  offsets will be those pollutants
        29The data clearly indicate that few new NPDES permits for process water discharges have been issued to
new facilities discharging "into" impaired waters. The definition of "into" is a key determinant of the number of
facilities affected. If "into" also encompassed upstream sources that might affect a downstream impaired water
body, then the number of new facilities from 1995-98 that might have been required to obtain offsets would increase
from 2 facilities annually to about 100 facilities annually.

                                                                                                  35

-------
               that these facilities typically discharge.

        3)      Nationally, based on the 1998 303(d) listed waster bodies, there are 20,929 water bodies
               that are impaired by 40,226 causes, or an average of about 2 causes per water body.
               However, as discussed later, many of these causes are not necessarily related to
               pollutants from process dischargers. Further, the 1995-98 analysis indicated that perhaps
               half of the facilities that must obtain offsets may need to provide two offsets. Therefore,
               we apply the results of the 1995-98 analysis and assume that half of the facilities that
               provide offsets will provide one offset, and the other half will provide two offsets.

With this approach, we can use the magnitude of discharge flows in combination with water quality
criteria to estimate the total annual loads for each pollutant that must be offset. Assuming that non-
nutrient offsets will generally be obtained from existing point sources and that nutrient offsets can be
obtained from agricultural sources, we can estimate the unit cost of offsets for specific pollutants.
Combining these unit costs with the loads that result from various flows, we can estimate the cost of
associated offsets.  Each step is described more fully below.

3. Assumed Number of Facilities That Will Require Offsets and the Number of Offsets Needed.

        If the. proposed offset requirements had been in effect during 1995-98, perhaps 2 new facilities
annually would be required to obtain a total of 2-3  offsets, and no significantly expanded facilities would
be required to provide offsets.  Further,  all of these facilities would be minor dischargers.

        To provide a conservative estimate of the cost of the offset requirement, we assume that twice
the number of facilities and offsets will be needed annually than would have been required from 1995-98.
Thus, we base the cost  on 4 new or significantly expanded facilities which must obtain 6 offsets
annually. From the year 2000 through 2015,  this amounts to a total of 64 facilities providing 96 offsets.
Further, we use a range to estimate cost, assuming for the low cost that all of the 4 facilities are minor
dischargers while the high cost is assumes that one of the 4 facilities is a large major discharger.

4. Flow Distribution of Process Dischargers

        The available data regarding design flow for the new facilities from 1995-98 is limited.
However, the distribution of new sources by 4-digit SIC is roughly similar to the distribution in PCS
generally. Therefore, we used all available data in PCS for design flow to estimate the flow
characteristics for existing industrial dischargers, and we assume that new industrial dischargers will
span a similar distribution for flow. Further, using data from EPA's 1996 Clean  Water Needs Survey
Report to Congress we can describe the flow characteristics of planned expansions for existing POTWs
and for planned new POTWs through 2016.  Exhibit I-10 provides the resulting average flow by quartile
for major and minor industrial dischargers, as well as the average flow by quartile for planned POTW
expansions and planned new facilities.

        For this analysis, we use the average flows associated with industrial dischargers to represent the
flows from new and significantly expanded industrial sources and from new and significantly expanded
POTW sources for several reasons. From Exhibit I-10, it is clear that the span of average flow by
quartiles for POTW expansions and new facilities  lies within the span of flows for major and minor
industrial facilities. Furthermore, 87% of the new facilities in 1995-98 were industrial facilities. Finally,
for the significantly expanded facilities from 1995-98, the increased flow ranged from  .006-3 MOD, so

36

-------
that it appears likely that the flow increases associated with such expansions also falls within the flows
represented by the major and minor industrial facilities.

                                           Exhibit MO
                     Flow Distribution for Industrial Dischargers and POTWs
DISCHARGER CLASS
INDUSTRIAL1
POTW2
Minor
Major
Future Expansions
Future New Facilities
AVERAGE FLOWS (MGD)
BY QUARTILES
1st
0.010
0.676
0.021
0.006
2nd
0.077
1.474
0.052
0.029
3rd
0.327
4.097
0.109
0.153
4«n
1.151
10.652
1.997
4.337
        1 Data based on actual or imputed flow data for 5,020 minor and 887 major industrial dischargers.
        - Data from U.S.EPA, 1996 Clean Water Needs Survey Report to Congress

        Therefore, to represent the flows for the 4 facilities assumed to require offsets, we assume that
each of the four facilities can be represented by the average flow in each of the four quartiles for
industrial sources. Thus, for the low case, the total flow for 4 minor dischargers would be 1.6 MGD (one
facility with a flow of 0.01 MGD, one with a flow of 0.077 MGD, one with a flow of 0.327 MGD, and
one with a flow of 1.151 MGD).  Similarly, for the high case, the total flow including 1 large major
discharger (using the largest quartile flow of 10.652 MGD for the major dischargers instead  of the
smallest quartile flow of .010 MGD for the minor dischargers) would be 12.2 MGD.  Therefore, the
range of flow assumed for the 4 facilities that will be required to obtain offsets is 1.6 - 12.2 MGD.  We
believe that this generally represents a conservative range for the total flow of the facilities potentially
needing offsets since 1) it is based on twice number of facilities requiring offsets for the 1995-98
experience, and 2) the upper end of the range assumes that each year one of the facilities will be a large
major discharger when none of the facilities that might  have required offsets in 1995-98 were major
dischargers.  This  range is provided, in part to provide for the possibility that a single new facility might
have a flow that is much larger than the averages shown in Exhibit MO30 — each year, the range provides
an extra 10 MGD  beyond  what might be expected based on the 1995-98 experience, amounting to 160
MGD from 2000 through the year 2015, providing an allowance for much larger facilities than average.

5. Pollutants for Which Offsets Are Most Likely To Be Required

        As discussed in the next subsection, the cost associated with offsets for different pollutants can
vary widely. Therefore, it is necessary to estimate the frequency with which offsets for different types of
pollutants will be  needed. The most straightforward way to estimate this is to examine the frequency
with which water  bodies are impaired by the pollutants that the new and significantly expanded process
waste water sources tend to discharge. Exhibit 1-11 on  the next page shows the incidence of each of the
        30The maximum planned expansion identified in the Needs Survey was for an increased flow of 158 MGD,
and the maximum planned new facility was for 193 MGD. Such flows, however, are rare.
                                                                                              37

-------
causes of the impairments identified in the 1998 303(d) lists of impaired water bodies. For each cause,
the exhibit indicates the likelihood that new or significantly expanded process waste water sources will
discharge the pollutants of concern.  There are many causes which these sources are less likely to
contribute to: for example, habitat alteration, flow alteration and biological impairments are not likely to
be caused by pollutants these sources discharge; pH and temperature are not likely to be pollutants which
will require offsets; pesticides and pathogens are not typical pollutants for these sources; and others such
as legacy, fish consumption advisory, "other" are not specific enough to be addressed directly in this
analysis.

                                          Exhibit Ml
              Frequency of Reported Causes of Impairments in the 1998 303(d) Lists
Reported Cause of
Impairment
Sediments
' Nutrients
Pathoeens
Metals
Dissolved Oxveen/BOD
Other
Other Habitat Alteration
uH
Temperature
Biological Impairment
Fish Consumption Advisorv
Flow Alterations
Pesticides
Ammonia
Leeacv
Unknown
Oreanics
Toxics
Inorganics
Total
Frequency
16.0%
14.1%
12.0%
9.9%
9.5%
5.4%
5.3%
4.4%
4.3%
3.3%
3.1%
3.0%
2.7%
1.9%
1.3%
1.3%
1.1%
1.0%
0.5%
100.0%
Potential Contribution
By Process Waste Water
Dischargers
Hieh
Hish
Low
Hieh
Hish
Unknown
Low
Low
Low
Low
Unknown
Low
Low
Hish
Unknown
Unknown
Hish
Hieh
Hish -

       New or significantly expanded dischargers are most likely to discharge pollutarfts that contribute
to impairments due to sediment nutrients, dissolved oxygen, metals, ammonia, organics, toxics and
inorganics. As can be seen from Exhibit 1-11, these pollutants contribute to about 54% of the causes of
impairments. Four of these pollutants amount to 49.5% of all causes and 92% of the causes that these
sources are most likely to contribute to: sediment (16%), nutrients (14.1%), metals (9.9%) and BOD
(9.5%). These four major categories of pollutants are the focus of the remainder of this analysis.  The
other four pollutants which are not further evaluated in this analysis (ammonia, organics, toxics and
inorganics) represent less than 5% of all of the causes of impairments nationally. These pollutant: are
38

-------
implicitly represented in the analysis to the extent that the cost of offsets for these pollutants can be
represented by the weighted average cost of the offsets for sediment, nutrients, oxygen and metals.

        Therefore, the distribution of offsets to be obtained by new or significantly expanded sources
that discharge into and contribute to the impairment of a water body is shown in Exhibit. Thus, for the
96 offsets required through 2015 (i.e., 6 per year for 16 years), we assume that 32.3% of the offsets will
be for sediment, 28.5% of the offsets will be for nutrients, 20% of the offsets will be for metals, and
19.2% of the offsets will be for BOD. Note however, that the occurrence of offsets for these pollutants
will be a little less frequent to the extent that some of the offsets will be for other pollutant groups (such
as toxic inorganics) instead of the four pollutant groups in Exhibit 1-12.

                                           Exhibit 1-12
                             Incidence of Pollutants Requiring Offsets
Pollutant
sediment
nutrients
metals
BOD
Total
% Occurrence
Nationally
16.0%
14.1%
9.9%
9.5%
49.5%
% of Offsets
32.3%
28.5%
20.0%
19.2%
100.0%
6. The Unit Cost of Offsets

        The steps described thus far estimate the number and type of offsets that might be needed. The
cost of offsets for a particular facility would depend on the particular mix of existing point and nonpoint
source dischargers that also contribute to the impairment and which might be willing to provide offsets.
The typical costs per pound associated with the removal of different pollutants by the existing
dischargers can provide a basis for providing perspective on the potential cost of offsets. The resulting
costs must be marked up to account for transactions cost to provide an estimate of the cost of the offsets.

        For this analysis, the annualized cost per pound of obtaining offsets for  total suspended solids
(TSS) and for biological oxygen demand (BOD) is based on the annualized cost of advanced treatment
for these pollutants. The annualized cost per pound of obtaining offsets for nutrients is based on recent
national experience and analyses for water trading in which agricultural BMPs provided the offsets.  The
annualized cost per pound for each pollutant category and its basis is further summarized in Exhibit 1-13
below, and is described in greater  detail in Attachment 4.
                                                                                               39

-------
                                          Exhibit 1-13
                   Cost per Pound of Offsets (March, 1999 Dollars) by Pollutant
Pollutant
TSS
nutrients
metals
BOD
Cost/lb.
$1.29
$25.50
$162.00
$1.21
Basis
The cost of upgrading to BCT, and its performance.
The cost of agricultural BMPs for nitrogen and phosphorous. For each pollutant,
applies the average cost for the range of costs that were developed. Also assumes
that nitrogen and phosphorous are equally likely to be the nutrient of concern.
The expected cost to a medium sized existing discharger to install multimedia
filtration for metals removal, and its performance.
The cost of upgrading to BCT, and its performance.
        It should be noted that these are considered to be generally representative costs, which are
appropriate for estimating national cost.  However, the cost of offsets in particular cases may be different
than those assumed above. This is especially likely for metals, in which the estimated cost per pound for
metal removal is for the removal of all metals in a waste stream.  This cost would be most representative
of the cases where a variety of metals are causing an impairment, and an offset must be provided for
these metals. However, in some cases the cause of an impairment might be identified as a particular
metal (such as zinc, lead, mercury, etc), and the required offset would be specific to that metal.  If that
metal represented, say, l/5th of the metals discharge for the source that is providing the offset, then the
cost per pound of providing the offset for that specific metal would be 5 times higher than $162 (i.e.,
about $1,000 per pound).  However, since the water body is specifically impaired due to a particular
pollutant, it may be reasonable to expect that there would be existing sources of discharge for that
pollutant that could provide more efficient opportunities for offsets. Nevertheless, the cost per pound
might be significantly higher for some new sources than estimated here. This influenced our approach
for estimating the metals discharge for new sources (discussed in the next subsection), which likely
overstates the overall magnitude of the metals discharge in a manner that would compensate for
understating the cost per pound of offsets in these  cases.

        It should also be noted that the costs shown in Exhibit 1-13 are not necessarily representative of
the costs associated with obtaining reductions of pollutants other than TSS, nutrients, metals and BOD.
For example, Attachment 6 provides an estimate for the cost of controlling toxic brganics that exceeds
$7,500 per pound. The total cost associated with such offsets is likely to be a relatively small fraction of
the total cost because the number of water bodies that are impaired for toxic organics represents a small
percentage of the impaired water bodies, and it is likely that the discharge that contains toxic organics   .
will  generally be a relatively low flow so that the size of the offset will be more limited in magnitude that
the other pollutants. Nevertheless, it may be difficult for a new source to find appropriate sources to
provide the offset, and in such a case the cost of the offset might be significant enough  to affect the
location or expansion decision of the firm.

7. Allowed Pollutant Discharges For Facilities Required To Obtain Offsets

        It is assumed for the purpose of this analysis that new or significantly expanded existing process
water sources that discharge pollutants of concern to impaired water bodies will generally be required to
achieve effluent standards equivalent to water quality criteria at end of pipe.  Ideally, we could apply
40

-------
 such standards to the estimated flows for new sources that need offsets to estimate the magnitude of the
 pollutant discharges that must be offset. Unfortunately, such criteria are often either not available or not
 readily applied. Our basis for estimating the pollutant concentrations for BOD, TSS, metals and nutrients
 is summarized below, and described in greater detail in Attachment 4.

        BOD & TSS

        There are no water quality criteria for these pollutants. We assume that the most any facility
 would be allowed to discharge is the BCT value for POTWs. EPA previously determined that 16 mg / L
 is the long term average performance of POTWs meeting secondary treatment requirements. (July 1986,
 51 FR 24974). This is likely to be an upper bound estimate since facilities located on impaired water
 bodies would be likely to have more stringent requirements.

        Nutrients

        The water quality criteria for phosphorous is 0.1 mg/L. There are no national criteria yet
 developed for nitrogen, and many states have only criteria for drinking water standards. We assume the
. allowed nitrogen discharge is 6 mg/L based on the total  nitrogen monthly average for new dischargers in
 the Tar-Pamlico watershed, which was designated as a nutrient sensitive water by the State of North
 Carolina, and which also has one of the most developed effluent trading programs in the country.

        Metals

        To be conservative, we assumed that the facilities requiring offsets would discharge every metal
 for which there was an available freshwater criteria. The resulting estimate of 0.426 mg/L represents a
 source that is discharging a wide variety of metals. The composition of the assumed discharge (based on
 the maximum allowed discharge at water quality criteria) is arsenic (35%), cadmium (0.5%), chromium
 III (17%), chromium VI (2.6%), copper (2.1%), lead (0.59%), mercury (0.18%), nickel (12%), selenium  .
 (1.2%), and zinc (28.1%). This total metals load is the load we assume must be offset.

        The total metals load that results from this discharge is likely to be an overestimate of the metals
 load that must be offset for two reasons. First, few facilities are likely to discharge all of these metals.
 Second, a single impaired water body is not likely to be impaired by all of the metals that comprise the
 allowed 0.426 mg/L discharge. If a water body is impaired for only one of these pollutants, then the
 magnitude of the total cost will be significantly overstated. For example, if the water body is impaired by
 zinc only then the relevant metals load would be overstated by a factor of 3.6; if the water body is
 impaired by copper only, then the metals load would be overstated by a factor of 47; and if the water
 body is impaired by mercury only, then the relevant metals load would be overstated by a factor of 555.

 8. Cost of Offsets for Various Facility Flows

        For a given facility flow, we can determine the  maximum allowed discharge for the four
 pollutant groups: TSS, nutrients, metals and BOD.  The needed offset would be 1.5 times greater than the
 load discharged, since the proposed regulation requires  a 1.5:1 offset. The cost of the offset is the unit
 cost for the pollutant being offset times the magnitude of the offset required. The resulting costs for
 facilities with flows ranging  from 0.025 MOD to 10.0 MOD are'shown in Exhibit 1-14 on the next page
 (not including transactions costs).  Thus, for example, for a  1 MOD facility that must provide offsets for
 metals and TSS, the cost of metals offsets would be $223,392 and the total cost of offsets for TSS would

                                                                                             41

-------
be $66,812, for a total cost of $290,204 annually.  Note that the costs scale linearly with flow, so that the
costs for a  15 MOD facility would be 15 times the costs for the 1.0 MOD facility.

        The total costs shown in Exhibit 1-14 illustrate the importance of the effluent limitations for the
new or significantly expanded existing sources in  determining the total cost of offsets. Even though the
unit cost of offsets for phosphorous ($17-$62/lb) are greater than the unit cost of offsets for nitrogen
($1 l-$20/lb), the total costs for obtaining offsets for phosphorous are 1/30 of the total costs for nitrogen
because the amount of phosphorous that is allowed to be discharged is l/60th of the amount of nitrogen
that is allowed to be discharged.

                                          Exhibit 1-14
                          Annual Offset Cost for Different Pollutants at
                     Various Facility Flows at  Water Quality Criteria Levels
                     (March,  1999 Dollars. Does not include transactions costs.)

Flow
(MGD)
0.025
0.050
0.500
1.000
2.000
3.000
4.000
5.000
10.000
Metals
(.426 mg / L)
$162/Ib
$5,585
$11,170
$111,696
$223,392
$446,784
$670,176
$893,567
$1,116,959
$2,233,918
BOD
(16 mg/L)
S1.21 /lb
$1,567
$3,133
$31,334
$62,668
$125,337
$188,005
$250,673
$313,342
$626,683
TSS
(16 mg/L)
SI. 29
SI ,670
$3,341
$33,406
566,812
$133,623
$200,435
$267,247
$334,058
$668,117
Nitrogen
(6 mg/ L)
Sll/lb
$5,342
$10,683
$106,830
$213,660
$427,320
$640,980
$854,640
$1,068,300
$2,136,600
S20/lb
$9,711
$19,422
$194,220
5388,440
$776,880
$1,165,320
$1,553,760
51,942,200
53,884,400
Phosphorous
(0.1 mg/L)
S17/lb
$138
$275
$2,751
55,503 .
$11,006
$16,509
$22,012
$27,515
$55,029
$62 /lb
$334
$669
$6,690
$13,380
$26,759
$40,139
$53,518
$66,898
S.133,796
Calculation of annual costs:  1 mg/L = 8.3 Ibs. /MGD. Multiply criteria times 8.3, and limes flaw in MGD for
total pounds/ day and times 260 for annual pounds. Multiply by 1.5 for the pounds of offsets needed under a 1.5:1
offset program. Multiply by cost/ Ibfor annual cost.
9. Total Private Sector Cost of Offsets

        Compliance Cost

        We have assumed that every year another four facilities will need to provide six offsets, and that
the total flow from these facilities ranges from 1.6 - 12.2 MGD. From Exhibit 1-12, we have estimated
the incidence of the pollutants that will need to be offset based on the frequency with which these
pollutants tend to cause impairments. We can use this information in combination with the facility costs
presented in Exhibit 1-14 to develop a weighted-average annual cost for obtaining offsets per MGD of
42

-------
flow as shown below in Exhibit 1-15. Therefore, the weighted average cost for a process water
discharger of obtaining an offset for discharging the pollutants of concern at 1MGD is $122,535.

                                           Exhibit 1-15
                  Weighted Average Annual Cost of an Offset per MGD of Flow
                     (March, 1999 Dollars.  Does not include transactions costs.)
Pollutant
sediment
nutrients*
metals
BOD
Cost/MGD
$66,812
$155,246
$223,392
$62,668
Total Weighted Average Annual Cost
% of Offsets
32.3%
28.5%
20.0%
19.2%
100.0%
Weighted
Annual Cost/MGD
$21,580 ,
$44,245
$44,678
$12,032
$122,535
                *Based on the average cost for phosphorous and for nutrients, and assuming
                that need for offsets for phosphorous and nutrients are equally likely.

        The flow range for the four facilities we assume will need offsets annually is 1.6 - 12.2 MGD.
Therefore, the resulting total annual cost for these four facilities to each provide a single offset is
$196,056 - $1,494,927. However, we assume that these four facilities will need to provide six offsets, so
that the cost associated with the six offsets is $294,084 - $2,242,391, before consideration of transactions
costs.

        As discussed earlier in this subsection, transactions costs have amounted to 35-45% of the cost of
obtaining offsets for new source review trading for air emissions — generally the higher percentage cost
applies to smaller value offsets and the lower percentage cost applies to higher value offsets. If similar
transactions costs apply for these offsets, then the total annual cost of offsets might range from about
$535,000 to $3,450,000.

        If each year, an additional 4 facilities are required to obtain 6 offsets at an annual cost of
$535,000 to $3,450,000 annually, then the annual cost would increase to $1,070,000 to $6,900,000 in the
second year, and so on. However, each year States would make progress in developing TMDLs.  If
TMDLs are developed uniformly for all of the current impaired water bodies from 2000 through 2015,
then the number of water bodies requiring offsets would be declining steadily  at a rate of about 6% per
year (assuming that there are few new water bodies listed during this period).  The result is that the
annual cost of offsets steadily increases for about 8 years to $2.4 - SI 5.5 million, but then declines until it
is eliminated in 2015.  Using the OMB-recommended discount rate of 7% annually, the present value of
these offsets would be $15.9- $102.2 million, and the annualized cost through 2015 would be $1.68 -
$10.83 million.

        Administrative Cost for the Private Sector

        We assume that the additional administrative burden associated with the offsets for both new and
substantially expanded existing dischargers can be represented by the effort associated with the burden
                                                                                               43

-------
for a permit modification plus the burden associated with a medium request for section 308 information.
Based on the November 6, 1998 ICRfor NPDES Modification and Variance Requests, the applicant
burden for requesting a modification is 5 hrs at $36.12 per hour; and based on the December 3, 1998ICR
for NPDES and Sewage Sludge Management Permits, the applicant burden for a medium request for
section 308 information is 50 hours at $36.12. Thus, each of the four facilities that must obtain offsets
would incur an additional burden of 55 hours for a total of 220 hours. We assume that the facilities that
provide the offsets must also provide similar information.  If, on average, two sources are needed to
provide each of the 6 offsets, then each of the 12 facilities would incur 55 hours of effort for a total of
660 hours. Thus the total administrative effort would be 880 hours for a total cost of $31,786 annually.

        Unlike the compliance cost, this annual cost will not accumulate from year to year, although the
permits will need to be renewed every five years (and we assume these costs would be incurred again
upon renewal). As in the case of the compliance cost, TMDLs will be developed uniformly through
2015, reducing the number of facilities subject to offset requirements. Taking all of these factors into
account, results in a present value of $275,300, and an annualized cost through 2015 of $29,143.

        Total Cost

        The total  compliance and administrative cost associated with the proposed offset requirement is
estimated to range from $ 1.71 to $ 10.86 million annually.

10. Cost to Permit Authorities

        The corresponding cost to permitting authorities for processing offsets is 40 hours for each
permit modification and 10 hours to review the section 308 information for a total burden of 50 hours at
an average cost of $31.68 per hour. The total burden for 16 facilities would be 800 hours for a total
annual cost of $25,344.  Taking into account a 5-year renewal cycle and the development of TMDLs
from 2000 through 2015 results in a present value of $219,508, with an annualized cost of $23,237.

11. Summary

        The annualized costs associated with the proposed offset requirements for industrial process
water discharges is as follows:
Type of Cost
Cost To Private Sector
Cost To Permit Authorities
Total Cost
Annualized Cost
(S Millions)
1.71 - 10.86
.023
1.73- 10.88
44

-------
                  H.  PROPOSED SILVICULTURAL AUTHORITIES

PROPOSED PROVISIONS ALLOWING SOME SILVICULTURAL ACTIVITIES TO BE DESIGNATED AS
POINT SOURCES SUBJECT TO NPDES

        Currently, most silvicultural activities are not point sources subject to NPDES requirements.

        Silvicultural point source means any discernible, confined and discrete conveyance
        related to rock crushing, gravel washing, log sorting, or log storage facilities which are
        operated in connection with silvicultural activities and from which pollutants are
        discharged into waters of the United States. The term does not include non-point source
        silvicultural activities such as nursery operations, site preparation, reforestation and
        subsequent cultural treatment, thinning, prescribed burning, pest and fire control,
        harvesting operations, surface drainage, or road construction and maintenance from
        which there is natural runoff..  (40 CFR § 122.27(b)(l))

The proposed provisions delete the second sentence quoted above, thereby eliminating the definition of
most silvicultural activities as non-point sources that are not subject to NPDES. The result of removing
this exemption is that discharges from aH silvicultural activities can be designated as point sources
subject to NPDES under existing authority to designate sources of storm water:

        ... discharges composed entirely of storm water shall not be required to obtain a NPDES
        permit except:... (v) A discharge which the Director, or in States with approved NPDES
        programs, either the Director or the EPA Regional Administrator,  determines to
        contribute to a violation of a water quality standard or is a significant contributor of
        pollutants to the waters of the United States. (40 CFR § 122.26(a)(l))

This existing designation authority is available to both EPA and the States. However, the proposed new
provision adds that "EPA shall only designate silvicultural point sources discharging to waters for which
EPA establishes a TMDL." Thus, under the proposed provisions, EPA could designate silvicultural
activities only in connection with establishing a TMDL. States, however, could designate any
silvicultural activities that contribute to violation of WQS or significantly contribute pollutants,
independent of whether or not a TMDL is involved.

OVERVIEW OF COST IMPLICATIONS

        Neither States nor EPA currently have authority under Federal NPDES regulations to address
silviculture/1  The removal of the exemption  and subsequent permitting of silvicultural operations may
result in compliance costs for these sources beyond what these sources would incur in the absence of the
proposed regulations.                                                                      '

        These costs, however, are indirectly rather than directly attributable to the proposed silviculture
        31 Apart from the limited circumstances involving rock crushing, gravel washing, etc. Existing authority
over these few silvicultural activities address very little of the potential threat to water quality from silviculture, and
we will not concern ourselves in this paper with this limited existing authority.

                                                                                             45

-------
designation provisions.  The provisions do not require that any silvicultural operations be designated, nor
do they directly require any silvicultural operations to implement any new practices.  The provisions only
authorize EPA and the States to designate some silvicultural operations as point sources under certain
circumstances. The circumstances may not arise, an even if they do arise, EPA and the States may
choose not to exercise the designation authority. Costs will be incurred only if and when the authority is
invoked. This paper presents some rather uncertain estimates of how often the proposed designation
authority might be invoked, and, if so, the costs that will likely ensue.

SUMMARY OF TYPES OF COSTS AND PROCEDURES FOR ESTIMATING THEM

        In this paper, we will estimate the following costs attributable to the proposed silviculture
provisions:

•       Compliance costs for the silvicultural operations that will be permitted and required to change
        their practices in various ways that will protect water quality. These compliance costs will
        include: 1) The additional capital and operating costs of the BMPs these operations will be
        required to implement; and 2) The administrative costs of obtaining and maintaining coverage
        under NPDES permits.

•       Administrative costs to Federal and State governments of permitting these operations.

In addition to developing national cost estimates for all entities, we will also provide an estimate of the
regulatory costs and impacts specifically for small entity silvicultural operations.

        We will estimate these costs through a sequence of eight steps:

1.       Determine the volume of silvicultural activity that occurs, who performs it, and where it takes
        place. This will provide the data base from which to begin the analysis.

2.       Determine the geographic areas/types of silvicultural operations for which there are existing,
        enforceable non-EPA legal authorities that already can be used to compel silvicultural operations
        to act  in the desired manner.  We assume that existing authorities will be used preferentially over
        the proposed designation authority in these areas and for these types of operations.  The proposed
        regulation will thus impose no incremental costs for these areas/operations.

3.       Determine the volume, location and number of silvicultural operations that will be affected by
        the provision. This will involve matching the areas where silviculture occurs (#1) less the areas
        where existing authorities will be used (#2) against 303(d) waters that are impaired by
        silviculture.

4.       Determine the set of silvicultural BMPs that will typically be required of designated and
        permitted silvicultural operations and the per unit costs for these BMPs.

5.       Estimate total national BMP costs by multiplying the  volume of silviculture affected (#3) by unit
        BMP costs (#4).

6.       Estimate the administrative costs for silvicultural operations to obtain and maintain coverage


46

-------
        under NPDES permits. This will be estimated by multiplying the number of operations that will
        be designated (#3) by an estimate of the administrative cost for a typical operation.

7.      Estimate the administrative costs to Federal and State governments for designating and
        permitting these operations.  Assuming that permitting will be accomplished by general rather
        than individual permits, these costs will be estimated by multiplying the number of general
        permits to be developed (assuming one general permit for each silvicultural area where
        designation will occur — #3) by an estimate of the administrative cost for developing and
        implementing a typical general permit

8.      Develop estimates of compliance costs and administrative costs for small entities specifically.
        This will be done by disaggregating the estimated costs for all entities developed under #5 and
        #6.

Most of this paper will consist of describing the data, calculations,  results and uncertainties in each of
these eight steps.

KEY ASSUMPTIONS IN ESTIMATING COSTS

        Estimating the costs of the proposed provisions is particularly difficult because the provisions
authorize EPA  and State program Directors to take action, but do not mandate specific actions. No duties
are established for EPA, States, or silvicultural operations. There is no certain answer to the fundamental
questions of what will EPA and the States do  and what will they require of silvicultural operations if the
proposed regulations are finalized.

        As the  starting point for estimating the costs of the proposed provisions, we assume that the
designation authority will be implemented as  follows:

•       The designation authority will be invoked bv States or EPA only for silvicultural operations that
        affect a 303(d) water that is impaired  or threatened by silviculture. Even though the proposal
        gives the State Director designation authority that extends  beyond 303(d) waters (e.g., wherever
        silviculture "...is a significant contributor of pollutants to waters of the United States"), we
        assume that the designation authority will not be used so broadly.  The provision is intended to
        facilitate attaining WQS in water bodies where silviculture is an important contributor to
        nonattainment and a TMDL will be needed, and we will assume that it will not be used beyond
        these instances.

•       The designation authority will not be  invoked when there is some other existing, enforceable.
        less onerous authority that can be used to elicit the desired controls from silvicultural operations.
        Designating many silvicultural operations as point sources and then permitting them will be very
        burdensome, both for the permit authority and for the regulated sources.  We  assume that this
        authority will be used only as a last resort — it will be used only when there is no easier way of
        achieving the desired behavior from silvicultural operations.  In fact, many States have laws and
        programs (pursuant to State law, not Federal) that give them adequate enforceable authority over
        silvicultural activities that may affect water quality.  Also, all private timber cutting on National
        Forest  lands is pursuant to timber sales contracts, and the U.S. Forest Service (USFS) has
        authority to require compliance with  water quality-related  BMPs as part of these contracts.


                                                                                              47

-------
        Indeed, USFS is obliged under its land and resource management planning regulations to require
        management practices that protect water quality. The owners of other public lands (e.g., BLM,
        States, counties) also typically have authority to require use of appropriate BMPs of the entities
        that cut timber on public lands. Finally, many States have voluntary or incentive programs that
        encourage good silvicultural practices on private as well as public lands.  We will describe the
        existing State, USFS, and other authorities over silviculture in more detail in a subsequent
        subsection of this paper. Our assumption is that where enforceable authorities or workable
        voluntary programs already exist, they will be used preferentially over the proposed designation
        authority. The effect of this assumption is that we assume no use of the proposed designation
        authority (and no compliance costs attributable to the proposed regulation) for: 1) States that
        have enforceable forest practices laws; and 2) Timber harvesting from all public lands, including
        National Forest lands.32

        In cases where there is no existing comprehensive enforceable authority over silviculture, we
        assume that the designation authority may be used. We believe that virtually any existing
        authority — whether it is only partially enforceable, it applies to some silvicultural operations and
        not others, or it is a voluntary or incentive program — will be used preferentially over
        designation. Any of these existing authorities could be used effectively in a site-specific manner
        to establish reasonable assurance for a TMDL. Designation will be used in whatever instances
        remain.  We will estimate a lower and an upper bound for the frequency with which we expect
        that the designation authority will be used when no comprehensive enforceable authority is
        available. This estimate for the frequency of designation is highly uncertain. There is little
        experience that is directly relevant to judging how often the designation authority might be
        used.33

        Permitting will be achieved through a general permit covering all operations in the area, not
        individual permits for individual silvicultural operations. This assumption is a matter of
        practicality. It is likely that all the silvicultural operations in the geographic area affecting the
        impaired water body will need to implement roughly similar BMPs. The sources and desired
        controls will probably be sufficiently similar to warrant treatment via a general permit.  A
        general permit approach will be far less costly for the control authority than individual permits.

        The general permit will require all the silviculture operations in the specified geographic area to
        employ appropriate BMPs. The BMPs that are required can be expected to be tailored somewhat
        according to region of the country, topography, soils, nature of the water body impairment, etc.,
        but there will likely be general similarity across the country in the silvicultural practices that are
        32 Another rationale that gets us to the same conclusion regarding the costs of the proposed silviculture
provisions is to define existing enforceable State, USFS and other silviculture authorities to be part of the pre-
regulatory baseline. If existing State, USFS, Bureau of Land Management (BLM), and other authorities are
assumed in the baseline to be employed whenever necessary to effectuate TMDLs, then the proposed silviculture
provisions can potentially impose incremental costs only for locations where these existing authorities are not
already available.

        33 Experience with the long-existing NPDES designation authority for concentrated animal feeding
operations (CAFOs) supports an expectation that the silviculture designation authority will be little used if there are
any reasonable alternatives. Only a handful of CAFOs have been designated over the past fifteen years.

48

-------
        addressed by the BMPs: road construction and maintenance, streamside buffers, log skidding,
        revegetation, etc. We believe that there will be enough consistency in the silvicultural BMPs that
        will be required in different areas to allow us to pursue a unit costing approach (e.g., for all
        silvicultural operations of type "X", the BMPs that will be required will cost roughly $Y per
        acre).

ESTIMATED COSTS OF THE PROPOSED SILVICULTURE PROVISIONS

        In this section, we describe the eight steps resulting in estimates for the different varieties of
costs attributable to the proposed regulations.

1. Determine how much silviculture occurs, and  where

        Silviculture occurs in all 5D States. The most detailed information available on how much timber
harvesting occurs and where is provided by the Timber Product Output Database, generated by the USFS.
USFS has developed statistics on the volume of timber of various sorts cut in each county in the country.
The data is broken out by ownership of the land from which the timber is cut (private/forest industry,
private/non-forest industry, National Forest, other public). The most recent data available is for 1996,
and we have used this year's worth of data to represent the volume and location of timber expected to be
harvested in future years.34 This data base shows total 1996 timber cut as follows:

                                          Exhibit n - 1
                               National Timber Cut in 1996 (mcf)
From private lands
From National Forests
From other public lands
Total
14,330,530
912,060
1,025,628
16,268,218
        We have been unable to find any geographically disaggregated data base on timber revenues or
on acres undergoing harvesting. In order to convert the USFS timber harvest data into revenue and
acreage information, The Timber Data Company developed for us regional estimates of: 1) The current
price received per thousand cubic feet (mcf) for harvested timber of various sorts;35 and 2) The number of
        34 Reliance on this single year's worth of data is a source of some uncertainty in our estimates.  Timber,
like other commodities, is subject to significant price volatility.  The amount of timber harvested in any given year
is somewhat less variable than is price, but nevertheless has increased and decreased substantially in the past few
years, and is likely to continue to change in the future.

        3> TDC estimated prices using direct and inferred information from its Stumpage Price Report (containing
U.S. timber and log prices by region) and the Random Lengths Yardstick reprinting of Timber-Mart South log
prices.
                                                                                              49

-------
acres on which harvesting operations would occur in order to obtain 1 mcf of timber.36 These data are as
follows:

                                           Exhibit H - 2
                    Timber volume per acre and log prices (per mcf), by region
Region
Westside Oregon &
Washington
California
Arizona & New Mexico
Inland & Rocky Mountain
Lake States
Midwest
Northeast
Westside Southeast
Eastside Southeast
mcf/Acre
3.0
2.0
1.5
1.5
1.3
0.6
0.6
1.0
1.0
Sawtimber
Softwoods
$2,500
$2,500
$2,000
$1,850
$1,650
$1,350
$1,350
$2,400
$2,400
Sawtimber
Hardwoods
$2,500
$2,000
$2,000
n/a
$1,750
$1,500
51,500
$1,600
$1.600
Softwood
Pulpwood
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1.000
Hardwood
Pulpwood
$1,000
$1,000
$900
$900
$900
$800
$800
$800
$800
We need acreage data for our analysis in order eventually to scale up the costs of silvicultural BMPs,
which are given in terms of dollars per acre undergoing harvesting.  We need price data for our analysis
in order to calculate revenues, against which compliance costs will be compared.

2. Determine the areas and types of silvicultural operations for which enforceable authorities
already exist

        We assume that existing enforceable authorities will be used preferentially over the proposed
designation authority in circumstances where the existing authorities are applicable. Three sorts of
existing authorities are of interest: enforceable State laws that can be used to control forestry practices on
private and State lands, authorities available to Federal land management agencies to require silvicultural
BMPs for timber harvesting on Federal lands, and the ability of other public sector landowners (e.g.,
counties, Tribes) to prescribe conditions for silvicultural activities on their lands.

        Many States have existing, enforceable forest practices laws that are used to regulate silviculture
activities on private  lands.37  These acts typically require planning for timber harvest, review of submitted
        36 TDC estimated the acres harvested in order to produce a given log volume based on queries of Timber
Data Company's database of USFS and selected State timber sale appraisal data for 1998 sales. TDC adjusted this
data somewhat based on discussions with industry professionals regarding differences between public and private
timberland management practices.  Adjustments also were made for known regional differences in land growth
potential and harvest methods.

        37 The following information on State authorities is drawn from two studies by the Environmental Law
Institute: Enforceable State Mechanisms for the Control ofNonpoint Source Water Pollution (October,  1997) and
50

-------
plans, use of prescribed forest practice standards or BMPs, monitoring, and enforcement. Example
States that have and implement such laws are California, Nevada, Massachusetts, Connecticut, Maine,
Alaska, Idaho, Oregon and Washington.  Rather than impose statewide or watershed-based enforceable
standards for forest practices, other States have forestry "bad actor" laws that allow the State to direct
forest operators who are contributing to pollution problems to implement specific improved practices.
Such States include Delaware, Virginia, West Virginia and New Hampshire.  Other States have more
general laws requiring enforceable erosion or sediment controls that are applicable to forestry as well as
other activities. Such States include Ohio, Vermont, Maryland, Nebraska, Kentucky, and North Carolina.
Wisconsin has even more general authority to order abatement of nonpoint source pollution in priority
watersheds (as we presume 303(d) waters would be) and require use of BMPs and corrective action. The
Exhibit below lists the States that we determined to have comprehensive enforceable authorities that can
be used to require silvicuftural BMPs on all private and non-Federal public land in the State.

                                          Exhibit H - 3
              Does  the State Have Adequate, Comprehensive, Enforceable Authority
        to Require  Use of Silvicultural BMPs on All Private and non-Federal Public Land?
Yes
AK, CA, CT, DE, ID, KY, ME, MD, MA, NE,
NV, NH, NC, OH, OR, VT, VA, WA, WV, WI

No
AL, AZ, AR, CO, FL, GA, HI, IL, IN, IA, KS,
LA, MI, MN, MS, MO, MT, NJ, NM, NY, ND,
OK, PA, RI, SC, SD, TN, TX, UT, WY
       Twenty States are considered to have applicable comprehensive enforceable authorities, thirty-do
not.  We believe our categorization to have been conservative, in the sense that we did not count a State
as having a comprehensive enforceable authority if there was any question whether the State's authority
was enforceable and applicable to all silviculture on private and non-Federal public land that might affect
impaired waters.  For example, Montana requires BMPs for timbering within streamside management
zones, but BMPs are voluntary beyond these zones. Michigan law provides for BMPs and enforceable
standards, but only in forested regions that have been designated "forest improvement districts", and such
designation requires the consent of a percentage of the affected landowners. Alabama's statute gives
broad authority to the Forest Commission "to adopt and promulgate rules and regulations pertaining to all
phases of forestry within this state, which rules and regulations when adopted shall have the force and
effect of law", but the commission has not yet adopted enforceable standards.. Many States require use of
BMPs for forest activities on State-owned lands, but not on private lands. For all of these and similar
States where there are not universally applicable enforceable authorities, we assume that the proposed
designation authority will be used to some degree.

       Many States have adopted and implemented their forest practices authorities in recent years.
Exhibit II-4 shows how recently the 20 States with comprehensive enforceable authorities adopted their
key provisions.
Almanac of Enforceable State Laws to Control Nonpoint Source Water Pollution (1998).
                                                                                              51

-------
                                         Exhibit D - 4
    When Did States With Comprehensive Enforceable Silviculture Authorities Adopt Them?
Before 1990
CT, ME, MD, NV, NH, NC,
OR,WI
1990 to Present
CA, ID, KY, OH, VT, WA, WV

Unknown
AK,DE,MA,NE,VA

Eight States adopted their key authorities before 1990, seven did so since then, and we were unable to
determine (from an Internet search of State laws) when five States did so. Excluding the States for which
we could not obtain the desired information, approximately half of the 20 States now having enforceable
authority to require BMPs for silviculture on private and non-Federal public lands adopted these
authorities since 1990.  We assume that this rate of progress in States adopting silviculture authorities
will continue into the future. In another nine years (by 2008) an additional 10 or so States that do not
now have enforceable silviculture authorities will have adopted them. By 2015, when all the TMDLs for
waters listed in 2000 will be required to be completed, an additional set of States will likely have adopted
enforceable authorities.- In effect, the universe of States without enforceable authorities and in which the
designation procedure might likely be used will shrink over time. The extent to which the designation
authority will be used will also shrink.  States are expected to develop their TMDLs at an even pace over
the 2000 - 2015 period.  Some TMDLs will be developed soon when 30 States are without enforceable
authorities and the need for the designation procedure is the greatest. Some TMDLs will not be
developed until 2015 when perhaps only 10 States will be without enforceable authorities, and the need
for the designation procedure will be much reduced. For this analysis, we will assume conditions as they
will be in 2008, at the midpoint in the period over which the TMDLs will be developed. We expect at
that time that an additional 10 or  so States will have adopted enforceable authorities,  bringing the total to
30. 20 States will remain without enforceable authorities, and these 20 States will constitute the realm in
which the proposed designation authority will potentially be used.

        The 20 States that we consider to have existing, enforceable authority to require silvicultural
BMPs accounted for 32.9 % of the national timber harvest in 1996.  By 2008, the median year in which
TMDLs will be completed, we assume that another 10 unspecified States will have adopted authorities.
One-third of the 30 States that do not now have enforceable authorities will adopt them by 2008, leaving
by then only 20 States that do not have enforceable authorities and in which the proposed silviculture
designation procedure may be used.  The fraction of the nation's timber production accounted for by
States with enforceable authorities in 2008 will expand to an expected 55.3 % (32.9 % plus 1/3 x 67.1
%). By assuming that the proposed designation provision will not be used in States that now have
enforceable authorities or will have by them by 2008, we are assuming that at least 44.7 % of the nation's
silvicultural activity will be unaffected by the proposed regulation.

        The USFS also has enforceable authority to require silvicultural BMPs - on National Forest
lands as necessary to protect water quality. The National Forest Organic Act, the Multiple Use-Sustained
Yield Act, and the National Forest Management Act have become progressively more prescriptive in
requiring the USFS to manage for multiple uses, including conservation of soil and water resources. The
USFS approach to water quality protection is as follows. Each USFS Region has established a minimum
set of soil and water conservation practices that must be required throughout all National Forests in the
Region. Each individual National Forest must develop a Forest Management Plan, incorporating at a
52

-------
minimum both these practices and any applicable State BMP requirements.38 Each Management Plan
goes through the NEPA process, in which the need for additional water quality-related requirements is
considered. The Management Plan defines management areas for potential timber harvesting within a
National Forest. For each management area in which timber harvesting is to occur, a project plan must
next be developed.  The project plan undergoes NEPA review also, and furthermanagement area-specific
BMPs are developed and specified in the NEPA document.  These BMPs, in turn, must be incorporated
by USFS into any subsequent contract for a timber sale. Contract administrators then ensure that the
BMPs are applied and ultimately determine whether the timber operator's performance has been
consistent with the requirements.

       It seems highly likely that any future potential adverse impacts from silviculture on National
Forest lands will continue to be addressed through this long-standing planning and management process
rather than through the proposed new  designation authority. In fact, USFS'  forest planning regulations
(36 CFR § 219) are now being revised to provide for a more targeted, watershed-based approach to water
quality protection, which, if finalized, will make use of the proposed designation authority even less
likely.

       Other Federal land managing  agencies that allow timber harvesting (the Bureau of Land
Management, the Fish and Wildlife Service) have similar enforceable authorities to require silvicultural
BMPs. We have not investigated these authorities in detail because, except  in Oregon (which effectively
requires BMPs throughout the State),  the amount of timber cut on these agencies' lands is minimal.

       National Forest system lands, for which we believe that there is existing, enforceable authority to
require silvicultural BMPs, accounted for 5.6 % of the nation's timber harvest in 1996. We expect that
there will be no need for use of the proposed designation authority with respect to this 5.6 % of the
nation's silvicultural activity.

       Like Federal land managing agencies, other public sector landowners (States, counties,
municipalities, Tribes) also typically have enforceable authority to manage timbering on their lands by
writing desired management measures into their timber sales contracts with the private entities that do the
cutting.39

       When we combine the three sets of projected enforceable authorities to require silvicultural
BMPs in the baseline - those for 30 States in 2008; for the USFS and other Federal land managing
agencies; and for other public landowners — we are assuming that there will be no need for use of the
proposed designation authority with respect to 62.4 % of the nation's timber harvest.
        38 Section 313 of the Clean Water Act (Federal facilities pollution control) requires USFS to comply with
all Federal, State and local water pollution control requirements. This includes requiring USFS contractors to meet
such requirements.

        j9 Source: discussion with James McElflsh, Environmental Law Institute, and primary author of the ELI
studies on enforceable mechanisms for control of nonpoint source pollution.

-------
3. Determine the location, volume, and number of silvicultural operations that will be affected by
the provision

        We employed four steps in determining the amount of silviculture activity that will be affected
by the provision:

        a..     Identify the areas where silviculture occurs and is contributing to the impairment of a
               303(d) water body;

        b.      Eliminate from further analysis the silvicultural activity occurring where we believe
               there is very little likelihood that the proposed provision will be applied. This includes
               all silviculture: a) In the 20 identified States with existing comprehensive enforceable
               authorities; b) On Federal and other public lands; and c) That contributes to impairment
               to only a minor degree;

        c.      In the remaining areas where the proposed designation authority may be used, develop
               low and high estimates of how often it will be used; and

        d.      Scale these low and high usage estimates statistically in order to account for: a) States
               that have not reported information allowing a determination of how often silviculture
               contributes to impairment of their 303(d) water bodies; and b) The assumption that by
               2008, an additional 10 unidentified States will have adopted enforceable authorities,
               thereby further reducing the areas where the proposed designation authority may be used.

After we complete these steps and estimate the volume of silvicultural activity to which the designation
authority will be applied, we will estimate the compliance costs that will be incurred to apply appropriate
BMPs for this volume of timber harvest.  In this subsection, we discuss the four steps we took to
determine the location and volume of silviculture activities that we estimate will be affected by the
designation provision.

        3a. Identify the areas where silviculture occurs and is contributing to the impairment of a 303(d)
        water body

        We began with a search of the 1996 and 1998 data bases of 303(d) waters for water bodies with
silviculture cited as the cause of impairment; The search was performed by TetraTech,-Inc. Such waters
were reported for only 10 States.  We judged that coverage in these data bases was substantially
incomplete — some States are not yet included in the data base at all, and for many States the source
information for listed waters is incomplete.  The 1996 National Water Quality Inventory (compiling the
results from the States' 305(b) reports) appears to provide a much more complete picture of the extent of
water quality impairment by silviculture, indicating impairment in at least 33 States.  We therefore
decided to use the National Water Quality Inventory as our point of departure.40
        40 Use of 305(b) rather than 303(d) likely causes us to overestimate costs for this proposed provision.
States tend to list more water bodies as impaired in their 305(b) reports than they do for 303(d). Listing a water
under 303(d) typically entails important consequences (i.e., the need to develop a TMDL and implement potentially
costly load reductions in order to achieve water quality standards), while listing a water for 305(b) carries much
lesser implications.  Most States apply a much more stringent burden of proof in listing a water for 303(d) than they

54

-------
        The 1996 National Water Quality Inventory reports that States cited silviculture as a major or
 moderate/minor contributor to water quality impairment in:

 •       2% of surveyed river and stream miles in the U.S. (6% of impaired river and stream miles);

 •       2% of surveyed lake acres (5% of impaired lake acres); and

 •       1% of surveyed estuary square miles (3% of impaired estuary square miles).

 In order to identify and locate the specific water bodies that were cited in the underlying State 305(b)
 reports as being impaired by silviculture, we consulted EPA's 1996 National Assessment Database. The
 database was created to provide a uniform and consistent method for States to report water quality
 information in their 305(b) reports.  We performed a search of the database to find all impaired water
 bodies across the U.S. for which silviculture was identified as a major, moderate, or minor source of
' impairment. 965 such waterbodies were identified in  25 States.

        We do not believe this list of water bodies impaired by silviculture is a complete representation
 of all such water bodies throughout the nation.  The data base shows no water bodies impaired by
 silviculture in 25 States.  While it is likely true for several of these States that there really is no
 silvicultural impairment,  most of the States with no impairments shown in the data base simply either: 1)
 Did not report the cause of impairment for any of their impaired water bodies; or 2) Used a set of codes
 for reporting the source of impairment that did not include silviculture as a possibility.  We thus believe
 that the degree of silvicultural impairment reported in the 25 States citing silviculture should be viewed
 as a sample, and the results from these 25 States should be appropriately extrapolated or scaled up to
 estimate the degree  of impairment in the Nation as a whole. We will perform this scaling at the end of
 this subsection.

        Before attempting to  locate each of the 965 impaired waterbodies and determine the amount of
 silviculture affecting them, we first performed the review of enforceable State authorities relating to
 forestry that was described in the previous subsection. This review identified 20 States with existing
 comprehensive enforceable authorities, and we assumed that the proposed designation provision will not
 be used in these States. Based on this assessment, 23 8 waterbodies within these 20 States were
 eliminated from further consideration.

        After this culling, 727 silviculture-impaired waterbodies in 18 States remained from the original
 database search. For each of these water bodies, the amount of silviculture activity (measured in volume
 of timber harvest, acres, or otherwise) that contributes to that water body's impairment is presumably the
 amount of silviculture activity that will be regulated if the designation authority is invoked. The finest .
 level of geographic  resolution at which silvicultural activity data (timber harvest volume, see step # 1) is
 reported is the county level.  Similarly, the finest level of geographic resolution at which we were able to
 locate identify individual water bodies in the time available was also the county level.41  With more time,
 do in listing a water for 305(b).

        41 The data base printout we received identified both the "HUC" (watershed code) within which the
 silviculture-impaired water body is located and the counties covered by the HUC. HUCs often span 10 or more
 counties.  Thus, the data base printout allowed us to locate an impaired water body only to the extent of knowing
                                                                                               55

-------
we might be able to provide a more precise location for each water body, and to identify the basin in
which silviculture-related loadings contribute to impairment of the water body.  However, the difficulty
of obtaining timber harvest volume data on a sub-county level for all affected water bodies may will
likely still present an obstacle to pinpointing the exact amount of silviculture activity that contributes to
impairment. For this analysis, the best we could attempt to do is identify a single county within which
the silviculture-impaired water body is located, and assume that al] the silvicultural operations within that
county might be designated and then permitted in order to eliminate the impairment.  Our assumption  is
that the designation authority may be used for up to all of the silvicultural operations occurring within a
county in which there is a silviculture-impaired water body.

        We identified the county in which each silviculture-impaired water body is located by using 3
sources of information.  First, for each waterbody in the database, there is an associated 8-digit "HUC",
or watershed code. Using other information provided by EPA and the Research Triangle Institute, we
identified the counties covered by each watershed, thereby narrowing the range of possible counties in
which each water body might be.42  Second, we compared the set of possible counties for each water
body with results from searching the U.S. Geological Survey's National Atlas of the U.S.
(http://www-atlas.usgs.gov/scripts/start.html).  If these steps did not result in identifying a single county
within which a particular water body is located, we took the third step of consulting State maps.  In cases
where these three steps did not narrow down the choice of possible counties to one, we included all
possible counties for that waterbody in our initial list of counties to consider.

        Out of the 727 silviculture-impaired water bodies, we identified the individual county associated
with each of 608 water bodies. For an additional 76 water bodies, we were able to narrow the list of
possible counties to two. For the remaining 43 water bodies, we could not narrow the list of possible
counties to less than three.  In total, we identified 230 counties as being associated with the 727 water
bodies.

                                           Exhibit H - 5
      Number of Reported Silviculture -Impaired Water Bodies and Corresponding Counties

Water bodies in one county
Water bodies in two counties
Water bodies associated with 3 or more counties
Total
Water
Bodies
608
76
43
727
Number of
Corresponding
Counties
• 158
38 more
34 more
230
that it is somewhere within a group of counties. EPA is currently developing a database that will map the latitude
and longitude of impaired water bodies, allowing the user to identify their location with much more precision,

       42 William Cooter of RTI provided us with a copy of the database and performed the initial queries to
identify silviculture-impaired water bodies, their watersheds, and the counties included in these watersheds.
56

-------
        Cases in which silviculture activities in three or more counties contribute to the impairment of a
particular segment of a water body are probably ^extremely rare.  Most silviculture-impaired water bodies
are small, and they are likely affected by the logging that takes place within only a portion of the
surrounding county — they are unlikely to be affected by logging in the surrounding 3 counties, much less
the surrounding (in some cases) 10 counties. We believe that inclusion of all 230 counties in our final
list would lead to an overestimation of the amount of silvicultural activity that might be subject to the
designation authority. We therefore decided to consider only those counties within which were located
silviculture-impaired water bodies that could be tied to two or fewer counties. Although this approach is
inexact, our inclusion of counties from water bodies for which we identified two possible counties may
roughly offset our excluding counties from water bodies for which we identified three or more possible
counties.43  Using this two-or-fewer-counties-per-water-body approach, }ye found a total of 196 counties
associated with silviculture-impaired water bodies.

        3b.  Eliminate from further analysis the silvicultural activity occurring where we believe there is
        very little likelihood that the proposed provision will be applied.

        For each water body listed as impaired by silviculture, the 1996 National Assessment Database
contains a data field indicating the source magnitude — the extent to which silviculture contributed to the
impairment of the water body. Thus, for each affected water body, silviculture may be listed as a
"major", "moderate", or "minor" source of impairment.  Water bodies for which silviculture was listed as
a minor source of impairment are impacted more significantly by one or several other sources of
impairment. For these water bodies, addressing the silviculture contribution to impairment through BMP
requirements would not fully eliminate the impairment, and would likely be of much lower priority than
addressing the major or moderate sources of impairment. In most cases, we would not expect the TMDL
for such a water body to require extensive control of silviculture. Accordingly, we will assume that those
counties that contain one or more silviculture-impaired water bodies  listed as "minor", and that do not
contain any silviculture-impaired water bodies listed as "moderate" or "major", will not be affected by
the designation provision.  29 of the 196 counties in our list met these criteria, and were therefore
excluded from the list Our final list of counties associated with major or moderately silviculture-
impaired water bodies in States without existing enforceable authorities contains 167 counties. This
represents 5 % of the total number of counties in the U.S.

        We determined the amount of silvicultural activity in these 167 counties, located  in 18 states,
using timber harvest volume data from the U.S. Forest Service's Timber Product Output Data Retrieval
        43 34 additional counties (not already identified from the 1- or 2- county water bodies) were derived from
the group of 43 water bodies that were each associated with 3 or more possible counties. Some of these "3-or-more"
counties probably do include one or more of the 43 water bodies.  In this respect, excluding all the "3-or-more"
counties undoubtedly results in an undercount of counties. On the other hand, we would expect that our inclusion of
"2-county" counties in our base list results in an overcount of counties. While it is possible that silviculture
activities from adjoining counties may impact a particular water body if it lies near their shared border or if the
impaired water body crosses a county line, thereby implicating silvicultural activities in two counties, we believe
this to be the exception rather than the rule. Therefore, for a few of the "2-county" water bodies, we should be
considering silviculture  activities from both counties, while for the majority of these "2-county" water bodies, we
are probably including counties that are not actually affecting that water body. The effect of excluding timber
harvest volume from those "3-or-more" counties that do not already appear on our base list, and the effect of
including timber harvest volume from those "2-county" counties that probably do not impact the listed water body,
will tend to offset one another.

                                                                                                 57

-------
System (http://srsfia.usfs.msstate.edu/rpa/tpo/), developed in support of the 1997 Resources Planning Act
Assessment.  Timber harvest volume for 1996 is available by type of product at the regional, sub-
regional, state and county level, and can be broken out into volumes harvested on National Forest land,
other public land, forest industry land, and other private land41.  Using this harvest data, we calculated the
total volume of timber harvested in the 167 counties (1,828,936 thousand cubic feet (MCF)).  This
volume represents 11.2 % of the total U.S. timber harvest on all lands in 1996.

       Part of the total volume of timber harvested  in the 167 counties was taken from National Forest
system lands. This amount is 131,544 MCF, about 7 % of the total harvest in these counties.  We assume
that this timber harvesting will be regulated preferentially under existing USFS authorities rather than the
proposed designation authority. We exclude this amount from the timber volume that might potentially
be covered by the designation authority.

       Another part of the total volume of timber harvested in the 167 counties was taken from other
public lands. This amount is 94,608 MCF, about 6% of the total harvest in these counties. We assume
that timber harvested from these other public lands also will not be affected by the provision. Although
the statutory and regulatory authorities governing silviculture practices on other public lands may not be
as definitive as those for National Forest lands, other public land owners (Bureau of Land Management,
State, county and municipal governments, Indian tribes)  generally have the ability to require protective
measures in timber sales contracts written for harvesting undertaken on their lands. We believe that
timber harvesting on these lands also will be regulated preferentially under existing authorities rather
than the proposed designation authority. As a result of the USFS and other public lands exclusions, we
ultimately assume that only timber harvested from privately owned lands may be covered by the
proposed designation authority.

       We estimate that a total of 1,602,784 MCF of timber was harvested from all private lands in the
affected counties in 1996.  This represents 9.9 % of the U.S. timber harvest in 1996, or 11.2 % of all
timber harvested on private lands. Applying regional harvesting intensity estimates developed by the
Timber Data Company, we also estimate the areal extent of forest land from which this volume was
harvested as 1,546,066 acres. Furthermore, applying regional timber price estimates developed by the
same source, we estimate the gross revenues from this quantity of timber harvest to be about $2.48
billion.45 The revenue estimate involves breaking out the timber volume in the counties into four
different categories for pricing purposes, and applying the Timber Data Company's estimated prices for
softwood sawtimber, softwood pulpwood, hardwood sawtimber, and hardwood pulpwood for 7 different
regions of the country.

       Finally, we  estimate  the number of entities pursuing silvicultural activities that could be affected
by the possible use of the designation authority  in the 167 counties. Silvicultural BMPs required
pursuant to designation and permitting could affect two sorts of entities: those conducting logging on
designated lands, and those owning the designated timber lands. Much more information is available on
logging entities than is available on entities owning timber land.
       44 Other private land is defined as private land owned by an entity that does not operate a wood-using
facility.

       45 Given the broad scope and tight schedule of the work that we requested of the Timber Data Company,
their data  on acres and log prices can only be described as "best estimate".

58

-------
       The Small Business Administration's analysis of 1996 Census data provides the following
information on economic activity in the logging industry (SIC 2410):

                                         Exhibit H - 6
                               Logging Economic Statistics, 1996
Industry
Logging, SIC 24 10
Total # of
Firms
14,278
# of Small
Entity
Firms
14,247"
Total
Receipts
(million S)
15,538
Receipts of
Small Entity
Firms (million S)
13,02 139
       In addition, there is some additional amount of logging that we cannot quantify that is conducted
by firms classified in other SIC codes. Some firms in SICs 2421,2426 and 2429 (various types of
sawmills) conduct logging as well as wood milling. Some firms in SICs 0811 (timber tracts) and 0831
(forest nurseries) conduct logging on timber land they own. The amount of logging conducted by firms
in these S  "s other than 2410 cannot be determined, but we believe it is relatively small. Industry
sources agree that the great majority of logging is performed by independent logging companies that
contract to provide wood to mills, and that only a small additional amount of logging is conducted by
vertically integrated firms that are involved primarily in either growing or milling wood. The data in
Exhibit II-6 should be regarded as probably a modest underestimate of the number of firms involved in
logging and their receipts.

       Some of the management measures to minimize adverse water quality impacts of silviculture will
apply to timber land owners rather than to loggers (e.g., measures to address use of chemicals applied to
timber land during forest regeneration after logging). A very wide range of entities  own the timber land
where designation might occur: large forest products companies, small forest products companies,
farmers, ranchers, entities holding timber land primarily for other purposes (real estate developers, ski
area operators, land speculators, second home owners, etc.) Only a very few of these timber land-owning
entities are identified by their SIC code as primarily engaged in management of timber land — those in
SIC 0811 (timber tracts) and SIC  0831 (forest nurseries).  Data compiled by the Small Business
Administration on firms in these two SICs is presented in Exhibit II - 7.
          Small entity logging firms are those with less than 500 employees.
                                                                                            59

-------
                                         Exhibit D - 7
                     Statistics on Finns in Timber Land-Owning SICs, 1995
Industry
Timber Tracts, SIC 08 10"
Forest Nurseries, SIC 0830
' Total
Total # of
Firms
571
98
669
# of Small
Entity
Firms
525
87
612
Total
Receipts
(million S)
743
161
904
Receipts of
Small Entity
Firms (million $)
248
68
31648
Apart from these two SIC codes, no recent comprehensive information is available on the number of
entities that own timber land or on their revenues or their size.  Some timber land-owning entities are
very large firms (e.g., the large, integrated forest products companies), while many others are small
farmers, ranchers or home owners.  The great majority of these entities obtain the bulk of their income
from activities other than owning timber land. Some data (much deriving from a 1978 nationwide
survey) on private ownership of timber lands is summarized below:49

•      7.8 million ownership units hold 333.1 million acres of private forest land in the U.S.;

•      About half of this forest land is  in tracts greater than 500 acres. These are owned by less than 1
       % of the ownership units.  About half of the land in these tracts is in tracts of greater than 10,000
       acres, owned largely by forest products companies. The ten largest forest industry firms in 1991
       owned more than 40 million acres of U.S. timber land. Typically, timber land owned by the
       forest products industry is more productive, managed more intensively, and more likely to be
       harvested than other timber land, accounting for a much higher proportion of the nation's timber
       harvest than the industry's  share of timber acreage.

•      30 % of private forest land is in ownerships of 100 to 500 acres;

•      About 20 % of private forest land is in ownerships of less than 100 acres. These small tracts
       likely account for more than 90 % of the ownership interests.

       After the forest products industry, fanners are the  occupational group that owns the largest
       fraction of private timber land.  They own 16 % of non-forest-industry timber land. The average
       holding for farmers owning timber land is more than 100 acres.
        47 The small business standarrj-for these two SICs is annual receipts of less than $5 million.

        48 Small entities in these two forest land-owning industries account for only 35 % of revenues in these two
industries. This is a lower share than the 84 % of revenues that is accounted for by small entities in the logging
industry. The logging industry is comprised predominantly of small businesses; this may be less true for timber
land ownership.

        49 This information is drawn from: Research Triangle Institute. Economic Analysis of Coastal Nonpoint
Source Pollution Controls: Forestry. 1992. Pages 2-26 through 2-32.
60

-------
Even though most timber land owners are classified in non-silvicultural SICs, they nevertheless conduct
many important silvicultural activities that need to be managed carefully in order to protect water quality,
including forest regeneration following logging, use of pesticides and fertilizers, road building, pre-
harvest planning, etc.. Some portion of the BMPs likely to be required of silvicultural operations that
will be designated under the proposed new authority will be the responsibility of the entities that own the
forest land rather than the other entities that conduct the logging. Unfortunately, though, we have no way
to identify specifically which or how many of these non-silviculture SIC entities (wood milling firms that
conduct logging, and forest land owners) will likely be affected by use of the designation authority. We
can project the number of likely affected entities only for entities classified in the silvicultural SICs for
which information is available.

        Timber production from private lands in the counties in which the designation authority might be
applied accounts for 9.9 % of all the timber harvested in the U.S.. How many logging firms might be
involved in producing this amount of timber? Absent specific information on the geographic scale at
which loggir.2 firms operate,50 we will assume that the designation authority will affect 9.9 % of all the
firm- in the . egging industry, or about 1,414 firms. We also assume that this set of potentially affected
firn.  will continue to operate in the areas where the designation authority is used, so that, in future years
folk  'ng designation, the cumulative number of affected firms will not expand significantly beyond the
origu ,u maximum of 1,414.Sl

        We use similar logic to estimate the number of entities owning timber lands that might be
affected by designation. The maximum acreage of timber land that might be designated annually is
1,546,606 acres, or about 0.464 % of the 333.1 million acres of private forest land in the U.S. If a similar
0.464 % of the 7.8 million owners of private timber land are affected by designation, roughly 36,200
owners might be affected. (In all  likelihood, the number of owners affected would be substantially less
than this,  because the designation authority will be used in areas where intensive logging is occurring.
Small landowners are much less likely to be  conducting intensive logging on their acreage than are large
land-owning forest products firms.)

        In sum, we estimate that the proposed designation provision could potentially be used to affect
the following amount of silvicultural activity:
        50 Exhibit II - 6 shows that more than 99.7 % of logging firms are small entities (employment less than
500). Industry sources indicate that the logging industry is highly localized, and that logging companies generally
operate within a radius of only 100 miles from their headquarters. We will therefore assume that such small firms
will operate only locally. Consequently, the projected widely scattered application of the designation authority will
affect the particular small firms that operate in these scattered areas, and these scattered areas only.  If designation
affects in a scattered way roughly 10 % of total logging activity, then it is also likely to affect roughly 10 % of the
small logging firms. A higher percentage of the large firms, with operations in multiple areas, could be expected to
be affected.

        51 Historically, the number of logging establishments has not varied to a significant extent. The 1992
Census of Manufactures provides data on the number of logging establishments in 1992, 1987, 1982 and 1977.
The number of establishments in these years has remained within a fairly narrow range - from a low of 11,658 in
1982 (18% below the 1996 figure of 14,359 establishments) to a high of 15,469 in 1977 (8% above the 1996
figure).

                                                                                                 61

-------
                                          Exhibit H - 8
              Summary of Silviculture! Activities Potentially Affected by Designation
Volume of wood cut
Area on which harvesting occurs
Logging receipts
Number of entities
1, 602,784 MCF per year
1,546,066 acres per year
$2.48 billion per year
logging firms: 1,414
forest land owners: < 36,200
This Exhibit summarizes the maximum extent of the silviculturai activity to which the proposed
designation authority may be applied.  The next step is to estimate how often the authority will be
applied.

       3c. Estimate how often the authority will be applied

       We believe that the proposed designation authority will ultimately be applied to much less than
this amount of silviculturai activity.  There are several reasons why designation will likely be needed for
less than the total volume of silviculture occurring on private lands in all 167 counties:

•      Most of the silviculture-impaired water bodies are small, with typical length of 5 - 25 miles.  It is
       probably rare that as much as an entire county's worth of silviculture will need to be controlled
       in order to mitigate the adverse impact of forestry on a water body of this size.  More typically,
       adverse impacts could be mitigated by controlling silviculture in only a small fraction of the
       county. Often, controls may be needed only in the impaired water body's immediate riparian
       zone, perhaps for only several hundred feet on both sides of an impaired river reach, for example.

•      We repeat that use of the designation authority is viewed as very burdensome by both permit
       authorities and potentially designated sources. The designation authority for AFOs that has
       existed for several decades has been used only a handful of times. States will undoubtedly
       attempt to obtain the desired improvement in performance of silvicultural operations by all other
       means possible  — voluntary and incentive programs, technical assistance, partially enforceable
       programs, perhaps lawsuits under broad State authorities seeking to enjoin harmful practices as a
       nuisance or injurious to fish and wildlife — before invoking the designation authority.

•      Silvicultural operators themselves will often prefer voluntarily to adopt acceptable practices
       rather than to be designated and become regulated under the NPDES program.
      /
There is little basis, however, for estimating quantitatively how much less than all the silviculture in
these counties is likely ultimately to be designated. We will assume a range for the extent to which the
proposed designation authority will be used:

•      For a county containing at least one water body that is impaired by silviculture to a major extent,
       the designation  authority will  be used to cover 50 - 100 % of all the silviculture  occurring on
       private land in the county;
62

-------
  •      For a county containing at least one water body that is impaired by silviculture to a moderate
         extent but no water body impaired to a major extent, the designation authority will be used to
         cover 25 - 50 % of all the silviculture occurring on private land in the county.

  And, as discussed previously, we assume that the designation authority will not be used at all in States
  with comprehensive enforceable authorities to require silvicultural BMPs, nor for silviculture on public
  lands, nor for silviculture when it causes only a minor impairment of a water body. Exhibit II - 9 shows
  the resulting estimates (not yet scaled) of how often the designation authority will be used:

                                            Exhibit n - 9
                         Unsealed Estimates for Use of Designation Authority

All silviculture on
private lands in
these counties
High estimated use
of designation
authority
(100% of major
counties, 50 % of
moderate counties)
Low estimated use
of designation
authority
(50 % of major
counties, 25 % of
moderate counties)
# of counties
volume cut (mcf/yr)
acres cut/yr
receipts (106$/yr)
# of entities
(loggers, owners)
volume cut (mcf/yr)
acres cut/yr
receipts (106$/yr)
# of entities
(loggers, owners)
volume cut (mcf/yr)
acres cut/yr
receipts (106S/yr)
# of entities
(loggers, owners)
Counties w/ Major
Impairments
14
74,975
51,080
128
66
1,693
74,975
51,080
128
66
1,693
37,488
25,540
64
33
847
Counties w/ Moderate
Impairments
153
1,527,809
1,494,986
2,352
1,348
34,507
763,905
747,493
1,176
674
17,254
381,952
373,747 '
588
337
8,627
Total for AH
Counties
167
1,602,784
1,546,066.
2,480
1,414
36.200
838,880
798,573
1,304
740
18.947 •
419,440
399,287
652
370
9.474
        3d.  Scale the estimates to account for non-reporting States and the projected increase in States
        adopting comprehensive enforceable authorities.

        Our final step in estimating the volume of silvicultural activity to which the proposed designation
authority will be applied was to scale these unsealed estimates. The scaling procedure is intended to reflect
two factors:
                                                                                                 63

-------
•       Some States did not report in the 305(b) data base in a manner allowing identification of water
        bodies impaired by silviculture. Only 33 States used a system for reporting that included detailed
        codes for causes of impairment that included silviculture as a possibility.  17 States either:  1)
        reported impairments without identifying any causes; or 2) reported impairments with causes
        identified only at a higher level of aggregation (e.g., impairment by point sources vs. nonpoint
        sources vs. mixed sources).  25 of the 33 States that might have identified silviculture as a cause of
        impairment actually did report water bodies impaired by silviculture. Thus 8 States that did not
        report any water bodies impaired by silviculture either: 1) Reviewed their water bodies for
        silvicultural impairments and actually had none; or 2) May have had silvicultural impairments, but
        did not report any in a manner such that they appeared in EPA's national 305(b) data base. We
        therefore need to extrapolate from the 25 States reporting silvicultural impairments to either the
        additional 17 States that, in effect, did not report at all in a relevant manner, or to the remaining 25
        States that did not report any impairments. To be conservative, we will take the latter course.  We
        will extrapolate from the 25 States reporting impairments to the 25 States that report none,
        assuming conservatively that each of the 25 States reporting no impairments does have
        impairments, but simply did not report.

•       We also need to scale to reflect our assumption that the number of States with comprehensive
        enforceable silviculture authorities (States in which the proposed designation authority will not be
        used at all) will grow over time.  20 States have such authorities now. If the rate at which States
        develop such authorities continues at the same pace as in the past, by 2008, 10 more States will
        have such authorities. Our unsealed estimates for the extent to which the proposed designation
        authority will be used now reflect the 30 States currently without such authorities. We need to
        scale these estimates down by 1/3 to reflect the projection rate at which these authorities will be
        established as TMDLs are completed.

        We choose to reflect these two scaling concerns by using a numerical procedure based on the
volume of timber cut from private lands.52 To extrapolate from the 25 States reporting silvicultural
impairments to the 25 other States that are assumed to have impairments that they did not report, we will
assume that the rate of impairments in the two groups of States is proportional to their respective volumes
of timber harvested from private lands. To reflect the assumption that 1/3 of the States (10 of 30) now
without enforceable authorities will adopt them by 2008, we assume that 1/3 of the timber now harvested
from private lands in the 30 States will be eliminated from potential coverage under the designation
authority as a result of additional States adopting enforceable authorities.  Exhibit II - 10 shows the data on
timber harvest from private lands from which these two scale factors are calculated. •
          52 We might have chosen instead to scale based on other factors: the acreage cut, timber revenues, timber
  harvested from all lands rather than only private lands, etc.. We believe that since our compliance cost estimates
  flow most directly from data on timber harvested from private lands, this data element provides the best basis for
  scaling.
64

-------
                                         Exhibit H -10
                 Timber Harvest Data From Which to Calculate Scaling Factors

Number of States
Volume of timber cut from private lands in the 25 States
that reported silvicultural impairments (mcf/yr)
Volume of timber cut from private lands in the remaining
25 States (that did not report impairments) (mcf/yr)
States Now With
Enforceable Authorities
20
5,058,424
States Now Without
Enforceable Authorities
30
6,139,289
3,032,817
       For the first extrapolation, we want to scale from the States that reported (6,139,289 mcf
harvested) to the States that did not report and in which the designation authority might be used
(3,032,817 mcf harvested). This scale factor is 1.49 ([6,139,289 + 3,032,817]/6,139,289). In other
words, to extend our estimate of the impact of the designation provision from the reporting States to the
non-reporting States in which designation might be used, we should multiply our unsealed impact
estimates by  1.49.

       The second extrapolation, however, nearly exactly cancels out this first one.  For the second
extrapolation, we want to reflect the assumption that the number of States without enforceable authorities
will shrink by 1/3 relative to what it is now. We assume that both the 6,139,289 mcf cut in reporting
States presently without enforceable authorities and the 3,032,817 mcf cut in non-reporting States
presently without enforceable authorities will shrink by 1/3. Summing the two sets of States presently
without enforceable authorities, the total timber harvest to which the designation authority will
potentially be applicable in 2008 will be (2/3) x (6,139,289 + 3,032,817), or 6,114,738 mcf.

       In effect, in the previous subsection of this report, we estimated the degree to which the proposed
designation authority will be used in a sample of reporting States presently without enforceable
silviculture authorities. These States have annual timber harvest of 6,139,289 mcf.  We now want to
extrapolate these usage estimates to a revised sample of both reporting and non-reporting States presently
without enforceable silviculture authorities, and the revised sample includes annual timber harvest of
6,114,738 mcf.  Our previous unsealed estimates covered States with harvest of 6,139,289 mcf, and we
want to scale to harvest of 6,114,738 mcf. To scale our previous unsealed usage estimates, we multiply
by a scale factor of .996 (6,114,738/6,139,289). Applying this scaling factor, Exhibit II - 11 shows our
scaled low and high estimates for the volume of silvicultural activity to which the proposed designation
authority will be applied:
                                                                                               65

-------
                                        Exhibit n-11
                Final, Scaled National Estimates for Use of Designation Authority

# of counties
volume cut (mcf/yr)
acres cut/yr
logging receipts (106 $/yr)
# of entities (loggers)
(owners)
Low Estimate
166
417,762
397,690
649 '
369
9,436
High Estimate
166
835,524
795,379
1,299
737
18,871
We will apply unit cost estimates developed in the next several subsections of this report to these activity
estimates in order to'project the costs of the proposed regulation.

4. Determine the set of silvicultural BMPs that will typically be required of designated and
permitted silvicultural operations and the per unit costs for these BMPs

       There exists a fairly typical set of BMPs that silviculture operations should employ to minimize
their impacts on water quality. These practices, and the particular combination of practices used, will
vary somewhat according to the type of silviculture operation and the conditions (e.g. region, climate,
topography, soils) of the silviculture site. Examples of BMPs for silvicultural activities are available
from several sources, among them EPA's 1993 Guidance Specifying Management Measures for Sources
ofNonpoint Pollution in Coastal  Waters. The Guidance specifies 10 management measures for use in
coastal States (including the Great Lakes States) to protect waters from silvicultural sources of nonpoint
pollution, and lists and describes management practices that have been found by EPA to be representative
of the types of practices that can be applied successfully to achieve the management measures. The 10
management measures, and examples of BMPs to fulfill the objectives of each measure, are as follows:

1) Preharvest planning - Performing advance planning for forest harvesting to minimize negative water
quality impacts. BMPs include laying out harvest units to minimize the number of .stream crossings,
systematically designing transportation systems to minimize total mileage; minimizing road and skid trail
grades; and surfacing roads (with gravel, grass, crushed rocks, etc.) where grades increase the potential
for surface erosion.

2) Streamside Management Areas - Establishing and maintaining a streamside management area (SMA)
along surface watery which is sufficiently wide and which includes a sufficient number of canopy
species to buffer against detrimental changes in the temperature regime of the water body, to provide
bank stability, and to withstand wind damage. BMPs include providing a minimum SMA width of 35 to
50 feet; avoiding operating skidders or other heavy machinery in the SMA; and applying harvesting
restrictions in the SMA to maintain its integrity.

3) Road Construction/Reconstruction - Minimize delivery of sediment to surface waters during road
construction/reconstruction projects. BMPs include following the design developed during preharvest
66

-------
planning to minimize erosion by properly timing and limiting ground disturbance operations; using straw
bales, grass seeding, and other erosion control and revegetation techniques to complete the construction
project; and installing surface drainage controls to remove storm water from the roadbed before the flow
gains enough volume and velocity to erode the surface.

4) Road Management - Managing existing roads to maintain stability and utility and to minimize
sedimentation and pollution from runoff-transported materials. BMPs include maintaining road surfaces
by mowing, patching, or resurfacing as necessary; and revegetating to proving erosion control.

5) Timber Harvesting - Minimizing sedimentation resulting from the siting and operation of timber
harvesting, and managing petroleum products properly.  BMPs include felling tr^es away from
watercourses; removing slash from the water body and placing it out of the SMA; minimizing the size of
landings; skidding uphill to landings whenever possible; avoiding cable yarding in or across
watercourses; and taking precautions to prevent fuel leakage and spills.

6) Site Preparation and Forest Management - Regeneration of.harvested forest lands. Examples of BMPs
include avoiding mechanical site preparation on slopes greater than 3 0 percent or in SMAs; distributing
seedlings evenly across the site; and hand planting highly erodible sites, steep slopes, arid SMAs.

7) Fire Management - Minimizing potential NFS pollution and erosion resulting from prescribed fire for
site preparation and from the methods used for wildfire control or suppression. BMPs include planning
burning to achieve the desired results while minimizing impacts on water quality; avoiding intense
prescribed fire or construction of firelines in SMAs; avoiding burning on steep slopes with high-erosion-
hazard areas or highly erodible soils.

8) Revegetation of Disturbed  Areas - Preventing sediment and pollutants from entering water bodies by
revegetating disturbed soil. BMPs include using seed mixtures adapted to the site; using native woody
plants planted in rows, cordons, or wattles on steep slopes; and seeding during optimum periods for
establishment, preferably just prior to fall rains.

9) Chemical Management - Minimizing the water quality impact of the use of pesticides and fertilizers.
BMPs include maintaining a buffer area around all water bodies for aerial spray applications; and
applying pesticides and fertilizers during favorable atmospheric conditions, and during maximum plant
uptake periods to minimize leaching.

10) Wetlands Forest Management - Taking special measures to protect beneficial wetlands functions and
avoid water quality impacts in wetlands areas. BMPs include providing adequate cross  drainage to
maintain the natural  surface and subsurface flow of the wetland; and establishing a SMA adjacent to
natural perennial streams, lakes, ponds, and other standing water in the forested wetland.

        In order to cost out a  typical set of BMPs, we used the set of BMPs and costs provided in EPA's
1992 report, "Economic Analysis of Coastal Nonpoint Source Pollution Controls: Forestry", prepared by
the Research Triangle Institute (RTI). The report analyzes the economic achievability of the
management measures cited above and contained in the Guidance Specifying Management Measures for
Sources of Nonpoint Pollution in Coastal Waters. RTI costed out the potential BMPs for each
management measure that it judged would most likely be adopted by all the coastal states in each major
forest region.
                                                                                              67

-------
       RTI provided two sets of BMP costs for Forest Management Units (FMUs) with Region-Average
Volumes — one applicable to South Coastal Areas, the other to North Coastal Areas.  Accordingly, we
assigned data from each of the 18 States (containing the 167 counties) into South and North for cost
calculations.  For both the South and North, RTT reported cost data for dry (no streams or water bodies)
and wet FMUs. Because we are interested in silviculture activities in areas where they impair water
bodies, we chose to use RTI's cost data for wet FMUs only. RTI further broke BMP costs out according
to physiographic class and forest type.  We were unable to make full use of these further distinctions,
because we have no data on physiographic class and forest type for our 167 counties.  We established
only two categories for differentiating BMP costs: BMP costs when applied on flat land, and BMP costs
when applied on hilly land.  RTI's BMP costs for hilly areas are markedly higher than those for flat
areas. Using available information about the topography of each of the 18 states in which the 167
counties are located, we roughly estimated the percentages of flat and hilly hand in each State, and
weighted RTI's flat and hilly land BMP costs according to these percentages for each State.

       Finally, RTI provided unit costs for BMPs based on both: 1) Acreage harvested, and 2) Volume
harvested. Exhibit II - 12 shows the topography percentages, the original RTI BMP cost figures in 1989
dollars on a per-acre and per-100-cubic-feet-harvested basis, and the cost figures for each escalated to
current dollars.
68

-------
                                                           Exhibit II-12
                                     BMP costs by state per acre and per 100 cubic feet harvested

Flat %
Hilly
%
Wei %
RTI cost/acre"
in 1989$
for Hal, wet.
forest land
Updated
cost/acre
for flat, wel
forest land
RTI cost/acre*
in 1989$
for hilly, wet
forest land
Updated
cost/acre
for hilly, wet
fores) land
RTI cost/100
cu .ft.*
in 1989$
for flat, wel
forest land
Updated
cost/100 cu. ft
for Hal, wel
forest land
RTI cost/100
cu .fl.*
in 1989$
for hilly, wel
forest land
Updated
cosl/IOOcu. fl
for hilly, wel
forest land
\
Alabama
Arizona
Arkansas
Florida
Ilinois
Louisiana
Michigan
Minnesota
Mississippi
Montana"
^ew Mexico
North Dakota
Oklahoma
Rhode Island
South Carolina
South Dakota
Tennessee
Wyoming"
75%
50%
75%
95%

75%


95%

65%

90%

65%

50%

25%
50%
25%
5%

25%


5%

35%

10%

35%

50%

100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
$20.43
$20.43
$20.43
$20.43
$15.04
$20.43
$15.04
$15.04
$20.43

$20.43
$15.04
$20.43
$15.04
$20.43
$15.04 '
$20.43 .

$23.49
$23.49
$23.49
$23.49
$17.30
$23.49
$17.30
$17.30
$23.49
$33.14
$23.49
$17.30
$23.49
$17.30
$23.49
$17.30
$23.49
$33.14
$27.82
$27.82
$27.82
$27.82
$15.04
$27.82
$15.04
$15.04
$27.82

$27.82
$15.04
$27,82
$15.04
$27.82
$15.04
$27.82

$31.99
$31.99
$31.99
$31.99
$17.30
$31.99
$17.30
$17.30
$31.99
$33.14
$31.99
$17.30
$31.99
$17.30
$31.99
$17.30
$31.99
$33.14
$0.96
$0.96
$0.96
$0.96
$1.36
$0.96
$1.36
$1.36
$0.96

$0.96
. $1.36
$0.96
$1.36
$0.96
$1.36
$0.96

$1.10
$1.10
$1.10
$1.10
$1.56
$1.10
$1.56
$1.56
$1.10
$1.68
$1.10
$1.56
$1.10
$1.56
$1.10
$1.56
$1.10
$1.68
$1.46
$1.46
$1.46
$1.46
$1.36
$1.46
$1.36
$1.36 .
$1.46

$1.46
$1.36
$1,46
$1.36
$1.46
$1.36
$1.46

$1.68
$1.68
$1.68
$1.68
$1.56
$1.68
$1.56
$1.56
$1.68
$1.68
$1.68
$1.56
$1.68
$1.56
$1.68
$1.56
$1.68
$1.68
*  Average price across all forest types
** We .assigned Wyoming and Montana the highest of RTI's costs (those for hilly Southern coastal forests) because the extensive mountainous
terrain in WY and MT will tend to increase costs over and above those for any of the terrains addressed in the RTI study.
                                                                                                                                69

-------
5. Estimate total national BMP costs

        To estimate total national BMP costs, we applied the per unit BMP costs (State-by-State per-
volume and per-acre cost estimates, weighted according to the flat and hilly terrain percentages, as
shown in Exhibit II - 12) to the scaled State-by-state amounts of timber harvest volume and acreage that
we projected to be designated using the proposed authority.5" The estimated BMP costs resulting from
use of the designation authority are shown in Exhibit II - 13.

                                         Exhibit D - 13
                BMP Costs Resulting From Use of Proposed Designation Authority
            ,                   (After all Sources are Designated)

Volume of timber harvest to be covered by designation
(mcf/yr)
Volume-based estimate of BMP costs ($/yr)
Acreage of timber harvest to be covered by designation
(acres/yr)
Acreage-based estimate of BMP costs ($/yr)
Low Estimate
417,762
$5,287,269
397,690
$10,058,907
High Estimate
835,524
$10,574,538
795,379
$20,117,813
        These estimates reflect the costs that would occur if all TMDLsfor silviculture-impaired waters
and all designations pursuant to these TMDLs were to occur in the first year following promulgation of
the proposed designation authority (assumed to be the year 2000). Alternatively, these estimates show
the annualized BMP costs once all the silvicultural operations that will be designated have been
designated.  However, all TMDLs will not be completed immediately in the year 2000.  Many TMDLs
for silviculture-impaired waters will be developed in years subsequent to 2000, and many designations of
silvicultural sources in order to implement these TMDLs will therefore not occur until well after the year
2000. The present value costs for many designated silvicultural operations to implement the required
BMPs will be  reduced because designation of many of them will occur later than 2000.

        Proposed TMDL regulations accompanying this proposed rule will require the TMDL for a water
body to be completed within 15 years following listing of the water body.  The proposed TMDL
regulations will presumably be finalized in the year 2000, and the first 303(d) list to which the 15-year
deadline will apply will  be the one submitted by States in 2000. All TMDLs for waters on the 2000 list
will therefore  need to be completed by the end of the year 2015. The proposed regulations will further
require States  to submit  schedules for completion of these TMDLs at a steady pace over this period.  We
therefore assume for this analysis mat all TMDLs will be completed at a steady pace during the 16 years
from 2000 through 2015. We also assume that  designation  of sources as is necessary to implement these
TMDLs will similarly occur at an even pace over these 16 years.  The result is a projection that one-
sixteenth of the silviculture operations to be designated as a result of this rule will be designated in 2000.
        53 In extending the BMP costs to additional States via the scaling procedure, we are assuming that the
terrain conditions in the States to which we are scaling are similar to those in the States for which we have data.
70

-------
another sixteenth will be designated in 2001, etc., and the final sixteenth will not be designated until
2015. This gradual designation process significantly reduces the present value of the BMP costs that will
result from the proposed designation authority. Exhibit II - 14 shows how, for a hypothetical gradual
designation process in which 100 sources will be designated at an even pace over the years from 2000
through 2015.

                                         Exhibit 11-14
                          Calculating the Impact of Gradual Designation
Total number of sources to be designated 1 00
Total annualized costs once sS sources are designated 51 ,000,000
Newly designated sources each year 6.25
Annualized costs for these sources $62,500

Year
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015


Cumulative Number of Total Annualized
Sources Designated Costs in This Year
6.3 $62,500
12.5 $125,000
37.5 $187,500
25.0 $250,000
31.2 $312,500
37.5 $375,000
43.8 $437,500
50.0 $500,000
56.2 $562,500
62.5 $625,000
68.8 $687,500
75.0 $750,000
81.2 $812,500
87.5 $875,000
93.8 $937,500
100.0 $1,000,000
Present value for all years 2000 through 2015
Annualized cost equivalent to this present value
Present value at beginning of 201 6 for annualized costs continuing forever after 2015

Present value in 1999 equivalent to this
Present value of annual costs beginning in 2000 and continuing forever
Annuity value of these costs (annualized cost continuing forever)
Present Value of
These Costs
$58,411
$109,180
$1 53,056
5190,724
5222,808
5249,878
5272,453
$291,005
$305,963
$317,718
$326,626
$333,009
$337,159
$339,340
$339,793
$338,735
$4,185,858
, $443,105
514,285,714
$4,839,066
$9,024,923
$631,745
                                                                                              71

-------
The exhibit shows the hypothetical 100 sources being designated at an even pace over the 16 years from
2000 through 2015. The average source, once it is designated, will incur an annualized cost of $10,000
per year, continuing forever. All 100 sources, when they are all designated, will incur a total annualized
cost of $1 million per year, continuing forever. In the first year (2000), one sixteenth of the sources will
be designated, and they will begin incurring their annualized costs. In 2015, the last of the 16 years, the
final cohort of sources will be designated, bringing the cumulative number of designated  sources to 100.
It will not be until this year that all the designated sources will incur a total annualized cost of $ 1 million
annually. In years beyond 2015, the 100 designated sources will continue to incur total annualized costs
of $1 million per year.  At the bottom of the exhibit, the present values (discounted at 7 % annually) of
several of these cost streams are calculated.
                                                                                              /
       The present value of the costs incurred in 2000 through 2015  is $4.186 million. The annualized
       cost equivalent (continuing forever) of this present value is 5443,105.

•      The present value at the beginning of 2016 of $1 million in costs in 2016 and for all years
       thereafter (i.e., $1 million also in 2017, in 2018, etc.) is $14.286 million.  Discounting this 2016
       present value back to 1999 gives a present value of $4.839 million.

•      Adding the two present values ($4.186 million for the gradually increasing costs  in 2000 through
       2015; and $4.839 million for the continuing $1  million in costs in 2016 and onward), the present
       value of the entire stream of costs is $9.025 million.

•      The annuity equivalent of this present value is $631,745. This is the amount which, if paid
       annually forever, is equivalent to the  entire stream of costs from 2000 through 2015 and onward.
       The annualized equivalent of the entire cost stream is 63.1745 % of the annualized cost that is
       reached in the sixteenth year when all of the sources are finally designated.

       This latter factor is the important result from this exhibit. The annualized cost (to be incurred
each year forever, beginning in the first year) is equivalent to 63.1745 % of the annualized cost once all
the sources have been designated. We will apply this .631745 factor to scale down the BMP costs we
have estimated so as to reflect that fact that the silvicultural operations will be designated gradually over
16 years  rather than immediately.

       The impact of gradual designation over 16 years can also be shown graphically.  Exhibit II - 15
shows the annualized costs incurred in each year by the gradually increasing number of silvicultural
sources designated under the low estimate, using volume-based  costs. In 2000, one-sixteenth of the
silvicultural operations have been designated, and the annualized cost in this year is 1 /16 x $5,287,269,
or about  $330,000. This annual cost grow through 2015, when all the operations have been designated
and the annualized costs reach $5,287,269 per year. Costs then continue at this level for  each year
beyond 2015.  The exhibit also shows the annuity amount (a constant annual amount, continuing forever)
that is equivalent to this increasing then level cost stream.  The annuity equivalent, shown in shaded gray
color, is  .631745 times the $5,287,269 annual figure that prevails after all the sources are designated.
The annuity equivalent is $3,340,206.
72

-------
        Exhibit n - 15. Graph of Annuity Equivalent to Costs With Gradual Designation
    I Costs/yr (million $)


       2000    2005
2010     2015
       Year
2020    2025
                                                                       Actual
                                                                    Annual Cost
                                                                     (2000 and
                                                                      onward)

                                                                      Annuity
                                                                     equivalent
                                                                     (2000 and
                                                                      onward)
       After applying this scaling factor of .631745 to reflect gradual designation of the silvicultural
operations over 2000 - 2015, the estimated annualized BMP costs resulting from the designation
provisions are shown in Exhibit II - 16.

                                       Exhibit n - 16
         Annualized BMP Costs Resulting From Use of Proposed Designation Authority
                         (With Gradual Designation Over 2000 - 2015)

Volume of timber harvest to be covered by designation (in
years 2015 and later) (mcf/yr)
Volume-based estimate of BMP costs ($/yr)
Acreage of timber harvest to be covered by designation (in
years 2015 and later) (acres/yr)
Acreage-based estimate of BMP costs ($/yr)
Low Estimate
417,762
$3,340,206
397,690
$6,354,664
High Estimate
835,524
$6,680,412
795,379
$12,709,328
       We thus estimate the annualized costs of the BMPs to be required of designated silviculture
operations ~ assuming gradual designation over 2000 - 2015 —to range from $3.3 million annually to
$ 12.7 million annually. Lower estimates are obtained by applying RTTs cost figures on a per-volume
basis, and higher estimates are derived by applying cost figures on a per-acre basis.  Our estimates also
                                                                                        73

-------
reflect substantial uncertainty about what fraction of the silviculture] operations in a county with
silviculture-impaired water bodies will ultimately need to be designated.  Our low assumptions with
regard to use of the designation authority result in national BMP costs estimated at $3.3 - $6.4 million
annually, while our high assumptions result in costs estimated at S6.7 - $12.7 million annually.

6. Estimate administrative costs for sources

        Silvicultural operations within areas where the designation authority will be invoked will need to
incur the administrative costs of seeking and maintaining coverage under the general permits that the
permit authority will  issue. We assume that a silvicultural entity operating in one of these areas will
need to:

•       File an initial Notice of Intent (NOI) to seek coverage under the general permit. The NOI will
        likely  include identification of the applicant, a description of the planned silvicultural activity
        and its location and duration, site information, information on potentially affected water bodies,
        and perhaps more. Such information is not unlike that required forNOIs for coverage of
        industrial facilities or construction activities under the storm water general permitting program.
        We will assume that the effort required to prepare such a silvicultural NOI will be the same as
        that for the storm water NOI - estimated at 2 hours per NOI in EPA's ICRfor Notice of Intent
        and Stormwater Pollution Prevention Plan for Discharges Associated with Industrial Activity
        (March, 1999).

•       Update or amend the NOI whenever timbering operations are expected to move to another
        location within the designated area.  If we assume such an amendment will be necessary
        approximately annually, then the burden associated with the NOI and amendments becomes 2
        hours annually rather than 2 hours on a one-time basis.

•       Prepare some sort of silvicultural pollution prevention plan, indicating how the management
        measures required by the general permit will be translated into specific BMPs to be employed in
        the applicant's operations.  We have not researched further details on what such a plan might
        contain. Some indications could probably be obtained by reviewing the requirements in States
        that already implement permit-like regulatory programs governing forest practices.  Absent
        further information, we will assume that the burden associated with such a requirement would be
        similar to that for the Storm water Pollution Prevention Plans required under the storm water
        general permit program.  This one-time burden is estimated in the relevant ICR at 80 hours.

To these burden estimates, we apply a costing factor of $36.12 per hour for fully  loaded private sector
labor costs, drawn from the recent ICR for Applications for the NPDES Discharge Permit and the
Sewage Sludge Management Permit (December, 1998).
                                      /
        Within a geographic area where the proposed designation authority has been invoked, we assume
that designation will apply specifically to the land on which silvicultural activities occur. However, we
assume that either the logger or the land owner can perform the administrative activities involved in
seeking and maintaining coverage under the general permit. Most likely, when a logger is conducting
harvesting within a designated area on many small privately owned timber tracts, the logger will pursue
general permit coverage for the timber lands owned by all of the different owners. Conversely, when the
owner of a large forest tract within the designated area contracts for logging with multiple operators, the

-------
 large landowner is likely to seek general permit coverage for multiple operations across all of his land.
 We assume that either the logger or the owner will apply for general permit coverage for a timber tract,
 and that both of them will not apply. The result of these assumptions is that the maximum number of
 entities that will seek general permit coverage in a designated area is the lesser of the number of affected
 loggers and the number of affected private timber land owners. Since in most areas of the country the
 number of private timber land owners is substantially greater than the number of logging entities, we
 assume that the number of loggers will be the maximum number of entities that will seek general permit
 coverage.

        The total estimated administrative cost to covered silvicultural sources is thus:
r
        Annual costs for filing and updating NOIs:                    S 26,657 - $53,241
        (369 - 737 logging firms x 2 hours annually x $36.12/hr)

        One-time costs for developing silvicultural PPPs:              Sl,066,262 - $ 2,129,635
        (369 - 737 firms x 80 hours x S36.12/hr)

        Converting the one-time costs to annualized costs by assuming that PPPs remain valid for 10
 years and applying a discount rate of 7 % (resulting in a capital recovery factor of 0.1424), we obtain an
 annualized cost for the PPPs of $151,836 - $303,260.

        Adding the costs for NOIs and the costs for PPPs, we estimate that the total administrative cost
 to sources will be $178,493 - $356,501/yr. This is the estimated cost if all sources were designated
 immediately. To reflect gradual designation over 2000 - 2015, we multiply by the scaling factor of
 0.631745. The result is an estimated annualized administrative cost to sources of $112,762 - $225,218.

 7.  Estimate the administrative costs to Federal and State governments for designating and
 permitting these silvicultural operations.

        We assume that there will be one silvicultural general permit issued for each county in which
 silvicultural operations will be designated as point sources.  We assume that the silvicultural BMPs
 required will be somewhat different as a function of topography, etc., from county to county, so a single
 statewide general permit would not be appropriate. 166 general permits will thus be issued.  These
 general permits will be issued over the 16-year time period  (2000 - 2015) during which all TMDLs are
 expected to be completed. We assume therefore that an average of 10.3 75 general permits will be issued
 annually during this period.

        EPA's "Revised Workload Model for the NPDES Program" (Jim raft, Rob Wood, May 5, 1995)
 estimates the following pricing factors for general permits:

 •      0.5 FTE per general permit for issuance.  In issuing 10375 general permits annually, there will
        be an annual cost of 5.19 FTE in each year through 2015.  After 2015, all the general permits
        will have been issued and there will be no further cost. We assume that it makes no difference
        whether a State or EPA issues the permit.

 •      0.125 FTE/yr/State for continuing review of existing general permits. We assume that these new
        silviculture  general permits will constitute about 1/5 of the volume of general permits in the

                                                                                              75

-------
        States that issue them. We assume that all 20 States in which the designation authority is
        projected to be used will issue silviculture general permits, and that this will be accomplished
        over the 8 years from 2000 - 2007. If so, the annual EPA review burden will increase from .063
        FTE (0.125 x 1/5 x 2.5) in 2000 to 0.5 FTE (0.125 x 1/5 x 20) in 2007 and thereafter.

        In addition, we assume that the permit authority will need to spend an average of 2 hours to
process and review each NOI that is submitted. The number of NOIs submitted that will need processing
and review will grow from 23.1 - 46.1 in the year 2000 (the initial NOIs for the one-sixteenth of the 369 -
737 silvicultural operations that will be designated in the first year) to 369 - 737 in the years 2015 and
onward (assumed annual NOI revisions for the 369 - 737 sources that will have been designated through
2015).

        To these burden estimates,  we apply a costing factor of $65,895 per FTE ($31.68 per hour) for
fully loaded State or  Federal labor costs,  drawn from the recent ICR for Applications for the NPDES
Discharge Permit and the Sewage Sludge Management Permit (December, 1998).

        Summing these streams of administrative costs for EPA and the States, discounting them back to
the present and annualizing them, the resulting estimates are:

•       Costs for general permit issuance:                                    $226,014/yr

•       Costs for EPA review of State general permits:                         $26,314/yr

        Costs for processing and review of NOIs                             $ 14,750 - $29,500/yr.

        We therefore estimate the total annualized administrative cost to EPA and the States of the
designation provision to be $267,105 - $281,855/yr.

8. Develop estimates of BMP costs and administrative costs for small entities

        Two sorts of small entities  may be affected by the proposed designation provision: small logging
firms, and small entities that own timber lands. We will estimate impacts on these two groups
separately.

        The costs for some of the BMPs that will likely be required of designated and permitted
silvicultural operations will fall on timber land owners, while the costs of others will fall on loggers.
Exhibit 11-17 shows each of the BMPs and the group that will initially incur the cost for each of the '
BMPs.54
        54 The dynamics of the timber market may be such that costs initially incurred by loggers get passed
backward to land owners in the form of lower stumpage prices paid for timber cut from an owner's land.  Or, costs
initially incurred by owners or loggers may eventually be passed forward to mill operators in the form of higher log
prices and even further in higher prices to forest product customers. The ultimate incidence of the BMP costs could
be estimated by use of a timber market model, but such an effort is beyond the scope of this project.  We will
estimate the impacts on small entities from silvicultural BMP costs by assuming that the entity (loggers or owners)
initially paying a cost is also the entity that will ultimately bear the cost after all market adjustments.

76

-------
                                         Exhibit H -17
                                 Who Incurs the Costs of BMPs?
BMP
Preharvest planning
Streamside management areas
(SMAs)
Road construction/reconstruction
Road management
Timber harvesting
Regeneration of harvested lands
Fire management
Revegetation of disturbed areas
Chemical management
Wetlands forest management
Affected
Group
Loggers
Owners
Loggers
Loggers
Loggers
Owners
Owners
Loggers
Owners
Owners
Discussion
The first activity conducted by loggers at a new site
Setting aside areas adjoining streams reduces the
volume of timber that a land owner can sell from his
acreage
Typically the responsibility of loggers, though
sometimes the owner constructs the roads
By loggers during harvest, by owners thereafter

Site preparation, reseeding and ongoing forest
management are conducted by land owners between
harvesting cycles. Note: most forest land is not
replanted following harvest.
Same as above
Conducted by loggers before they leave a site
A part of ongoing forest management. Pesticides and
fertilizers are used in forest regeneration following
harvest. Note: most forest land is not actively
managed.
Reduces the amount of harvestable timber on the
owner's acreage, like SMAs
The most costly of these BMPs for atypical silvicultural operation are Streamside management areas,
proper road construction, and careful timber harvesting. The relative cost of different BMPs varies
greatly from site to site, however. For example, preserving Streamside management areas adds no
incremental cost in harvesting a timber plot that includes no streams, but can be quite costly for plots
with streams.

        The RTI study that has provided the basis for our cost estimates provides costs for most of these
BMPs for several dozen varying modeled forest management units. Across the model units that include
streams, the total cost of the management measures that will be paid for initially by loggers is
approximately the same or slightly more than the total cost of the management measures that will
initially be the responsibility of land owners. Across the model units that do not include streams, and
that therefore do not require Streamside management areas, relatively little of the total cost of the
management measures falls on owners. To reflect the fact that a large proportion of the total acres to be
designated do not contain streams, we will assume that one-third of the compliance costs associated with
designation will accrue to land owners, and the remaining two-thirds of the costs will accrue to loggers.
                                                                                             77

-------
       Impacts on small logging entities

       Exhibit 11-18 shows the projected impacts of designation on small logging entities. The exhibit
ultimately compares the total logging revenues received by small logging firms within designated areas
with the summed administrative and BMP costs incurred by these firms within the designated areas.  We
conclude that the average small logging company will incur compliance costs equal to 0.36 % to 0.67 %
of revenues, depending on whether costs are estimated using RTI's volume-based or acreage-based
numbers.55 We have no information on the variation of costs or revenues across small logging firms, and
hence cannot estimate anything further about the distribution of impacts across firms other than the
average impact of 0.36 % to 0.67 %. We estimate that 368 - 735 small logging firms will eventually (by
2015, when all designation is complete) incur these impacts from use of the proposed designation
authority.

                                        Exhibit H - 18
                             Impacts on Small Logging Firms (S/yr)

Administrative costs
Total BMP costs — volume based
— acreage based
BMP costs to loggers — volume based
— acreage based
Total costs to loggers — volume based
— acreage based
Logging receipts
Costs as % of revenues — volume based
— acreage based
AH logging in
designated areas
Low estimate
$112,762
$3,340,206
$6,354,664
$2,227,917
$4,238,561
$2,340,679
$4,351,323
. $649,000,000
0.36%
0.67%
High
estimate
5225,218
$6,680,412
$12,709,328
$4,455,835
$8,477,122
$4,681,053
$8,702,340
$1,299,000,000
0.36%
0.67%
Logging only by small entities
in designated areas
Low estimate
$112,424
$2,799,093
$5,325,208
$1,866,995
$3,551,914
$1,979,418
$3,664,338
$543,862,000
0.36%
. 0.67%
High estimate
$224,542
$5,598,185
$10,650,417
$3,733,990
$7,103,828
$3,958,532
$7,328,370
$1,088,562,000
0.36%
0.67%
This Exhibit reflects the following data, assumptions and procedures:

•      Administrative costs to loggers are those estimated in Subsection 6. We assume, as a worst case,
       that all administrative costs will be paid by loggers and none will be paid by land owners.  Small
       logging firms constitute 99.7 % of all logging firms nationally (Exhibit II - 6), and we assume
       that they will account for this same percentage of the administrative costs incurred by loggers in
        55 These impacts will be incurred by the average firms that conduct logging only within designated areas.
To the extent that a firm conducts some logging outside of areas that are designated, this firm will earn revenues but
incur no compliance costs from this outside logging, thereby reducing the impact of designation on this firm's
revenues below the 0.36 - 0.67 % level.
78

-------
        the designated areas;

•       Total BMP costs are taken from Exhibit II - 16.  We assume, as noted previously in .this
        subsection, that two-thirds of total BMP costs are incurred by loggers and one-third are incurred
        by timber land owners.

•       Small logging firms account for 83.8 % of logging industry revenues nationally (Exhibit II - 6).
        We assume that small logging firms will similarly account for 83.8 % of BMP costs and logging
        receipts in the designated areas.

•       The final estimate of costs as a percent of revenues represents the summed BMP costs and
        administrative costs for small logging firms relative to their  logging receipts. The costs have
        been scaled down by the 0.631745 factor to reflect gradual designation over 16 years. Note that,
        in contrast, logging receipts have not been scaled down, as the small logging firms will earn the
        same amount of annual revenues in 2000 and all subsequent years.

        Impacts on small timber land owners

        We estimate that there may be up to 9,436 - 18,871 owners of timber land that could be affected
by use of the proposed designation provision (after all designation is completed, in 2015). Roughly one-
third of the costs of the additional BMPs to be required of designated and permitted silvicultural
operations might fall on these owners.  These owners range from large forest products companies with
large timber land holdings to small farmers, ranchers, second home owners and others with small tracts.
Economic data on these timber land owners and data on their timber land holdings are extremely limited.
We cannot estimate the exact proportion of these timber land owners that are small entities, though it is
undoubtedly a majority. The vast majority of all farms and ranches are classed as small entities, and
more than 91 % of the firms in the timber land-owning SICs are small businesses (see Exhibit II - 7).  It is
difficult to analyze the impacts of designation on these small entities that own timber land.

        A timber land owner receives stumpage fees when a logger cuts and sells timber from the
owner's property.  For every dollar that the logger realizes in selling the timber to a mill, the stumpage
fee paid to the land owner might represent anywhere from about 20 - 80 cents, depending on the value of
the tree species, the difficulty of logging conditions, and other factors.  RTI presents modeled data
showing the landowner's stumpage revenue amounting to about 32 % of the logger's mill revenue in
northern States, and 50 % in southern States. With this and other information, we can estimate the
typical impact of the costs associated with designation on a typical timber land owner's revenues from
selling his timber:

•       We have just estimated that the compliance costs bome by loggers amount to about 0.36 - 0.67 %
        of loggers' mill revenues.

•       The compliance costs borne by land owners are approximately one-third of the compliance costs
        borne by loggers.56 Hence land owners bear compliance costs equal to roughly 0.12 - 0.22 %  of
        56 See Exhibit II - 15. Two-thirds of the BMP costs are assumed to fall on loggers, and one-third fall on
land owners. We assume that loggers will tend to bear the administrative costs of designation, not land owners.
However, since administrative costs are relatively small, we will farther assume that two-thirds of total compliance
                                                                                             79

-------
        loggers' mill revenues.

•       About one third of the U.S. timber harvest occurs in northern States, while two thirds occurs in
        southern States. Weighting stumpage revenue (32% of mill revenue in northern States and 50%
        in southern states) by the proportion of timbering occurring in northern and southern states yields
        an average landowner's stumpage revenue equal to 44% of the logger's mill revenue.

•       Therefore, the compliance costs borne by land owners will equal about 0.27 - 0.50 % of the land
        owners' timber revenues (0.12 % divided by 44 % as a lower bound and 0.22 % divided by 44 %
        as an upper bound).

We have no reason to believe that these cost impacts as a percentage of timber sales revenues would be
different for small entity timber land owners than for large entities.  The cost impacts occur on a per acre
basis — a small entity owner with a small  timber tract will incur lower BMP costs and will receive lower
stumpage revenues than a large entity owner with a large tract, but the percentage impacts will be
identical in each case. We therefore estimate that small entity timber land owners whose land is
designated will incur a cost burden from designation equal to approximately 0.27 - 0.50 % of their timber
revenues.

        This projected percentage impact of designation on small entity timber land owners  is likely an
overestimate for most such owners for two reasons:

        1.      This percentage impact results if all of the land owner's timber land is designated. Many
               land owners, however, will have some timber land within a designated area and some
               that is not. Any timber land that is not within the designated area will provide revenues
               to the land owner without incurring compliance costs, thus diluting the aggregate impact
               of designation on the land owner's timber revenues below 0.27 - 0.50 %,

        2.      Similarly, the great majority of the small entity timber land owners have a primary
               business and source of income that is other than owning timber land — for example,
               fanners, ranchers, homeowners, etc. These entities will have other income beyond that
               from timber sales, and the compliance burden will represent a share of their total income
               typically much less than 0.27 - 0.50 %.

Consider, for example, a small farm with  a 20 acre wooded area. Assume the farmer contracts to have
this plot logged every 25 years. The farmer's  stumpage revenue might be 1/3 to V* of the $32,660 the
logger will realize from a mill for the timber cut from  the farmer's land, or $10,900 - $16,300.57  On an
annualized basis (straight-line, over 25 years), the farmer's timber sales revenue would be $436 -
$652/yr, while his compliance costs resulting from designation would be 0.27 - 0.50% of this, or $1.18 -
$3.26 annually. In addition to these timber revenues, however, this small farmer undoubtedly also has
farming income. The median farm having 20  acres of non-pastured woodland has 50 - 69 acres of total
costs will fall on loggers and the remaining one-third on owners.

        57  Exhibit II - 11 shows that on designated land, logging receipts will average about SI,633 per acre cut.
Multiply this figure by the 20 acres of the farmer's land that is cut.

80

-------
land, and average annual revenues from sale of agricultural products of $39,623.58 While the compliance
costs from designation might constitute 0.27 - 0.50 % of this typical fanner's timber revenues, these
compliance costs would constitute much less than 0.1 % of this farmer's total revenues (i.e., $1.18 -
$3.26 in annualized compliance costs relative to roughly $40,000 in annual total income).

       In sum, we estimate that the proposed silvicultural designation authority may have the following
impacts on small entities:

•      Small logging firms will incur annualized compliance costs equal to 0.36 % to 0.67 % of their
       annual revenues. Roughly 368 - 735 small logging firms may be affected.

•      Small entity timber owners will incur annualized compliance costs of 0.27 - 0.50 % of their
       timber revenues. The great majority of these small entities have other sources of income that are
       larger than their timber revenues, and compliance costs will constitute a much smaller percentage
       of their total revenues. Several thousand small entity timber owners, up to a maximum of
       perhaps 18,000 such owners, might be affected to this minimal degree.

We emphasize that these, estimates of small entity impacts are likely overestimates for several reasons:

•      We used States' much longer 305(b) lists to  indicate silviculture-impaired water bodies rather
       than their shorter and more carefully chosen 303(d) lists.

•      We assumed conservatively that all States not reporting silviculture-impaired water bodies on
       their 305(b) lists actually have such waters, but simply did not report them. If some non-
       reporting States in fact have no such waters, we again overstated the number of silviculture-
       impaired waters for which TMDLs will need to be developed

•      We made a worst case assumption in using RTT's unit costs for silvicultural BMPs that all
       silviculture activities subject to designation will occur in "wet" forest management units.

•      We assumed that logging firms and timber owners have revenues deriving only from silviculture
       activities and have revenues only from designated geographic areas. To the extent that affected
       entities have additional revenues, the additional revenues are not subject to compliance costs
       from designation and they will serve to dilute the  impact of designation below the levels cited.

SUMMARY OF IMPACTS OF PROPOSED SILVICULTURE PROVISIONS

       Exhibit 11-19 summarizes the estimated impacts  of the proposed silviculture provisions. These.
estimates are uncertain primarily because the provisions authorize EPA and State program Directors to
take actions, but do not mandate specific actions. A large number of assumptions have  been made  in
order to develop low and high estimates of how often the proposed designation authority is likely to be
used. For the most part, we believe these assumptions have been conservative, and the estimated impacts
shown below are likely overestimates.
        58 Source: U.S. Department of Agriculture, National Agricultural Statistics Service. 1997 Census of
Agriculture. Table 49.
                                                                                             81

-------
                                    Exhibit n-19
               Summary Estimated Impacts of Proposed Silviculture Provisions

BMP costs to sources (million $/yr)
Administrative costs to sources (million $/yr)
Administrative costs to EPA/States (million $/yr)
Total costs of provisions (million $/yr)
Low Estimate
S3.340 - $6.354
$.113
$.267
$3.72 - 56.73
High Estimate
$6.680 - $12.709
$.225
$.282
$7.19 -$13.22

# entities affected (after all designation completed)
logging firms
timber land owners
# entities affected annually
logging firms
timber land owners
369
9,436
23
590
737
18,871
46
1,179

Small entities affected (after all designation completed)
logging firms
timber land owners
Compliance cost as % of revenues for small entities
logging firms
timber land owners
368
up to 9,000 or so
0.36 % - 0.67 %
0.27 % - 0.50 %
735
up to 1 8,000 or so
0.36 % - 0.67 %
0.27 % - 0.50 %
82

-------
       HI.  PROPOSED AUTHORITIES REGARDING ANIMAL FEEDING
                                     OPERATIONS

PROPOSED PROVISIONS ALLOWING EPA TO DESIGNATE SOME ANIMAL FEEDING
OPERATIONS (AFOs) AS CONCENTRATED ANIMAL FEEDING OPERATIONS (CAFOs), THUS
MAKING THEM SUBJECT TO NPDES

       Under existing NPDES regulations, some AFOs59 are defined as CAFOs and hence point sources
subject to NPDES requirements. An AFO is necessarily a CAFO if it has:

•      more than 1,000 animal units [40 CFR 122, Appendix B(a)]; or

•      between 301 and 1,000 animal units and it may or does either 1) discharge pollutants into waters
       of the U.S. through a man-made ditch, flushing system or other similar man-made device; or 2)
       discharge pollutants into waters that originate outside of and pass over, across, or through the
       facility or otherwise come into direct contact with the animals.

In addition, under existing regulations, the NPDES Permitting Authority (the State Director in authorized
NPDES program States, and the EPA Regional Administrator in non-authorized States), may designate
any other AFO as a CAFO on a case-by-case basis if the permitting authority determines that the AFO is
a significant contributor of pollutants to waters of the U.S..  In practice, the types of AFOs that can be
designated as CAFOs fall into three general categories:

•      AFOs with between 301 and 1,000 animal units that do not meet either of the two "method of
       discharge" criteria listed above. If an AFO in this size range does meet either of the two criteria
       it is automatically a CAFO. If it meets neither of the criteria it can be designated as a CAFO.

•      AFOs with fewer than 301 animal units if the facility does meet either of the "method of
       discharge" criteria.

•      AFOs that raise animals other than the species for which animal unit definitions or quantity
       limits are specified in 40 CFR 122, Appendix B.  The most important example of such AFOs
       include those raising chickens (layers or broilers) when the facility does not have a liquid manure
       handling system (i.e., when the facility effectively handles its waste as dry manure). In addition,
       such AFOs may include more exotic animal operations, including mink, ostriches, llamas, etc..
       These operations may be designated as CAFOs regardless of the manner of conveyance of any
       discharge.

In addition, a NPDES permit application cannot be required of an AFO that has been designated as a
       59 Not all facilities (e.g., farms, livestock markets) at which animals are kept are AFOs. In general, to be
an AFO, the facility must: 1) Stable, confine and feed or maintain animals for at least 45 days in a 12-month period;
and 2) Not sustain crops, forage or post-harvest residues in the area where the animals are kept. The second
criterion is intended generally to distinguish feedlots from pasture land, which is not subject to the NPDES program.
The great majority of beef cattle are kept on pasture or open range land rather than in AFOs, while the reverse is
true for dairy cattle, swine and poultry.

                                                                                        83

-------
CAFO until the Director has conducted an on-site inspection of the facility and determined that it should
and could be regulated under NPDES.

       Under existing regulations, in authorized NPDES States, only the State Program Director has this
case-by-case designation authority. EPA now has no such designation authority in authorized States.

       The proposed regulation would extend the designation authority in authorized States to the EPA
Regional Administrator also. EPA's use of the designation authority would be limited to only those
situations where EPA is establishing a TMDL for 303(d) listed waters. It is intended that EPA will use
the designation authority only in instances where a 303(d) listed water body is impaired in part by AFOs.
and where for some reason the State has not developed an approvable (e.g., including reasonable
assurance for implementation) TMDL for the water body. When EPA then develops a TMDL for the
water body, EPA would have designation authority over the AFOs affecting the water body as a means of
enforceably achieving whatever reduction in AFO loadings is necessary to meet water quality standards.
EPA's use of this authority is expected to be infrequent, for several reasons:

•      States are expected to develop the great majority of TMDLs for their waters. Instances where
       EPA must establish part or all of a TMDL in an authorized State have been substantially less
       than 20 % of all such TMDLs that have been established thus far;

•      Under Federal NPDES regulations, an authorized State already has this case-by-case CAFO
       designation authority. The proposed extension of this authority to EPA serves only as a backup,
       to ensure that authority over AFOs is available to EPA when, for some reason, a State does not
       choose to exercise its authority.

This "backup" nature of the proposed designation authority over AFOs is quite different from the
designation authority over some silvicultural activities proposed elsewhere in this regulation. The
silviculture provision extends to the States and EPA authority  over these sources for the first time,
whereas the AFO provision only extends to EPA an authority that the States already have.

OVERVIEW OF COST IMPLICATIONS

       One might argue as follows that the proposed designation provision imposes no incremental costs
beyond what could be expected in the baseline. 1) The States already have this designation authority
under existing NPDES regulations; 2) States are obligated to do what is necessaryto achieve effective,
approvable TMDLs; 3) Therefore now giving this authority also to EPA simply gives EPA the ability to
ensure what the States are already required and able to do, and hence the provision adds no incremental
cost beyond what is already required.

       In order to be conservative in our regulatory cost estimates, however, we will reject this baseline
argument.  We will estimate the costs that may be incurred when EPA invokes the proposed authority
and designates AFOs as CAFOs, independent of whether some or all of these costs would have been
incurred in the baseline nevertheless. The remainder of this section estimates these costs.

       The costs estimated in this section represent indirect rather than direct costs attributable to the
proposed AFO designation provision. The provision does not require that any AFOs be designated, nor
does it directly require any AFOs to implement any new practices. The provision only authorizes EPA to


84

-------
designate AFOs as CAFOs under certain circumstances. The circumstances may not arise, and even if
they do arise, EPA may determine that it is not necessary to exercise the designation authority. Costs
will be incurred only if and when EPA invokes this authority.  This section presents some relatively
uncertain estimates of how often EPA may invoke the proposed designation authority, and, if so, the
costs that will likely ensue.

        Before beginning to estimate costs, we need to describe more fully the locations and categories
of AFOs that would be subject to designation under this proposed authority.

        First, this designation authority only applies in those States that are authorized to conduct the
NPDES program ("authorized" States). When this analysis was conducted, there were 41 authorized
States.60 The non-authorized States included AK, AZ, ID, ME, MA, NH, NM, TX and OK (for AFOs
specifically).  In non-authorized States, EPA already has the designation authority and the proposed
regulation adds nothing further.  Subsequent to completion of much of this cost analysis, two more States
TX and OK) were authorized to conduct the relevant portions of the NPDES program, and now there are
only 7 non-authorized States. There has not been sufficient time to revise the analysis to reflect the
increase from 41 to 43 in the number of States in which the proposed designation authority will be
applicable. The costs estimated for 41 States in this report would likely increase by roughly  5 % if
authorization for Texas and Oklahoma were to be reflected.

        Second, designation depends on the number and type of animals maintained at the facility and on
    method of discharge. To refer to the three general categories of AFOs that can be designated as
CAFOs, we have chosen the following terminology:61-62

•       Class A. AFOs with between 301 and 1,000 animal units. Class A AFOs that do not meet either
        of the two method of discharge criteria are subject to designation under the proposal. Class A
        AFOs that do meet either of the method of discharge criteria are already AFOs, and hence not
        subject to designation under the proposal.

•       Class B. AFOs with 300 or fewer animal units.  Class B AFOs that do meet either of the method
        of discharge criteria are subject to designation under the proposal.

•       Class C. AFOs that raise animals other than the species for which animal unit definitions or
        quantity limits are specified in 40 CFR 122, Appendix B. Species for which definitions or limits
        are provided include: beef and dairy cattle, swine, horses, sheep or lambs, turkeys, ducks, and
        chickens where the poultry operation uses a "liquid manure system". AFOs raising species other
        than these may be designated as CAFOs regardless of size and the manner of conveyance of any
        60 Information on State programs and NPDES authorization relevant to permitting AFOs was drawn from:
Science Applications International Corp., State Compendium: Programs and Regulatory Activities Related to
Animal Feeding Operations. Interim final report, February 10, 1999.

        61 Also, in order to be designated, the AFO must actually or potentially be a significant contributor of
pollutants to waters of the U.S.

        62 This terminology is not regulatory language; it is only our shorthand means of referring to the three
different groups of AFOs.
                                                                                             85

-------
       discharge.

EPA estimates that approximately 5 % of all AFOs, regardless of size and type, have conditions meeting
the "method of discharge" criteria.

       Liquid Manure Systems

       If a poultry AFO has a liquid manure system, the animal unit ranges in 40 CFR Part 122,
Appendix B apply, and the poultry AFO will be either already a CAFO (more than 30,000 laying hens or
broilers), a Class A AFO (9,000 - 30,000 laying hens or broilers) or a Class B AFO (< 9,000 laying hens
or broilers).  For a poultry AFO with a liquid manure system ~ as for beef, dairy, swine, etc. operations —
its method of discharge thus affects whether or not it is subject to designation. However, a poultry AFO
that does not have a liquid manure system is not covered by the animal unit ranges in 40 CFR Part 122,
Appendix B, and it will neither be already a CAFO, nor a Class A nor B AFO. Instead, a poultry AFO
without a liquid manure system is in Class C, along with other AFOs that raise other sorts of animals for
which no animal unit size categories have been established. All Class  C AFOs are subject to designation,
whatever their size or method of discharge. None are already CAFOs. A Class C AFO can be a CAFO
only through designation.

       In the cost analysis to follow, we assume that half of all layer and broiler operations have liquid
manure systems. Some layer operations  (probably less than half) use a water washing method to remove
manure from the poultry house. The remainder of the layer operations and nearly all broiler operations
remove litter from poultry houses without using water, and subsequently manage the litter as a dry waste
(e.g., spread it as a solid with a spinner spreader rather than as a slurry with a tanker truck or irrigation
guns).  However, the dry litter is frequently stored for some time before being spread, and only perhaps
half of the poultry farms have a covered storage facility that keeps the stored litter from being rained on.
EPA believes that an AFO that removes waste from a pen or a house and stacks it in areas exposed to
rainfall or an adjacent watercourse may have established a crude liquid manure system for process
wastewater that may discharge pollutants.63  We thus assume that roughly half of all poultry farms (those
removing waste from the house by water washing, and those without covered litter storage structures)
have liquid systems and are covered by 40 CFR Part 122, Appendix B. Half of the poultry layer and
broiler AFOs are in Class C, and half are distributed among CAFOs, Class A, and Class B.

SUMMARY OF TYPES OF COSTS

       In this section, we will estimate the following costs indirectly attributable to the proposed AFO
designation provisions:

       Compliance costs for the AFOs that will be designated, permitted and required to change their
       practices in various ways that will protect water quality. These compliance costs will include: 1)
       The additional capital and operating costs of the measures these operations will be required to
       implement; and 2) The administrative costs of obtaining and maintaining coverage under NPDES
       permits.
        63 U.S. EPA, Office of Water. Guide Manual on NPDES Regulations for Concentrated Animal Feeding
Operations. December, 1995. Page 6.

86

-------
•      Administrative costs to EPA for designating these operations, and to EPA and the States for
       permitting the designated operations. (Note that even if it is EPA that designates an AFO as a
       CAFO, it may be the State that ultimately develops the NPDES permit for the designated CAFO.)

In addition to developing national cost estimates for all AFOs that will be designated, we will also
provide a rough estimate of the regulatory costs and impacts specifically for the small entity AFOs that
will be designated.

OVERVIEW OF PROCEDURES FOR ESTIMATING COSTS

       We will estimate the compliance and administrative costs of the proposed AFO designation
authority through a sequence often steps:

1.     Determine the number, location, size and characteristics of AFOs in the U.S.. This will provide
       the data base from which to begin the analysis.

2.     Focus the analysis on only those "authorized" States that manage the NPDES program
       themselves, including permitting CAFOs.  In non-authorized  States, the EPA Regional
       Administrator already has designation authority, and the proposed regulation has no effect.

3.     Determine the locations where AFOs are likely to need  further controls in order for water quality
       standards to be met. This will involve locating the water bodies impaired by AFOs.

4.     Determine the number, size and characteristics of the AFOs that are likely to need further
       controls in these areas and for which the designation authority could provide the means of
       requiring the further controls. This will  involve extracting from the national data base on AFOs
       (#1) information on the specific AFOs in the geographic areas identified in (#3). We will further
       consider only those AFOs that are subject to designation as CAFOs, excluding those that are
       already CAFOs and those that cannot be reached through the designation authority because of the
       "method of discharge"  limitations.

5.     Determine the performance standards that designated and permitted AFOs will typically be
       required to meet, the management measures they will respond with, and the per unit costs for
       these measures.

6.     Estimate the maximum possible total national costs for these management measures by
       multiplying the number of AFOs that could be designated as  CAFOs (#4) by the estimated unit
       costs for the measures (#5).

7.     Estimate the frequency with which EPA will likely need to invoke the proposed Federal
       designation authority, and the management measure costs attributable to the proposed regulation
       that will result from these designations.  The maximum possible costs resulting from designation
       (#6) will be scaled down to reflect the relatively infrequent instances in which the further
       controls on AFOs needed to meet WQS will be obtained by EPA exercising the proposed Federal
       authority. Most often,  States will use their own authorities (either independent State authorities,
       or the State designation authority under existing NPDES regulations) to mandate the needed
       controls over AFOs. We will assume that  EPA will need to use the proposed Federal designation


                                                                                            87

-------
9.
       authority very rarely in States that now have strong AFO programs and authorities, and more
       often in States that have weaker programs.

8.     Estimate the administrative costs for designated AFOs to obtain and maintain coverage under
       NPDES permits. This will be estimated by multiplying the number of operations likely to be
       designated (#7) by an estimate of the administrative cost for a typical AFO.

       Estimate the administrative costs to Federal and State governments for designating and
       permitting these AFOs. Assuming that permitting will be accomplished by general rather than
       individual permits, these costs will be estimated by multiplying the number of general permits to
       be developed (assuming one general permit for each geographic area where designation may
       occur — #3) by an estimate of the administrative cost for developing and implementing a typical
       general permit.

10.    Develop estimates of compliance costs and administrative costs for small entities specifically.
       This will be done by disaggregating the estimated costs for all entities developed under #7, #8,
       and #9.

Most of this section will consist of describing the data, calculations, results and uncertainties in each of
these ten steps.

KEY ASSUMPTIONS IN ESTIMATING COSTS

       Estimating the costs of the proposed Federal designation provision is particularly difficult
because the provision authorizes EPA to act, but does not mandate specific actions. No duties are
established for EPA, States, or AFOs.  There is no certain answer to fundamental questions about what
will happen if the proposed regulations are finalized: What will the States do in developing TMDLs for
AFO-impaired waters?  How often will EPA need to step  in to accomplish what the States will not have
done? What permit conditions will EPA or the States impose on designated AFOs?

       As the starting point for estimating the costs of the proposed provisions, we assume that the
designation authority will be implemented as follows:

•      The designation authority will be invoked bv EPA only when EPA must develop a TMDL in an
       authorized State for a SOSCd") water bodv that is impaired or threatened by AFOs. This limitation
       is explicit in the proposed regulation.

       In the great majority of instances, the State will develop an effective, approvable TMDL for
       303(d1 waters impaired bv AFOs. and EPA will not need to use the proposed Federal designation
       authority. The TMDL program is intended to be implemented by the States, with EPA
       developingxTMDLs only in cases of State non-performance.  All States have authorities
       sufficient to address AFOs as necessary for TMDLs - most States have existing authorities and
       programs under State law, and all authorized NPDES States have the further option of using the
       NPDES designation authority  available under current Federal regulations. EPA will use the
       proposed Federal designation authority only when an authorized State chooses for some reason
       not to perform the TMDL tasks that the State is normally  able and expected to perform.  Some of
       the likely infrequent reasons why a State might not develop a TMDL and EPA would might


88

-------
       include: a resource shortfall in the State TMDL program; lack of State authority over AFOs and
       unwillingness by the State to use the NPDES designation authority available under Federal
       regulations; State unwillingness to take on the political issues attendant to a controversial
       TMDL; or perhaps some intractable policy difference between the State and EPA in which EPA
       contends that a State-developed TMDL is inadequate. Based on the judgment of EPA TMDL
       program officials, we will assume that EPA will need to establish a TMDL:

               For 1 - 5 % of the AFO-related TMDLs to be developed in States that have stronger AFO
               authorities and programs; and

               For 5 - 25 % of the AFO-related TMDLs in States that have weaker AFO authorities and
               programs.

       In these instances in authorized States where EPA will establish an AFO-related TMDL, we
       assume that EPA will use the proposed designation authority as the means of providing
       reasonable assurance for implementation of the AFO-related provisions of the TMDL. EPA
       currently has no alternative applicable and enforceable authorities over AFOs.  We further
       assume that EPA will be less likely to designate and permit smaller (Class B) AFOs than larger
       (Class A and C) ones.

       In a geographic area where EPA must develop the TMDL and will use the proposed designation
       authority, we assume that EPA will not consider designating as point sources subject to NPDES
       any AFOs in that area that are smaller than some de minimis size. Designating and permitting
       AFOs will be burdensome for governments (the permit authority must conduct a site visit to each
       designated AFO before requiring a permit application) and for the AFOs themselves. We  assume
       that EPA will not wish to incur and impose these costs for numerous very small AFOs, which in
       total may be responsible for a small fraction of the AFO-related environmental problem. Our
       definition of very small AFOs and our assumption that further controls will not be required for
       them matches the assumptions made in three other recent EPA cost analyses for AFOs.64

       Permitting will be achieved through a general permit covering all the designated non-de minimis
       AFOs in the area, not individual permits for individual AFOs. Again, this assumption is a matter
       of practicality. It is likely that all the  designated AFOs in the geographic area affecting the
       impaired water body will need to implement roughly similar control measures. The sources and
       desired controls will probably be sufficiently similar to •warrant treatment via a general permit. A
       general permit approach will be much less costly for EPA, States and AFOs than individual
       permits.

       The general permit will require all the designated AFOs in the specified geographic area to meet
       appropriate  performance standards. We expect that the standards to be met will be relatively
       consistent across the country. Judging from the existing EPA effluent guidelines for CAFOs and
       EPA's set of chosen CZARA and national nonpoint source management measures, we assume
        64 DPRA Incorporated. Economic Impact Analysis of Coastal Zone Management Measures Affecting
Confined Animal Facilities. October 7, 1992. Also, DPRA Incorporated. Economic Impact Analysis of National
Nonpoint Source Management Measures Affecting Confined Animal Facilities. May 17, 1995. Also, U.S. EPA.
1996 Clean Water Needs Survey Report to Congress, nonpoint source cost estimates,  www.epa.gov/own/npsneeds.

                                                                                            89

-------
        that the performance standards will likely require: 1) no discharge from waste management
        facilities except in the event of a 25-year/24-hour storm; and 2) conformity with broad manure
        application BMPs.  The management measures that designated and permitted APOs will adopt in
        response can be expected to be tailored somewhat according to farm and animal type, waste
        handling system, rainfall characteristics, etc., but there will likely be enough similarity across the
        country to allow us to pursue a unit costing approach (e.g., for all AFOs of type "X", meeting the
        general permit performance standards will cost roughly $Y per animal unit).

ESTIMATED COSTS OF THE PROPOSED AFO DESIGNATION PROVISION

        In this subsection, we describe the ten steps resulting in estimates for the different varieties of
costs attributable to the proposed regulation.

1. Determine the number, location, size and characteristics of AFOs in the U.S.

        The Census of Agriculture, conducted every five years, provides the most comprehensive data
available on AFOs. The most recent edition of the Census, having been made public only recently,
provides data for 1997.  The Census provides data at the national, State and county level, including the
following information that we used for this analysis:

•       The number of farms at which animals are kept,  for each of:  beef cows, dairy cows, swine, layers
        and broilers;65

•       The annual receipts by these farms (this information on the distribution of income for different
        size AFOs will be used in the analysis of impacts on small businesses);

•       The size distribution of these farms with respect to the inventory of these animals at the farm as
        of December 31,1997.  Inventory information is presented in ranges, so that, for example, data
        are available on the number of farms with 1-24 swine, with 25-49 swine, etc..

The Census thus gave us directly most of the information we needed on the number, location (by county),
size and revenues of AFOs. However, several assumptions or adjustments were necessary in adapting the.
Census information exactly to our needs:

•       We sought to analyze AFOs, while the Census provides information on farms. Some farms
        confine animals in a manner so as to meet the AFO definition, while many farms with animals
        keep them in pasture or range settings instead. We made the following assumptions:
        65 The Census of Agriculture also provides some, but not complete, information on other animal species:
turkeys, sheep, goats, horses, ducks, mink, etc.. For a variety of reasons, including incomplete county-level
inventory information, no information on manure management costs for farms raising these species, and a desire to
keep the amount of data collection and analysis to a manageable level, we decided to analyze in this study only
AFOs raising beef and dairy cows, swine, layers and broilers. This set of animals that we analyzed accounts for
nearly 95 % of the volume of manure produced annually at U.S. livestock and poultry farms.  (Source: U.S. EPA.
Preliminary Study of the Livestock and Poultry Industry: Section!, Economics. September 30, 1998.) We are
confident that omitting the numerous animal types other than the five we have analyzed results in only a very small
underestimate of the costs of the proposed regulation.

90

-------
               AFOs that confine beef are represented by the Census category "Cattle and calves
               fattened on grain and concentrates for slaughter". This category should generally
               include beef feedlots and exclude pasture and range operations.

               All Census milk cow, swine, layer and broiler farms with more than a very small number
               of such animals will meet the AFO definition.

•       We assumed that the animal inventory information at a farm as of December 31 was generally
        representative of the number of animals at the farm throughout most of the year. (This number is
        used to determine the approximate size of the manure management facilities necessary at the
        farm.) In fact, the number of animals at farms throughout the year may differ systematically
        from the number as of December 31 as a function of seasonal breeding and animal sales patterns.

•       For broilers and other meat chickens, the Census provided information on the number of animals
        sold over the course of the year rather than on the inventory on hand at the end of the year. We
        converted the  sales information to approximate inventory by dividing sales by 5.5, the
        approximate number of broiler flocks grown at a typical broiler farm in a year.

•       The Census does not provide complete inventory information for multiple animal species present
        simultaneously at a single farm. We were thus unable to identify AFOs that might be CAFOs
        because of multiple animal species even though  no single species was present in sufficient
        number to exceed the CAFO threshold. For example, we could identify as a CAFO a feedlot that
        exceeded 1,000 beef cows or 2,500 swine, but we could not identify as a CAFO a farm that had,
        say, 800 beef cows and 2,000 swine, which  together would exceed 1,000 animal units. As a
        result, we probably slightly undercount the number of farms that are already CAFOs and
        overcount the  number that are Class A or B AFOs that could be subject to designation. This
        probably results in a very slight overestimate of the potential costs of the proposed regulation.

2. Focus the analysis on only those "authorized" States that manage the NPDES program
themselves, including permitting CAFOs.

        In non-authorized States, the EPA Regional Administrator already has designation authority. In
these States, the proposed regulation will add no incremental authority and will have no impact.  The
non-authorized States  where EPA operates the NPDES permit program include: Alaska, Arizona, Idaho,
Maine, Massachusetts, New Hampshire, New Mexico, Texas and Oklahoma (for AFOs specifically).66
These States were dropped from further analysis.

3. Determine the locations where AFOs are likely to  need further controls in order for water
quality standards to be met.

        This involved locating  the water bodies impaired by AFOs that will need to have TMDLs
developed for them. We then assumed (in the absence of any more precise method for identifying the
        66 Again, this analysis does not reflect recent authorization for Texas and Oklahoma to conduct the
relevant portions of the NPDES program. Costs estimated in this analysis might increase by roughly 5 % to reflect
the addition of these States to the set of States in which the proposed Federal designation authority could be used.

                                                                                            91

-------
specific AFOs that will need further controls) that all non-de minimis AFOs in the county(s) in which an
AFO-impaired water body is located would need further controls in order to achieve WQS. Most of the
identified AFO-impaired water bodies are relatively small (typical length of the impaired river segments
was 5 to25 miles), and our assumption that the entire surrounding county's worth of AFOs would need to
be controlled will usually result in a substantial overestimate of compliance costs. However, some water
bodies are larger and seriously impaired; for them it is possible that more than the surrounding county's
worth of AFOs would need to be controlled.

        We began with a search of the 1996 and 1998 data bases of 303(d) waters for water bodies with
animal feeding operations cited as the cause of impairment (source names used by States that were
searched for in this investigation included "animal holding/management area", "feedlots [confined
animal feeding operations, non-point source]", "farm animals", and "animal operations"). The search
was performed by Tetra Tech,  Inc.  Such waters were reported for only 5 States  in the partly completed
1998 data base, and 12 States in the 1996 data base. We judged that coverage in these 303(d) data bases
was substantially incomplete — some States are not included in the data bases at all, and many States
provide information on the source types responsible for impairment of the listed waters at too high a level
of aggregation to allow identification of AFOs (e.g., many States cite impairment as due generally to
"agricultural nonpoint sources" without further information that allows us to determine whether the cause
more specifically might be row crops or grazing or AFOs, for example.)

        Much more complete and detailed information on sources of impairment is available from State
reporting under section 305(b) than from reporting for 303(d). In order to identify and locate water
bodies that were  impaired specifically by AFOs, we consulted EPA's 1996 National Assessment
Database. The database was created to provide a uniform and consistent method for States to report
water quality information in their 305(b) reports.  In this database, 28 States reported having water bodies
that were impaired by sources  related to AFOs. These sources were listed under the following source
sub-categories: 1) intensive animal feeding operations; 2) confined animal feeding operations (non-point
source); 3) animal holding area; and 4) manure lagoons. Based on the much greater number of States
reporting water bodies impaired by AFOs in the 305(b) data base relative to the 303(d) data base, we
decided to use the National Assessment Database as our point of departure.

        For each water body listed as impaired by AFOs, the 1996 National Assessment Database
contains a data field indicating the source magnitude — the extent to which AFOs contributed to the
impairment of the water body. For each affected water body, AFOs may be listed as a "major",
"moderate", or "minor" source of impairment. We decided to analyze only water bodies for which AFOs
constitute a major or moderate source of impairment. Water bodies  where AFOs are only a minor source
of impairment are impacted more significantly by one or several other sources of impairment. For these
water bodies, addressing the AFO contribution to impairment would not fully eliminate the impairment,
and would likely be of much lower priority in a TMDL than addressing the major or moderate sources of
impairment. For these waters, we assumed that AFOs (if they need to be addressed at all) are likely to be
addressed by means that are less oner&us than designating them as point sources; perhaps they will be
addressed sufficiently through voluntary or cost share programs alone.  Accordingly, we assumed that the
proposed designation provision is unlikely to be used in connection with these "minor AFO impairment"
water bodies, and excluded these water bodies from our initial list of affected water bodies.

        A total of 931 water bodies in 28 States were listed as having major or moderate impairments due
to AFOs. We then excluded water bodies in the non-authorized States (New Mexico, Oklahoma, and
Texas) among the 28.  This left us with a list of 25 authorized States that reported a total of 879 AFO-

92

-------
impaired waters, for which the proposed regulation might have some impact.

       For each of these 879 water bodies, we wanted to identify the AFOs that were responsible for
their impairment and that would need to be controlled in order to meet WQS. There were two significant
difficulties in doing so, however.  First, the finest level of geographic resolution at which AFO activity
data (see Step #1) is reported is the county level.  Second, the finest level of geographic resolution at
which we were able to locate identify individual water bodies in the time available was similarly at the
county level.67  With more time, we might be able to develop a more precise location for each water body
and then delineate the boundaries of the sub-watershed within which AFO-related loadings contribute to
impairment of the water body. However, the difficulty of obtaining AFO activity data on a sub-county
level would likely still present an obstacle to pinpointing the exact amount of AFO activity that
contributes to impairment.  For this analysis, the best we could attempt to do is identify a single county
(or counties) within which the AFO-impaired water body is located, and assume that al] the AFOs within
that county (or counties) would need to be controlled in order to eliminate the impairment. We assumed
that the maximum possible use of designation authority would be to address all the AFOs located within
every county in which there is an AFO-impaired water body.

       We identified the county  in which each AFO-impaired water body is located by using 3 sources
of information. First, for each waterbody in the database, there is an associated 8-digit "HUC", or
watershed code. Using other information provided by EPA and the Research Triangle Institute, we
identified the counties covered by each watershed, thereby narrowing the range of possible counties in
which each water body might be.68  Second, we compared the set of possible counties for each water
body with results from searching the U.S. Geological Survey's National Atlas of the U.S.
(http://www-atlas.usgs.gov/scripts/start.html') forthenarte of the waterbody. If mess st^ps did not result-.
in identifying the single county within which the particular impaired water body was located, we took the
third step of consulting State maps. In cases where these three steps did not narrow down the choice of
possible counties in which a water body was located to one, we included all possible counties for that
waterbody in our initial list of counties to consider.

       Out of the 879 AFO-impaired water bodies,  we identified a single individual county associated
with each of 585 water bodies. For an additional 115 water bodies, we were able to narrow the list of
possible counties to two. For the remaining 179 water bodies, we could not narrow the list of possible
counties to less than three.  In total, we identified 436 counties as being associated with the 879 water
bodies.
        67 The data base printout we received identified the geographic location of an AFO-impaired water body
only by indicating the "HUC" (watershed code) within which the AFO-impaired water body is located and the
counties covered by the HUC.  HUCs often span 10 or more counties. Thus, the data base printout allowed us to
locate an impaired water body only to the extent of knowing that it is somewhere within a group of counties. EPA
is currently developing a database that will map the latitude and longitude of impaired water bodies, allowing the
user to identify- the location of these water bodies more precisely.

        68 William Cooter of RTI provided us with a copy of the database and performed the initial queries to
identify AFO-impaired water bodies, the watersheds in which they were located,  and the counties included in these
watersheds.

                                                                                               93

-------
                                          Exhibit m-1
   Number of Reported AFO-Impaired Water Bodies and Corresponding Counties (Authorized
                                              States)

Water bodies
Water bodies
Water bodies
in one county
in two counties
associated with 3 or more counties
Total
Water
Bodies
585
115
179
879
Number of
Corresponding
Counties
240
52 more
144 more
436
        Cases in which AFOs in three or more counties contribute to the impairment of a particular
segment of a water body are probably extremely rare. Most AFO-impaired water bodies are small, and
they are likely affected by AFOs located within only a portion of the surrounding county — they are
unlikely to be affected by AFOs in the surrounding 3 counties, much less the surrounding (in one case)
11 counties.  We believe that inclusion of all 436 counties in our final list would lead to an
overestimation of the number of AFOs that could be affected by the provision. We therefore decided to
consider only those counties within which were located AFO-impaired water bodies that could be tied to
two or fewer counties.  Although this approach is inexact, our inclusion of counties from water bodies for
which we identified two possible counties, combined with our counting all AFOs in counties containing
affected water bodies, might offset the effect of our excluding counties from water bodies for which we
identified three or more possible counties. Using this two-or-fewer-counties-per-water-body approach,
we found a total of 292 counties in 25 authorized States associated with AFO-impaired water bodies
where AFOs were a major or moderate contributor to impairment.69
        69 While it is possible that AFOs from adjoining counties may impact a particular water body if it lies near
their shared county border or if the impaired water body crosses a county line, thereby implicating AFOs in two
counties, we believe this to be the exception rather than the rule.  Generally speaking,"we w'ould expect for the large
majority of cases that those AFOs that are actually causing impairment of a particular listed water body are located
within a single county.  If so, the actual number of counties in which AFO-impaired water bodies are located is
roughly equal to the 240 "1-county" counties, plus one-half of the 52 "2-county" counties, plus one-third or less of
the 144 "3-or-more-county" counties which had not already been counted in the other county groups.  (In fact, the
water bodies whose location could be narrowed only to being somewhere within "3-or-more-county" counties were
on average associated with 4.5 counties.) This rough estimate approach yields a total of 298 counties (240/1, plus
52/2, plus 144/4.5), and probably should be inflated somewhat to account for those exceptional cases where AFOs
in more than one county impact a given water body. As noted above, our approach of including "2-county"
counties and excluding "3-or-more-county" counties produces a roughly similar total of 292 counties. Although our
"2-or fewer-county" approach may undercount the number of "actual" counties in which AFOs are impacting water
bodies, it has the advantage of providing a better-targeted list of counties than the list we could obtain from
eliminating at random the "excess" counties associated with "2-county"  water bodies and "3-or-more-counry" water
bodies.  In addition, the effect of this undercount is most likely offset (and probably exceeded) by our decision to
consider all the AFOs located within our counties when we calculate the costs associated with the provision. This
94

-------
        In addition to the 25 authorized States reporting APO-impaired waters, there are 16 authorized
States which did not report any AFO-impaired waters in the 305(b) data base.  We do not believe this is
because there actually are no AFO-impaired waters in these 16 States.  Instead, in virtually all cases, we
believe this is because these States did not report the cause of impairment in their waters at all or did not
report causes of impairment in a manner that afforded any possibility of citing "AFOs" as a cause of
impairment.70 We believe the costs we estimate associated with use of the proposed Federal CAFO
designation authority in the 25 authorized States for which we have data should be scaled up to account
for likely costs in the additional 16 authorized States for which we do not have data on AFO-related
impairments.

        We calculated a scale-up factor to extrapolatexfrom authorized States that reported AFO-impaired
water bodies to  the authorized States that did not.  We began by obtaining data for each State on  the
State's inventory of layers, meat chickens,71 beef cows fattened on grain and concentrates, dairy cows
and swine.  Then, using animal-unit conversion factors for each type of animal,72 we calculated the total
number of animal units associated with 1) AFOs in the 25 authorized States that reported AFO-impaired
water bodies and 2) AFOs in the 16 other authorized States that did not report AFO-impaired water
bodies.  We assumed that the degree to which AFOs impair waters in these two groups of States  would
be directly proportional to the number of AFO animal units (or the amount of manure generated  by
AFOs) in the two groups of States. The  16 States without data accounted for 57 % as many AFO animal
units as the 25 States with data. We therefore used a factor of 1.57 to scale up the costs associated with
controls on  AFOs in the 25 reporting States to obtain a national cost estimate covering all 41 authorized
States.  We apply this scale-up at the end of our cost analysis.
decision probably leads to an overcount of AFOs, since it is unlikely that AFOs located at opposite ends of a county
(which may extend 50 miles or more) actually impact the same impaired water body, which is generally only 5 to 25
miles long.

        70 Many States use EPA's recommended system of "major source category codes" in reporting causes of
impaired water for their 305(b) reports, but to varying degrees. In EPA's system, "agriculture" (cause code  1000) is
a major category that is used by 38 States. More detailed AFO-related subcategories within agriculture include
"intensive AFOs" (cause code 1600), "CAFOs" (cause codes 1620, 1640 and 1650), "animal holding areas"  (cause
code 1800), and manure lagoons (cause code 1900). Only 30 States used these more detailed codes, and it was
these codes for which the 305(b) data base was searched for AFO-related impairments. Some States did not report
causes of impairment at all, some reported based on only very gross categories (e.g., point sources, nonpoint
sources), some reported at a more detailed level than EPA's system (e.g., California used a series of codes for
dairies, agricultural irrigation tailwater, agricultural subsurface drainage, etc.), and some used substantially different
classification systems.                                                                        '

        71 As noted earlier, the Census of Agriculture reports the number of broilers and other meat chickens sold
per year. We divided this figure by the average number of flocks raised by broiler farms each year (5.5) to obtain
an estimate of the inventory of the broilers at the farm at the end of the year.

        72 The conversion factors are .01 for layers and broilers (with the number of broilers sold multiplied by
0.182 to approximate inventory), 1.0 for beef cows fattened on grain ajid concentrates, 1.4 for dairy cows, and .4 for
swine.

                                                                                                 95

-------
4. Determine the number, size and characteristics of the AFOs that are likely to need further
controls and for which the designation authority could provide the means of requiring the further
controls.

       This involved two steps. First, we extracted from the Census of Agriculture (Step #1)
information on all the AFOs located in the counties within which there are AFO-impaired water bodies
(Step #3).  Second, we determined which of these AFOs would not potentially be subject to being
designated as CAFOs because they were: 1) too small to be worth the effort of designating; or 2)  so large
as to be CAFOs already; or 3) inconsistent with the "method of discharge" criteria.

       Our assumptions were as follows regarding AFOs that are too small to be worth the effort of
designating.  We assumed size cutoffs defining such very small AFOs at approximately 20 animal units
for each species. These were the same cutoffs that EPA used for its economic analyses of both coastal
zone management measures for CAFOs and national nonpoint source management measures for CAFOs.
AFOs with fewer than the  following numbers of animals were assumed to be unlikely targets for use of
the proposed Federal designation authority:

       20 beef cattle (20 animal units);

•      20 dairy cows (28 animal units; the Census provides data on dairies using a 20 cow break point,
       and provides no data on farms corresponding with 20 animal units or approximately 14 dairy
       cows);

•      50 swine (20 animal units);

•      3,200 layers (the equivalent of 32 animal units, the closest of the  Census break points to 20
       animal units for layers); and

•      60,000 broilers produced (10,900 broilers inventory, the equivalent of 109 animal units, but
       roughly the size of the smallest broiler house in common use).

EPA estimated in the Agency's  CZARA analysis that these cutoffs excluded the following fractions of
total animal inventories: beef cows - 24%; dairy cows - 2 %; swine - 9 %, and no figures were estimated
for poultry. These figures were based on 1992.data. Since 1992 the animal inventory has become much
more concentrated in larger farms, and the fraction of the 1997 animal inventory accounted for by the
very small AFOs that we exclude would be significantly less than the 1992 figures.

       We also eliminated from potential application of the proposed Federal CAFO designation
authority all those AFOs in the identified counties that were so large as to be CAFOs already. This
included all AFOs with more than the numbers of animals cited in existing regulations:
                                           s
        LOGO beef cattle;

•      700 dairy cows;

       2,500 swine;

       30,000 layers or broilers (equivalent tol65,000 annual production) at an AFO with a liquid

96

-------
       manure system (as noted, half of all poultry AFOs were assumed to have liquid manure systems).

       We thus eliminated all AFOs in the identified counties that were either too small or too large to
potentially be subject to the proposed Federal CAFO designation authority. Exhibit EI-2 shows how we
estimated from Census of Agriculture data the number of remaining AFOs in each of the three classes
that are potentially subject to designation:

•      Class A. AFOs with between 301- 1,000 animal units. They may be designated if they do not
       meet either of the "method of discharge" criteria.

•      Class B. AFOs with < 301 but more than a de minimis number of animal units.  They may be
       designated if they do meet either of the "method of discharge" criteria.

•      Class C. AFOs that raise animals other than species for which animal unit definitions or quantity
       limits are specified.  In our analysis, such AFOs include layer or broiler operations that do not
       have a liquid manure management system and that exceed the de minimis size. They may be
       designated regardless of their method of discharge.
                                                                                           97

-------
                              Exhibit III-2: Translation of Census Data on AFOs Into Classes A, B, C


Beelcatlle






Dairy cows






Swine






Layers





Broilers






Census Size Range
< 20 head
20 - 49 head
50 - 99 head
100 • 199 head
200 - 499 head
500 - 999 head
1000 + head
< 20 head
20 - 49 head
50 • 99 head
100 -199 head
200 - 499 head
500 - 999 head
1000+ head
< 50 head
100 - 199 head
200 • 499 head
500 - 999 head
1,000 -1,999 head
2.000 - 4.999 head
5,000 + head
< 3,200 head
3.200 -9.999 head
10,000-19,999
20,000 - 49,999
50,000-99,999
100,000 + head
60,000-99,999
100,000-199,999
200,000-299,999
300,000 - 499,999
500L000 - 749j999
750,000 + head
Fraction
Class A- 301-1,000 AU
0
0
0
0
2/3
1
0
0
0
0
0
1
2/5
0
0
0
0
Y,
1
1/6
0
0
0
Y,
1/6
0
0 _^_
54
1/3
0
0
0
0
of AFOs In Census Size Ranc
Class B •• < 301 AU and
> da mlnimls
0
1
1
1
1/3
0
0
0
1
1
1
0
0
0
0
1
1
'/«
0
0
0
0
y,
0
0
0
0
0
0
0
0
0
0
e That Are:
Class C •• Others (poultry
AFOs without liquid manure
jystems and > de mlnimls)
0 .
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
y,
Y,
y,
y,
Y,
y,
y,
Y,
y,
y,
y,

Notes
Too small



200 - 300 head are Class 8; remainder are A

Already CAFOs
Too small



200 dairy cows = 300 AU
500 • 700 head are Class B; larger are CAFOs
Already CAFOs
Too small


500 - 750 head are Class B; remainder are A

2,000 - 2,500 are Class A; larger are CAFOs
Already CAFOs
Too small
Liquid manure systems < 9.000 are Class 8; dry are Class C

Liquid > 30,000 already CAFOs
Liquid > 30,000 already CAFOs
Liquid > 30,000 already CAFOs

Liquid > 165,000 production already CAFOs




98

-------
        For each of the 292 counties in authorized States within which there were AFO-impaired waters,
we used Census data as shown in Exhibit III-2 to estimate the number of AFOs in these three classes and
their distribution among size categories. In the course of doing this, we made the assumption that the
distribution of AFOs by size category in the AFO-impaired counties of a State was identical to the s'ize
distribution of AFOs in the entire State. For example, in the Alabama counties with AFO-impaired
waters, there are 920 Class A, B and C broiler AFOs.  We assumed that the 920 broiler farms in these
counties are distributed across size categories in the same proportions (e.g., 10.3 % of them are 60,000 -
99,999 head in size) as the Alabama State-wide distribution of broiler AFOs by size.

        As the final step in identifying the potentially designatable AFOs, we applied the "method of
discharge" criteria. EPA estimates that 5 % of all AFOs meet the method of discharge criteria  and 95 %
do not. This estimate applies for all types of animal AFOs and for all size classes. Applying these
estimates:

•       95 % of Class A AFOs in these counties could potentially be designated.  5 % of the Class A
        AFOs meet the method of discharge criteria and hence are CAFOs already.

•       5 % of Class B AFOs in these counties could potentially be designated. 95 % of the Class B
        AFOs do not meet the method of discharge criteria and hence are not potentially subject to
        designation.

•       100 % of Class C AFOs in these counties could potentially be designated.

        Exhibit III-3 shows the State-by-State distribution of these AFOs that could potentially be
designated under the proposed regulation. In these 292 counties spread across 25 States, we estimate that
there are 25,065 Class A, B, and C AFOs. Scaled up to account for non-reporting States, there would be
nearly 40,000 AFOs. Approximately 9 % of these AFOs are Class A (301  - 1,000 animal units), 83 %
are Class B (< 301  animal  units, but larger than de minimis size), and 8 % are Class C (poultry farms
assumed not to have liquid manure  systems and that are larger than de minimis size). There are clearly
many more small, Class B AFOs (< 301 animal units) than there are medium-sized, Class A ones (301 -
1,000 animal units). However, given the infrequency with which AFOs meet the "method of discharge"
criteria, somewhat more Class A AFOs are potentially subject to designation than Class B AFOs. The
final column of Exhibit III-3 shows the total number of AFOs in each State and in the nation that are
potentially subject to designation - 95 % of the Class A AFOs plus 5 % of the Class B AFOs plus all of
the Class C AFOs. In the 25 States, 5,177 AFOs are potentially subject to designation.  Scaled up for
non-reporting States, 8,128 AFOs are potentially subject to designation.
                                                                                            99

-------
                  Exhibit m - 3. AFOs That Potentially Could Be Designated

Class A Class B Cl
farms farms ft
assC i Number of Farms that
urns i Potentially Could Be Designated
(95% of A, 5% of B. 100% of C)
i ' ;
AL
CA
DE
IA
n.
IN
KS
KY
LA
Ml
MO
MS
MT
ND
NE
OH
R]
SC
SD
TN
VA
VT
Wl
WV
WY
248 ; 346 898 1,152
65 ! 64 :
97 36 j
62 344
181 ; 1182
93 756
314 \ 2971 !
17 336
17 167
75 806
33 i 292
123 i 244
22 112
45 806
58 ! 293
157 i 1744 : . .
1 i 34 i
21 i 73
320 : 2640
54 552
7 62
65 ! 895
178 . 5781 i
12 118
42 J 161
11 | 76
321 415
1 76
1 . 233
15 '. 141 '
8 455
3 36
6 ! 31
4 115
25 70
S32 661
3 : 30
1 : 83
o ; 70
30 267
1 ': 4
27 50
5 441
28 ; 106
2 1!
1 108
11 : 469
11 29
0 ! 48
: • i
TOTAL

Scale-up
to cover
all 41
authorized
States
2305 20.814 i 1.946 > 5.177
.
3,619 | 32.677 ' 3
,056 | 8.128
5.  Determine the performance standards that designated and permitted AFOs will typically be
required to meet, the management measures they will respond with, and the per unit costs for these
controls.

       Permits written for AFOs that are designated as CAFOs will at a minimum require compliance
with the existing feedlots effluent guideline.  The guideline requires a CAFO to design, construct and
operate facilities that will contain all process wastewater plus the runoff from a 25-year, 24-hour rainfall
event. In essence, the facility shall not discharge except in the event of a 25-year, 24-hour rainstorm.
100

-------
         We might assume that permits for designated AFOs could establish any of a wide range of
 additional performance standards or require various further BMPs. State regulatory programs for
 CAFOs have developed many different sorts of requirements addressing, for example: treatment lagoons,
'retention ponds, dry" manure storage structures, equipment for irrigating with manure slurries, dead
 animal composting, filter strips, fencing, nutrient management planning, manure spreading limits,  animal
 diets, facility siting, ground water monitoring, etc..
         We decided that the best indication of what EPA might require in permits for designated AFOs is
 the set of recommended management measures for AFOs promulgated by EPA in the "Management
 Measures Guidance for Coastal Zone Nonpoint Source Pollution" required by the Coastal Zone
 Management Act Reauthorization Amendments of 1990. These management measures include:

         1.  Measures to store the AFO's process wastewater plus runoff from storms up to and including
         a 25-year, 24-hour rainfall event.  (This matches the feedlot effluent guideline requirement.)

         2.  Measures to prevent pollutant movement to ground water.

         3.  Measures to manage stored process wastewater and runoff, accumulated solids and dead
         animals in an appropriate waste utilization system.

 EPA conducted extensive analysis to estimate the likely cost to AFOs of implementing these
 management measures.73  This set of recommended measures for AFOs and accompanying cost analysis
 have been used many times subsequently in analyses by EPA and others (e.g., the measures and cost
 analysis were adopted by GAO in the 1995 study Animal Waste Management and Water Quality Issues,
 and by EPA for the 1996 Clean Water Needs Survey Report to Congress).

         For purposes of this analysis of the potential costs of the proposed designation provision, we
 have adopted both the set of CZARA management measures recommended for CAFOs and the
 accompanying estimates of their costs. DPRA's 1995 Economic Impact Analysis document describes
 these measures and costs in more detail:

         Management measures considered in this analysis focus on (1) facility runoff controls, (2) nutrient
         management practices with respect to the disposal of manure on agricultural land and (3) dead animal
         composting for poultry.  Facility runoff is controlled primarily through diversions (diking) for
         eliminating run-on and channeling on-site effluents to the ultimate control mechanisms... [The
         measures also] included lined retention ponds and irrigation for ultimate disposal of effluents-

         Variable rainfall patterns were also considered in  estimating management measures for controlling
         run-on and runoff. For example, retention ponds  were designed to hold runoff from a 25-year, 24-
         hour rainfall event, plus allowances for additional freeboard, process water, and storage for periods
         when water cannot be disposed. Because a storm of this magnitude is variable, control designs varied
         to provide capacity for 4-, 6-, 8- and 10-inch rainfall events. Ultimately, facilities in different states
         were required to have management measure controls designed appropriately for the expected rainfall
         7j DPRA Incorporated. Economic Impact Analysis of Coastal Zone Management Measures Affecting
 Confined Animal Facilities. October?, 1992. Also, DPRA Incorporated. Economic Impact Analysis of National
 Nonpoint Source Management Measures Affecting Confined Animal Facilities. May 17, 1995.
                                                                                              101

-------
        event (e.g., Louisiana facilities were designed on the basis of a 10-inch 25-year/24-hour storm,
        Michigan facilities for a 4-inch storm).

        Investment costs for effluent control mechanisms vary by size of facility and expected intensity of
        storm events. The ranges of total investment costs are... [from $900,000 for a small swine operation
        to $60,500 for a large dairy]... (Effluent control costs for poultry operations, which are presumed to
        be entirely enclosed, are not significant and were not estimated.)... Annual operating costs for these
        facilities, including farmer provided labor, range from a low of approximately four percent to a high
        of eight percent of total investment costs.

        Manure management includes provisions for disposing of manure on agricultural land.  Manure
        management consisted of spreading manure so that application of the nitrogen and  phosphorus
        constituents do not exceed  60 and 20 pounds per acre, respectively.  This requirement increased
        manure land application costs by requiring operators to apply manure over larger areas, consequently
        increasing transportation  costs. Transport distances were estimated to increase up to five (5) miles
        (one-way) for the largest model facilities considered.  Manure management cost estimates ranged
        from approximately $ 100 per year for small swine feedlots to over $ 12,000 per year for large dairies..

        Dead animal disposal was considered only for poultry (layers and broilers). The disposal method
        considered was composting dead birds with  poultry litter. Composting facility investment costs
        ranged from approximately  $400 for small 5,000-bird facilities, to more than $6,700 for 80,000-bird
        facilities.

        Total annualized costs (including annualized investment costs and annual operating costs) for all
        components of the effluent control and nutrient management and composting systems are presented
        in [a large spreadsheet,  an electronic version of which is available to accompany this report].
        Investment costs were annualized over the expected life of the effluent control mechanisms at a
        discount rate of 10 percent.  (Source: DPRA,  1995. op  cit., pp 7-8.)

        We made several adjustments in using DPRA's unit costs for AFOs for this study.  First, we
deleted a cost reduction of $3,500 per farm assumed by DPRA to be provided in  environmental cost
sharing funds by Federal and State agricultural agencies.  Our analysis aims to estimate the total social
cost of the proposed designation provision. Cost sharing funds reduce compliance costs to farmers, but
for our analysis  they represent transfer payments  from governments rather than any reduction in social
costs.  Second, we extended DPRA's cost estimates to several new model AFOs  that were sized smaller
or larger than  the range that DPRA had dealt with in its analysis.  We did this by developing regression
equations giving DPRA's estimated costs as a function of AFO size, and using these estimated equations
to extrapolate costs for model AFO sizes outside  of DPRA's range. Third, we updated DPRA's costs to
1999 dollars, using the  index of producer prices.

        There are both pros and cons in our using the  DPRA costs to estimate the unit costs for AFOs
that will need to improve their  environmental performance to meet the requirements of TMDLs:

        Pro: The DPRA costs cover a broad set of minimum management measures that are very typical
        of the least that a representative AFO will need to do. The costs reflect the effluent guideline
        requirement of no discharge except for a  25-year and 24-hour storm, and add some modest
        manure spreading requirements. The costs do not reflect several sorts of more extensive
        management measures that might or might not ultimately be required in  practice: development of
        and adherence to a comprehensive nutrient management plan, and assurance by the CAFO of


102

-------
        appropriate use for any of its manure transported off-site.  The management measures that are
        costed out for this analysis fall somewhat short of those that USDA/EPA project will be required
        at AFOs that will be permitted under these agencies' "Draft Unified National Strategy for
        Animal Feeding Operations".74

        Pro: The DPRA costs represent the most comprehensive set of costs for AFO water quality
        management that EPA has developed, and they have been widely used by the Agency and others.

        Con: The DPRA costs presume that no treatment is yet in  place at AFOs. DPRA costed out
        construction of waste lagoons, retention basins and liquid  waste irrigation equipment assuming
        that no portion of any of these facilities or equipment was yet in place.  In fact, the great majority
        of AFOs likely already have some portion of these items in place (but not all of them; many
        AFOs already have waste lagoons, but many of the lagoons are not lined).  Our use of unit costs
        that presume no treatment in place probably results in overestimating the incremental costs that
        will need to be incurred at most AFOs. However, we are not aware of any quantitative
        nationwide information on waste management facilities existing at AFOs that might allow us to
        scale down.DPRA's estimated costs to reflect the amount  of treatment that is actually in place.

        Con: In some other respects, DPRA underestimates likely incremental compliance costs by
        underestimating the extent of the management measures that many AFOs will need to incur.  For
        example, DPRA does not assume that poultry farms need manure storage structures. In fact,
        virtually all poultry farms do need such structures (perhaps only half of them now have them) in
        order to store dry litter appropriately after it is removed from the poultry house until it can be
        spread on fields. Also, DPRA's assumptions seem quite low regarding the typical distance from
        an AFO that manure will need to be transported before it can be spread consistent with
        agronomic needs. DPRA assumes transport distances of 0 - 5 miles. In  many areas of the
        country, however, there are large excesses of confined animal manure relative to local crop
        needs.75 In these areas, shipping manure 20 - 50 miles or more is likely to be necessary before
        crop land can be found that is not already oversaturated with nutrients.  In the four major poultry
        producing counties of the Delmarva peninsula (DE and MD), for example, it is estimated that 95
        % of the manure produced is excess considering local crop needs and existing soil nutrient
        concentrations. Much of this manure would need to be transported more than 100 miles before
        74 For example, permits eventually issued to AFOs to reflect the strategy will require compliance with a
revised, more stringent feedlots effluent guideline, and will require all permitted AFOs to develop and comply with
an appropriate comprehensive nutrient management plan (CNMP), which will likely include practices relating to:
feed management, manure handling and storage, land application of manure, land management, management of
dead animals, and record keeping.  See USDA, USEPA. Draft Unified National Strategy for Animal Feeding
Operations. September 11, 1998.

        75 USDA estimates that in 107 - 485 counties across the U.S., the volume of confined animal manure
produced exceeds agronomic needs within the county. These "excess manure" counties account for a large share of
all the AFOs in the U.S.  Source: Lander. Mofitt and Alt. Nutrients Available from Livestock Manure Relative to
Crop Growth Requirements. Resource Assessment and Strategic Planning Working Paper 98-1.  February, 1998.

                                                                                            103

-------
                                              •
        finding sufficient crop land on which it could appropriately be spread.76

        Con: Embedded within the DPRA estimate of annualized costs is an assumed 10 % discount rate.
        We would prefer to use a 7 % discount rate as currently recommended by the Office of
        Management and Budget However, it would be very difficult to disentangle DPRA's annualized
        costs into capital and annual O&M costs and then substitute our preferred discount rate.  We
        have chosen to retain DPRA's annualized costs despite their 10 % discount rate assumption.
        Annualized compliance costs with a 10 % discount rate may be, by our rough estimate, about 0 -
        5 % higher than they would be if we were to recalculate them with a 7 % discount rate. This
        error is small compared with other likely uncertainties in our estimates.
                                     X
In sum, the DPRA cost estimates have some shortcomings relative to our desires, but they are the best
available without devoting significant new effort to costing.

6. Estimate total national maximum costs for management measures at AFOs by multiplying the
number of AFOs that are potentially subject to designation as CAFOs (#4) by the estimated unit
costs for these management measures (#5).

        From the Census of Agriculture, we obtained the total number of AFOs for each animal type
(beef cattle, dairy cows, swine, layers, broilers) within the counties with AFO-impaired waters that are
potentially designatable considering both size (large enough so as not to be de minimis, not so large as to
be CAFOs already)  and method of discharge. These AFOs were further broken out by size category
within the designatable range and by the conditions under which they could be designated (Class A, B, or
C). For each category of AFO (e.g., a swine farm with 200 - 499 head in Iowa), we then assigned the
unit cost appropriate to this AFO category77 and multiplied the number of AFOs in the category by the
corresponding unit costs. We then summed across all the categories of AFOs (all sizes of beef, dairy,
swine, layer and broiler AFOs within the potentially designatable size range) to obtain an estimate of the
maximum possible compliance cost if all the designatable AFOs in the 25 authorized States were to be
designated.  The estimated maximum  compliance costs  are shown  in Exhibit III - 4.
       76 Environomics, Inc. Combustion of Poultry Litter for Energy Recovery: an Analysis of the Economic and
Environmental Desirability of Litter Combustion on the Delmarva Peninsula. Study for U.S. EPA/OP. March 17,
1999.

       77 DPRA's unit costs were developed for categories of AFOs by rainfall region (4, 6,  8, or 10 inches of
precipitation in a 25-year/24-hour storm) and by north vs. south (farms in northern climates were presumed to be
unable to spread manure when fields are frozen in the winter, and hence must incur higher costs for manure storage)
rather than by State. However, DPRA also provided information indicating the fraction of each State that fell into
each rainfall region, and whether each State was northern or southern.  We could thereby translate DPRA's unit cost
data, for example, for swine farms with 200 - 499 head in each different rainfall region in the north and in the south
into a single average unit cost applicable to swine farms of 200 - 499 head in Iowa.

104

-------
Exhibit HI - 4. Maximum Management Measure Costs if All Possible AFOs Were Designated

MAXIMUM » OF AFOs THAT COULD BE
DESIGNATED
Class A
Farms

AL
CA
DE
IA
IL
IN
KS
KY
LA
Ml
MO
MS
MT
NO
NE
OH
R,
sc
so
TN
VA
VT
Wl
WV
WY
236
62
92
59
172
88
298
16
16
71
31
117
21
43
55
149
1
20
304
51
7
62
169
11
40
Class B
Farms
Class C
Farms
i
17
3
2
17
59
38
149
17
8
40
15
12
6
40
15
87
2
4
132
28
3
45
289
6
8
898
11
321
1
1
15
8
3
: 5
4
: 25
532
3
1
0
30
1
27
5
28
2
1
11
11
0
MAXIMUM COSTS IF ALL POSSIBLE AFOs ARE DESIGNATED
Class A
Farms
Class B
Farms
Class C
Farms
TOTAL
i I
$753,162
$780,198
$264.946
$275,864
$815.888
$415,016
51,699,693
$106.120
S186.497
$409.321
$140.508
$426.076
$152.596
$272,017
$275,897
$680.235
S7.977
$119,723
$1,828.930
$373,543
S56.404
$645,763
$1.381,967
$47.513
$361.983
$40.645
$8,830
$3,585
S33.601
$112,554 •
$72,232
$274,149
$39,978
$40,444
$84,039
$27.923
$36.790
$12.379
$76.697
$23.552
$153,337
$3.996
$7,197
$233.265
$72,889
$9.247
S122.212
$622.517
$10,392
i S1 6.909
57.563.613 $8.357.420
$122,333
$2.331.066
$4,824
$9.911
$85.307
$46.178
$20.828
$911.361
$2.599.597
$314.289
$938.353
$572,555
$2,020,020
$166,925
$51.863 ! $278.804
$25.963
$519.323
$272,392 | $440.823
$5,370,618
. S2.245
$5.833.484
$167.220
$4.730 S353.444
$1,293 I $300,742
$170,192 $1.003.764
SI .882
$267.927
$13,855
$394.847
$19,092 ! $2,081.286
$136^19 S582.651
$6.930 $72.581
$4.628 $772.603
$101,829 j S2.106.313
$30.890
$88,796
SO I $378.893
; i i
TOTAL
2.192
1.041
1.946
$12.477.836
, S2.139.360
$16.652.753 : $31.269,949
Scale-up
to cover
all 41
authorized
States

j


3.441 1.634 3.055 $19.590,202 | S3.358.795 | $26.144,822 | $49.093.820 |
                                                                             105

-------
7. Estimate the frequency with which EPA will likely need to invoke the proposed Federal
designation authority, and the management measure costs attributable to the proposed regulation
that will result from these designations.

       In subsection #6, we estimated the management measure costs that could result if EPA invoked
the proposed designation authority for all of the potentially designatable AFOs in the identified counties.
However, EPA expects that the Agency will need to invoke the proposed Federal designation authority
only infrequently, far less often than this maximum. Most often, States will want to implement
themselves their TMDLs involving AFOs.  In general, there are several options available to most States
for implementing AFO-related TMDLs that are far preferable to leaving implementation to EPA and use
of the proposed Federal designation authority.  Federal designation of AFOs as NPDES point sources and
permitting them will be costly both for the State (which will likely need to permit the AFOs following
Federal designation, and enforce the permits) and for the AFOs.  States will want to avoid this, and the
proposed Federal designation authority will likely be a last resort:

•      Sometimes the impairment will be of a sufficiently limited magnitude such that existing
       voluntary programs (e.g., USDA, EPA and State cost-share programs) available to AFOs will be
       enough to induce improved performance by AFOs and eliminate the impairment. We expect
       that, in any instance where voluntary programs are sufficient to achieve attainment, a State will
       prefer to use them over designating and permitting the AFOs as point sources.

•      Sometimes achieving water quality standards will require regulatory efforts in addition to
       voluntary programs. Often, existing State regulatory authorities will be sufficient for this
       purpose. Many States have their own well established regulatory programs and authorities to
       address water quality impacts from AFOs that can be used to assure implementation of whatever
       controls are called for by a TMDL.

•      Even for a State that does not have an AFO regulatory program, the State has the option of using
       the designation authority available under existing Federal NPDES regulations.  We assume that
       most  States would prefer to use this designation authority themselves and control the outcome
       rather than leave designation to EPA with unpredictable results.

       For this cost analysis, we will assume that for most TMDLs,  States will achieve the necessary
controls over  AFOs preferentially through voluntary programs, existing State.regujatory programs, or
State use of the existing NPDES designation authority. We will assume that EPA will need to use the
proposed Federal designation authority very rarely in States that now have strong AFO programs and
authorities, and  somewhat more often in  States that do not have such programs.  We will also assume
that EPA will be less  likely to use the designation authority for small (Class B) AFOs than for larger
(Class A or C) AFOs.  This is for several reasons:
                            /
•      There is generally a greater environmental gain available in improving waste management at a
       large  facility that there is at a small facility. The cost of inspecting, designating and permitting a
       large  facility, though, is similar to that for a small facility.

•      The existing NPDES regulations signal some lenience toward smaller AFOs. AFOs with
       between 300 and 1,000 animal units can be designated and permitted if they are actual or
       potential significant polluters whether or not they meet the method of discharge criteria.  It is

106

-------
        more difficult to bring smaller AFOs under the permit program, however. Smaller AFOs with
        less than 300 animal units can be designated and permitted only if they are both actual or
        potential significant polluters and they meet the method of discharge criteria.

        Given these assumptions, based on rough estimates by EPA TMDL program officials, we will
assume that EPA will need to invoke the designation authority:

        For Class A and C AFOs:

               For 1 - 5 % of all AFOs affecting AFO-impaired waters in States that have strong AFO
               regulatory programs; and

               For 5-25% of all AFOs affecting AFO-impaired waters  in States that do not have
               strong AFO regulatory programs.

        For Class B AFOs:

               For 1 - 2 % of all AFOs affecting AFO-impaired waters in States that have strong AFO
               regulatory programs; and

               For 2 - 10 % of all AFOs affecting AFO-impaired waters  in States that do not have
               strong AFO regulatory programs.

These estimates are quite uncertain.  There is no basis on which one can predict confidently how often
States will choose to implement their TMDLs for AFOs themselves, and how often they will defer
implementation to EPA. TMDL program experience to date is that EPA has been developing
significantly less than 20 % of the TMDLs (and thus that EPA might have reason to address significantly
less than 20 % of the AFOs affecting AFO-impaired waters), but whether this experience in the early
stages of the program provides much of a guide to the future is open to debate.

        We will use these ranges (in States with strong AFO programs: 1  - 5 % of Classes A and C; 1  - 2
% of Class B; and in States without strong AFO programs: 5 - 25 % of Classes A and C; 2 - 10 % of
Class B) to scale down the maximum possible costs for use of the proposed Federal designation  authority
and obtain a cost estimate reflecting the likely frequency with which the proposed authority will actually
be used.78

        Before presenting these scale-down calculations, we will briefly describe the nature of existing
State AFO programs and how we identify States as having strong AFO programs or not.

        The 50 States have adopted a very wide variety of authorities and programs to address potential
        78 A reader interested in assuming a frequency of Federal designation different from our assumptions can
perform the calculations necessary to obtain a corresponding cost estimate. We have already presented estimates of
the maximum possible costs assuming the Federal designation authority is invoked for all potentially designatable
AFOs affecting AFO-impaired water bodies. This maximum possible cost can simply be scaled down by some
other assumed frequency with which TMDL implementation might be deferred to EPA.

                                                                                            107

-------
water pollution from AFOs.  Most but not all authorized NPDES States follow the NPDES regulations by
treating very large AFOs as CAFOs and specifying waste management performance standards or
requirements for them. Some States have implemented their CAFO program thoroughly (issuing
individual or general permits for most CAFOs and enforcing them), while others have issued permits to
only a fraction of the AFOs nominally meeting the NPDES size standards for CAFOs. EPA estimates
that there may be as many as 10,000 AFOs with more than  1,000 animal units, and that virtually all of
them are required to be permitted as CAFOs under the NPDES regulations. However, there are only now
about 2,000 CAFOs that have NPDES permits.79 An earlier review found that of these permits, only 760
were current as of 1995.  On average, less than 100 CAFOs with NPDES permits are inspected annually.
However, many States issue permits equivalent to NPDES under State programs, and these may not get
counted in these statistics.80

       Many States have also adopted regulatory requirements (sometimes involving permits as a
vehicle for implementation, sometimes not) for some sorts  of AFOs that are smaller than the CAFO
thresholds. Many States  have developed a strong regulatory program for the variety of animal operation
that is most problematic in the State, but have less  intensive or no programs for other sorts of animal
operations. States also vary in which of an AFO's activities are addressed: many States have
requirements regarding liquid waste lagoons and the standards to which they must be constructed, fewer
have requirements regarding land application of the manure removed from lagoons, and very few have
comprehensive nutrient management requirements that address both nitrogen and phosphorus.

       Several sources have recently compiled information on State programs and authorities relating to
AFOs.  These sources include:

•      EPA's "Feedlots Study", which is gathering information on AFOs to support consideration of
       revisions to the existing Feedlots Point Source Category effluent guideline.
       (www.epa.gov/ost/guide/feedlots/').  Summaries of many State AFO programs appear in
       Appendix I (Introduction) and Appendix II (Preliminary Study of the Livestock and Poultry
       Industry, Chapter 9: Current Regulation of Livestock and Poultry Operations). In addition, a
       further summary of "State CAFO Permitting Activities" is available in draft from EPA/OST staff
       responsible for the study.

•      Agricultural economists from a dozen universities conducted a 1998 National Survey of Animal
       Confinement Policies covering 35 States. The State-by-State survey responses are available at
       http://cherokee.agecon.clem son.edu/confine.htm.

•      The Environmental Law Institute (ELI) conducted  a State-by-state review of enforceable State
       authorities for control of nonpoint source water pollution. The ELI study  includes a description-
       of each State's authorities, with the agriculture section often indicating authorities relevant to
       AFOs. (Environmental Law Institute. Almanac of Enforceable State Laws to Com'rol Nonpoint
       79 U.S. Department of Agriculture, U.S. EPA. Draft Unified National Strategy for Animal Feeding
Operations. September 11, 1998.

       50 Ira H. Linville, Regional Agriculture Coordinator, USEPA Region 4.  Animal Waste Management
Issues: a Federal Perspective, www.bae.uea.edu/extension/pubs/linville.

108

-------
       Source Water Pollution.  1998.)

       The Natural Resources Defense Council completed a report in December, 1998, that reviewed 30
       States' practices in managing pollution from animal feeding operations. The report, America's
       Animal Factories: How States Fail to Prevent Pollutionfrom Livestock Waste, is available at
       http://www.nrdc.org/nrdcpro/factor/aafinx.html.

We reviewed these four sources in order to compile a picture of each State's authorities, and, to a lesser
degree, its performance in managing water pollution from APOs. Based on this review, we have termed
a State's program as "strong" if it generally meets the following criteria:
                             X
•      The State has enforceable authority to require appropriate water pollution prevention and control
       measures for both large AFOs (> 1,000 animal units) and medium-sized AFOs (300 - 1,000
       animal units) confining the animal, species that are most prevalent in the State (e.g., poultry in
       MD, dairy cows in WI, swine and poultry in NC, etc.); and

•      The State is making reasonably comprehensive efforts to implement and enforce its regulatory
       program or is conducting an active, well-funded voluntary program.

These two criteria are admittedly quite judgmental, and we have no explicit definitions of "appropriate",
"comprehensive", "well-funded", etc..  Given the subjectivity of our judgments and the frequent non-
comparability of the available information from State to State, a reviewer might disagree with us on
whether a given State's AFO program is strong or not. The following exhibit shows our judgment as to
whether or not a State currently has a "strong" AFO program. The exhibit covers the 25 authorized
NPDES States (where the proposed Federal designation authority might have some impact) that reported
water bodies impaired by AFOs to a major or moderate extent.

                                         Exhibit m-5
              States With Strong AFO Programs Currently (Authorued States Only)
Strong
CA, IA, IL, KS, MS, MT, NE, ND, SC, TN, VT,
VA, WI, WY
Not Strong
AL, DE, IN, KY, LA, MI, MO, OH,


RI, SD, WV

       To develop lower bound estimates for the number of AFOs and control costs resulting from the
proposed Federal designation authority, we assumed that EPA would use the designation authority to
address 1 % of the Class A, B and C AFOs in States with "strong" AFO programs and 5 % of the Class A
and C AFOs and 2 % of the Class B AFOs in States without "strong" AFO programs. For upper bound
estimates, we assumed that EPA would use the designation authority to address 5 % of the Class A and C
AFOs and 2 % of the Class B AFOs in "strong " States and 25 % of the Class A and C AFOs and 10 % of
the Class B AFOs in "not strong" States. Thus, we scaled down our previous estimates of the maximum
number of potentially designatable AFOs and maximum possible control costs by these percentages,
reflecting the proportion of all AFO-related TMDLs in which we assumed that EPA would need to use
the proposed designation authority.
                                                                                          109

-------
       After obtaining the number of AFOs likely to be designated by EPA (and their control costs) in
the 25 authorized States that reported AFOs as a cause of impairment, we developed a scale-up factor in
order to develop national estimates of the number of affected AFOs and their control costs. There are a
total of 41 authorized States, 16 for which we do not have data, as explained earlier in subsection #3 of
this paper.  The scale-up factor of 1.57 is based on the relative number of animal units at CAFOs in the
25 States for which we have data and in the remaining 16 authorized States.

       The scaled-up control cost estimates reflect the costs that would occur if all TMDLs for AFO-
impaired waters and all designations pursuant to these TMDLs were to occur in the first year following
promulgation of the proposed designation authority (assumed to be the year 2000). However, TMDLs
will actually be completed gradually over the 16-year period from 2000 through 2015. As discussed in
greater detail in Subsection #5 of Section II (silviculture), we assume that designation of sources as
necessary to implement these TMDLs will similarly occur at an even pace over these 16 years. This
gradual designation process significantly reduces the present value of the control costs that will result
from the proposed designation authority. In  Section II, we calculated a factor to convert control cost
estimates from fully front-loaded annualized costs (assuming simultaneous designations of all affected
entities in 2000) to annualized costs reflecting gradual designation over the period 2000 - 2015. We
applied this conversion factor (.631745) to our control cost estimates to obtain our final  national control
cost estimates.

       Exhibit III-6 shows the national  estimates of regulatory impacts resulting from 1) scaling-down
(to reflect EPA using the designation authority for only a small fraction of all TMDLs and AFOs), 2)
scaling-up (to reflect 16 States that reported on causes of their impaired waters in a manner that did not
allow for identification of AFOs as a cause), and 3) scaling-down (to reflect gradual designation over 16
years rather than immediate, simultaneous designation in the year 2000).  We estimate that the proposed
Federal designation authority will result in:

•      234 -1,136 AFOs eventually being designated as CAFOs;

•      Annualized costs  for management measures at these designated AFOs of $0.94 - $4.60 million;

The range in estimates reflects uncertainty about the frequency with which the Federal designation
authority will need to be used.
110

-------
Exhibit III - 6: Estimated Number of AFOs to be Designated and Compliance Costs
AL
CA
DE
IA
11
IN
KS
KY
LA
Ml
MO
MS
MT
ND
NE
Oil
Rl
sc
SD
TN
VA
VT
Wl
WV
WY
TOTAL
Scale-up la
cover all 41
nuthorlzed
Stales
Scale-down
lo cellecl
jraduEil
designation
Mi
Class A
farms
230
62
92
59
172
88
298
16
10
71
31
1J7
21
43
55
140
1
20
304
51
7
02
1G9
11
40
2,192
3,441
\XIMUM »O
THAT COUL
OESIGNAT
Class B
farms
17
3
2
17
59
38 '
149
17
8
40
15
12
6
40
15
87
2
4
132
28
3
4_5
2B9
6
8
1.041
1.034
FAFOs
DOE
ED
Class C
farms
898
11
321
1
1
15
8
3
8
4
25
532
3
1
0
30
1
27
5
20
2
1
11
11
0
1.946
3,055
MAXIMUM COSTS IF ALL POSSIBLE
AFOs ARE DESIGNATED
Class A
costs
$753,162
$780.198
$264,948
$275,864
$815.888
$415.016
$1.699,693
$108.120
$188.497
$409,321
$140.508
$428,078
$152,598
$272.017
$275.897
$1390.235
$7,977 	
S1 19.723
$1.828.930
(373.543
$56,404
$645.763
$1,381,987
$47.513
$361,983
$12.477.836
$19.590.202
Class B
costs
$40.645
$8,830
$3,565
$33,601
$112.554
$72.232
$274,149
$39.978
$40.444
$64,039
$27.923
$38,790
$12.379
$78,897
$23.552
$153,337
$3.998
J7.197
$233.265
$72.889
$9,247
$122.212
$922.517
$10.392
$16,909
$2.139.360
$3.358.795

Class C
costs
$7,583.613
$122.333
$2,331.066
$4.824
$9.911
$85.307
$46,178
$20.828
$51.883
$25.963
$272.392
$5,370.618
$2.245
$4,730
$1.293
$170,192
$1.882
$267.927
$19.092
$136,219
$6.930
$4,628
$101.829
$30.800
$0
$16,652.753
$26.144.822

Slrong
Slate AFO
Program?
NO
YES
NO
YES
YES
NO
YES
NO
NO
WO
NO
YES
YES
YES
YES
NO
NO
YES
NO
YES
YES
YES
YES
NO
YES



NUMBER OF AFOs
TO BE
DESIGNATED
LOW
57
1
21
1
2
6
5
1
1
5
3
7
0
1
1
11
0
1
18
1
0
1
5
1
0
149
234

i HIGH
285
4
103
3
10
30
18
8
6
23
16
33
1
3
3
54
1
2
90
5
0
4
15
8
2
724
1.136

ANNUALI2ED COMPLIANCE
COST FOR DESIGNATED
AFOs
LOW
$416.652
$9,114
$129.872
$3.143
$9,384
$28,461
$20,200
$7.147
$12.727
$23.445
$21,203
$58,335
$1.872
$3.534
$3.007
$45,588
$573
$3,948
$97.068
J5.827
$726
$7.728
$21,003
$4.128
$3,789
$936.330
$1.470.039
$928.690
HIGH
$2.083.258
$45.303
$649.361
$14.706
$43.541
$132.304
$92,777
$35.735
$83,634
$117.225
$106.017
$280,571
$7.990
$15,371
$14,331
$227.841
$2.884
$19,526
$485.332
$26.046
$3.352
$34,984
$86.640
$20.640
$18.437
$4.638.766
$7.282.863
$4,600.912
                                                                                         111

-------
8. Estimate the administrative costs for designated AFOs to obtain and maintain coverage under
NPDES permits.

       This will be estimated by multiplying the number of operations that will likely be designated
(#7) by an estimate of the administrative cost for a typical AFO.

       AFOs within areas where the designation authority will be invoked will need to incur the
administrative costs of being designated and then seeking and maintaining coverage under the general
permits that EPA or the State will issue following designation. We assume that an AFO that is a
candidate for designation will need to:

•      Host a site visit by an EPA representative. Under existing regulations, a permit application
       cannot be required of an AFO that is designated as a CAFO until "the Director has conducted an
       on-site inspection of the operation and determined that the operation should and could be
       regulated under the permit program." EPA's guidance manual on NPDES regulations provides a
       four page sample questionnaire representing the sorts of information that the permit authority's
       representative should obtain during the site visit.81 The information includes a description of the
       facility's location, setting, layout and number of animals confined; a facility diagram; details on
       the facility's waste management devices and procedures; information on any discharges from the
       facility; assessment of the potential for the AFO to affect water quality; and recommendations
       regarding further action. The site inspection is intended to facilitate making decisions on
       whether the operation is an AFO (are animals stabled, confined and fed or maintained for 45
       days or more in a 12-month period?) and whether it meets the criteria for designation (is it
       potentially or actually a significant contributor of pollutants to waters of the U.S., how many -
       animals are confined, does it meet the "method of discharge" criteria?). Given the importance to
       the AFO of the NPDES Program Director's decisions that will be based on the information to be
       obtained during the site visit, we assume that the AFO's owner or manager will want to be
       present during the inspection. The information that the AFO representative will be asked to
       provide during the visit should be readily available (e.g., approximate number of animals
       confined, dimensions of a waste storage lagoon, description of manure management practices)
       but may require modest effort to assemble. We estimate that roughly 2 hours will be sufficient
       to assemble the desired material,  and that another 2 hours will  be necessary to host the actual site
       visit.

Assuming that the AFO is then designated as a CAFO and the permit authority issues a potentially
applicable general permit, the owner or his representative will then need to:

•      File an initial Notice of Intent (NOI) to seek coverage under the general permit.  The NOI will
       likely include identification of the applicant, a description of the AFO and its waste management
       facilities, site information, information on potentially affected water bodies, and perhaps *nore.
       Such information is not unlike that required for NOls for coverage of industrial facilities or
       construction activities under the storm water general permitting program. We will assume that
       the effort required to prepare such an AFO NOI will be the same as that for the storm water NOI
        81 U.S. EPA, Office of Water. Guide Manual on NPDES Regulations for Concentrated Animal Feeding
Operations. December, 1995. Appendix B.

112

-------
        - estimated at 2 hours per NOI in EPA's ICRfor Notice of Intent and Stormwater Pollution
        Prevention Plan for Discharges Associated with Industrial Activity (March, 1999).

•       Update or amend the NOI whenever there is a substantial change in the AFO's activities or waste
        management practices.  We assume that such an amendment might be necessary on average once
        during the 5-year NPDES permit term.

•       Prepare some sort of animal waste pollution prevention plan, indicating how the performance
        standards and management measures required by the general permit will be translated into
        specific facilities, equipment and BMPs that will be employed in the applicant's operations.  We
        have not researched further details on what such a plan might contain. Some indications could
        probably be obtained by reviewing the requirements in States that already implement permit-like
        regulatory programs for AFOs.  Absent further information, we will assume that the burden
        associated with such a requirement would be similar to that for the Stortnwater Pollution
        Prevention Plans required under the stormwater general permit program.  This one-time burden
        is estimated in the relevant ICR at 80 hours per plan. However, we also assume that most AFOs
        will previously-have prepared much of the information required for such a plan in a farm nutrient
        management plan, an application for cost-share assistance  for soil conservation or waste
        management measures, or a similar document. We will assume that the incremental effort for
        the AFO to prepare the  pollution prevention plan will average 40 hours.

        Assuming that the animal waste pollution prevention plan  remains valid for an average of 10
years, and that a new NOI would need to be filed for general permit coverage every five years, the
administrative burden over a 10-year period to a designated and permitted AFO would be:

•       46 hours in the first year, consisting of 4 hours to host the  site visit, 2 hours to file the NOI, and
        40 hours to prepare the  animal waste pollution prevention  plan;

•       6 hours spread over the remaining nine years, including two hours each for twice filing updated
        NOIs, and two hours for filing the new NOI for general permit coverage in year six following the
        expiration of the first five-year permit.

This burden is equivalent to 45.33 hours in the first year, plus .67 hours annually over each of the 10
years.82 Annualizing the first year burden of 45.33 hours (by applying a capital recovery factor of .1424
for a 10-year asset life  at a 7 %  discount rate), we obtain 6.45 hours annually. The total annualized
burden estimated for a designated and permitted AFO is thus 7.12  hours.

        To these burden estimates, we apply a costing factor of S36.12 per hour for fully loaded private
sector labor costs, drawn from the recent ICRfor Applications for the NPDES Discharge Permit and the
Sewage Sludge Management Permit (December,  1998).
        82 The six hours spread over the nine years following the fust year are equivalent to 0.67 hours per year for
years 2 through 9. If we add to this 0.67 hours in the first year, this becomes 0.67 hours annually for the  10 years;
in effect an annual cost of 0.67 hours. Subtracting this 0.67 hours in the first year from the first year burden of 46
hours leaves 45.33 hours of first year costs, in effect a capital cost of 45.33 hours.

                                                                                             113

-------
       The total estimated annualized administrative cost to a designated and permitted AFO is thus
$257.17. We apply this per-AFO figure to the number of AFOs expected to be designated in the 41
States (234 - 1136, depending on the choice of low or high designation frequency, from Exhibit III-6).
To reflect gradual designation over 2000 - 2015, we also multiply by the scaling factor of 0.631745. The
resulting estimated annualized administrative costs to sources are:

       Low designation frequency (234 AFOs):                      S38,017/yr

       High designation frequency (1136 AFOs):                    S184,561/yr

9. Estimate the administrative costs to Federal and Stat£ governments for designating and
permitting these AFOs.

       In using the proposed designation authority, EPA will incur administrative costs to make the
required site inspection for all potentially designatable AFOs, and EPA or the States will then incur costs
subsequently to designate and permit the chosen AFOs.

       We assume that a site inspection of an AFO will require an EPA regional office staff person to
spend 1A day, including travel time to the site. We assume the staff inspector can visit 2 AFOs in a day.
AFOs will typically be some distance away from the EPA regional office responsible for the inspections.
We assume an average travel cost of 5100 for transportation and $100 for lodging and per diem for a
typical day spent inspecting 2 AFOs.  To inspect 2 AFOs will thus require 8 hours of staff time (or
$253.44, priced at $31.68/hr for fully loaded State or Federal labor costs, drawn from the recent ICRfor
Applications for the NPDES Discharge Permit and the Sewage Sludge Management Permit [December,
1998]) plus $200 in travel costs, for a total of $453.44. The one-time inspection cost per AFO will be
half of this, or $226.72 per AFO.

       We assume that designation of the inspected AFOs that are to become CAFOs and issuance of a
general NPDES permit prescribing the requirements that the designated AFOs must meet will be
accomplished simultaneously.  We also assume that there will be a single AFO general permit issued for
each State in which AFOs will be  designated as point sources.  This assumption of a single State-wide
AFO general permit matches the typical practice to date, in which many States and several EPA regions
have issued Statewide AFO general permits. It  has been rare that more than one AFO general  permit has
been issued covering different sets of AFOs within a State (e.g., perhaps one general permit for large
AFOs in a State and another general permit for smaller AFOs in the State, or different general  permits
for AFOs that raise different animal species, or  different general permits for AFOs in different
watersheds in the State). Evidently, EPA and the States have generally felt that there is sufficient
consistency in the requirements that different sorts of AFOs within  a State should meet to warrant a
single State-wide AFO general permit.

       We assume, for an upper end estimate, that EPA will use the proposed designation authority for
at least one TMDL in all of the 41 authorized States.  The upper end thus involves issuance of 41 State-
wide general permits. For a lower end estimate, we assume that EPA will use the proposed designation
authority for at least one TMDL in half of the 18 authorized NPDES States that do not currently have
strong AFO regulatory programs.  The lower end thus involves issuance of 9 State-wide general permits.

       We assume that these general permits will all be issued before 2008, during the first eight years

114

-------
following promulgation of the regulations. We assume therefore that an average of 5.125 (high end) or
1.125 (low end) general permits will be issued annually during this period.

       EPA's "Revised Workload Model fortheNPDES Program" (Jim Tart, Rob Wood, May 5, 1995)
estimates the following pricing factors for general permits:

•      0.5 FTE per general permit for issuance. In issuing 5.125  general permits annually, there will be
       an annual cost of 2.5625 FTE in each year through 2007. In issuing 1.125 general permits
       annually, there will be an annual cost of 0.5625 FTE in each year through 2007. After 2007, all
       the general permits will have been issued and there will be no further cost.  We assume that it
       makes no difference whether a State or EPA issues the permit.            x

•      0.125 FTE/yr/State for continuing EPA review of existing general permits.  We assume that
       these new AFO general permits will constitute about 1/5 of the volume of general permits in the
       States that issue them.  If so, the annual EPA review burden, for the high case, will increase from
       .128 FTE (.125 x 1/5 x 5.125) in 2000 to 1.025 FTE (.125  x 1/5 x 41) in 2007 and thereafter. For
       the low case, the burden will increase from .028 FTE in 2000 to .225 FTE in 2007 and thereafter.

       In addition, we assume that the permit authority will need to spend  an average of 2 hours to
process and review each NOI that is submitted. Each designated AFO will  submit an NOI in the first
year following designation and three more over the next nine years. The review burden per designated
AFO will be 2 hours in the first year and .67 hours annually in years 2 through 10.

       To these burden estimates, we apply a costing factor of $65,895 per FTE for fully loaded State or
Federal labor costs, drawn from the recent ICR for Applications for the NPDES Discharge Permit and
the Sewage Sludge Management Permit (December, 1998).

       Summing these streams of administrative costs for EPA and the States, discounting them back to
the present and annualizing them, the resulting estimates are:

•      Costs for inspection
               Low use                                                        $2,193/yr
               High use                                                        $10,644/yr

•      Costs for general permit issuance
               Low use                                                        $15,493/yr
               High use                                                       $70,580/yr

•      Costs for general permit oversight
               Low use                                                        $ll,841/yr
               High use                                                       $53,943/yr

•      Costs for NOI processing and review
               Low use                                                        $4,472/yr
               High use                                                       $21,708/yr

       The total estimated administrative costs for EPA and the States for inspection, designation and

                                                                                          115

-------
permitting of AFOs as a result of the proposed authority are thus:

               Low use                                                         $33,999/yr
               High use                                                         $156,875/yr

10. Develop estimates of compliance costs and administrative costs specifically for small entities
and assess impacts

       The Small Business Administration has established the following standards defining small
businesses for the sorts of AFOs that we have analyzed:

•      Chicken eggs (layers)                               < $9.0 million annual revenues

•      Beef cattle feedlots                                 <$ 1.5 million annual revenues

•      Dairy cows, swine, and broilers                      < $0.5 million annual revenues.

       We determined which among the AFOs that we projected to be designated under the proposed
regulation were smaller than these size standards.  To assess the likely impacts on these small businesses,
we then compared their projected compliance costs resulting from designation (costs of the management
measures they will implement, plus their administrative costs) with their annual revenues.

       As suggested by EPA's Revised Interim Guidance for EPA Rulerwriters: Regulatory Flexibility
Act (March, 1999), we determined the number and percentage of small businesses that would incur
compliance costs exceeding 1 % of their revenues, and the number and percentage that would incur
compliance costs exceeding 3 % of their revenues.

       The Impact of Cost-Sharing Funds on Compliance  Costs to be Borne bv AFOs

       Thus far in this analysis, we have estimated the social costs of the compliance efforts that
designated and permitted AFOs will need to make. Cost share funds are available under a wide
variety of Federal, State and local programs to assist farmers in installing and maintaining the sorts of
measures that we project as necessary for compliance. In developing the cost estimates for the AFO
management measures necessary for compliance, DPRA assumed that $3,500 in cost-share capital funds
would be available to each AFO. This was equivalent to an annualized $411 in 1992 dollars, and DPRA
estimated compliance costs to farmers net of this governmental contribution.  We added this amount to
DPRA's cost estimates when we calculated the social costs of the rule, but we now need to subtract
whatever amount will be provided in governmental cost share funds in order to consider only the costs to
be borne by the AFOs themselves.
                                                                                          /
       Governmental cost-sharing programs available to assist farmers in paying for environmental and
conservation measures have changed somewhat since DPRA conducted their cost analysis. Most
notably, four programs were consolidated and replaced in the 1996 Farm Bill  with the Environmental
Quality Incentives Program (EQIP). EQIP is funded at $200 million annually through 2002.
Conservation practices relating to livestock production (including all of the measures assumed as
necessary in this analysis) are targeted with the funds and will receive 50 % of total EQIP funding.  In
general, cost share  assistance is available for up to 75 % of the cost of the conservation practices at a

116

-------
farm, and a single farm owner can receive payments for up to $10,000 per year and $50,000 for the five-
to ten-year life of a contract.  65 % of EQIP funds are available only in locally-identified priority areas
with significant natural resource (particularly water quality) needs, while up to 35 % can be used for
other statewide natural resource concerns.  In effect, the bulk of program funds are targeted at assisting
AFOs in areas of significant water quality impairment. EQIP assistance for animal waste storage or
treatment facilities is not available to large AFOs. USDA's definitions of large AFOs generally match
the farms that are already CAFOs under EPA's NPDES regulations.  Hence, most of the AFO portion of
EQIP funds is available nearly exclusively to exactly the AFOs that will be considered for designation
under the proposed regulation (i.e., small and medium AFOs affecting impaired water bodies).

        EQIP funds have been oversubscribed. In 1997 and 1998, the $200 million annual EQIP
appropriation has been sufficient to fund only about 1/3 of the requests received each year.83  (Although
figures are not  available on this issue specifically, we expect the EQIP funding rate is much higher than
1/3 for requests by AFOs located specifically in water quality-impaired areas.)84  The Administration has
requested increased funding for future years.

        Several sources of cost share funds in addition to EQIP are potentially available to owners of
AFOs.

•       EPA provides grants to States to assist them in implementing approved nonpoint source
        management programs (the "section 319 programs"), and a portion of these grant monies are
        used by States for cost-sharing on a variety of nonpoint source control projects. The Clean
        Water State Revolving Fund (SRF) is also increasingly being used by States to fund nonpoint
        source eligible implementation projects, such as animal waste storage facilities.  Since 1997, the
        SRF  program has funded over $650 million in nonpoint source-eligible projects, including
        AFOs.85

        USDA also manages the Conservation Reserve Program (CRP), which provides funds for
        establishment of conservation buffers, which are often used to assist in reducing the transport of
        animal manure nutrients to water bodies. In 1998, approximately $500 million was spent
        through the CRP. USDA also funds the Small Watershed Protection Program, which annually
        provides about $20 million to implement watershed plans that address water quality.

•       At least 22  States also have programs providing financial assistance to-farmers in addressing
        animal waste problems. These programs vary widely,  and include measures to support better
        83 USDA/USEPA. Draft Unified national Strategy for Animal Feeding Operations.  September 11, 1998.
Page 24.
                                             ./•
        84 EQIP has recently held a number of regional round table discussions with parties interested in
improving the operation of the program. A common view that emerged in these discussions has been that EQIP
requests from AFOs in non-priority geographic areas are funded much less frequently than requests in priority areas.
Owners of AFOs located in the non-priority areas, which apparently comprise the bulk of the AFOs in most States,
have complained frequently about this. Transcripts of these EQIP "listening sessions" are available at many sites on
the Internet.

        85 USDA/USEPA.  Op cit. p. 25.

                                                                                             117

-------
        management of animal wastes by such means as cost sharing for capital investments, low interest
        loans, grants to projects in critical watersheds, credits against State income or property taxes,
        supplements to EQIP cost sharing, and no-cost or subsidized preparation of nutrient management
        plans.86

        For this analysis, we assume that sufficient cost sharing funds will be available for AFO waste
management measures in critical areas (i.e., areas where AFOs are impairing water quality) to meet the
requests for financial assistance from 2/3 of the AFOs that apply.  In effect, we assume that AFOs in
critical areas can receive financial assistance from all sources at double the rate at which assistance is
available to AFOs generally (including both AFOs in areas where they are contributing to impaired
waters and AFOs in areas where they are not contributing to impaired waters) from EQIP alone. This
assumption seems reasonable. EQIP has been funding 1/3 of all applications from AFOs generally. This
1/3 rate of funding is comprised of a funding rate probably substantially less than 1/3 for AFOs in non-
critical areas, and a rate probably substantially higher than  1/3 for AFOs in critical areas. When we then
add the  additional sources of funding available for AFOs from USDA, EPA and the States (with several
of these programs also being targeted toward critical areas), it seems quite likely that the total funds
available would be sufficient to bring the fraction of AFOs in critical areas that receive assistance up to
or beyond 2/3,87  For this analysis, we assume specifically that:

•       2/3 of the designated AFOs will receive EQIP or EQIP-like financial assistance.  These AFOs
        will receive financial assistance for 3/4 of their waste management measure costs, up to a
        maximum of $50,000 over a ten year contract.

•       1/3 of the designated AFOs will not receive financial assistance.

We used these assumptions to adjust downward the costs of the compliance measures for designated
AFOs in order to reflect only the portion of these costs that the AFO owners would ultimately pay.

        The Impact of Gradual Designation on Compliance Costs

        As discussed in Subsection £7 of this paper, we assume that AFOs will be designated gradually
        86 Clemson University. 1998 National Survey of Animal Confinement Policies, covering 35 States.

        87 We can look at this another way. Exhibit III - 6 indicates that the maximum annualized cost of AFO
waste management measures for all the AFOs in those counties with AFO-impaired water bodies would be about
SI 15 million (summing for Type A, B and C AFOs). Only some of these AFOs will ultimately be required to
implement controls (some because they will be designated under the proposed authority, and many more because
the State will require controls of them under State authorities). The tot?,! costs that AFOs in critical areas will
eventually incur and presumably seek financial assistance for will be somewhat less than the maximum of $115
million annualized.-- perhaps $50 million annually, perhaps $75 million annually, perhaps $100 million annually;
we don't really know.  EQIP now provides approximately $65 million annually in assistance to AFOs in critical
areas ($200 million annual appropriation, half going to AFOs, and 65% of that targeted in critical areas). The other
USDA, EPA and State sources of financial assistance to AFOs generally total well over $100 million annually.  It
seems entirely likely that the funds available in critical areas from these other sources, plus the $65 million annually
from EQIP, will amount to significantly more than enough to fund 2/3 of the requests for assistance from AFOs in
critical areas.

118

-------
over the 16-year period from 2000 through 2015. The costs of compliance at the farm level therefore
need to be adjusted downward to reflect the fact that only a fraction of affected AFOs will be designated
in the year 2000. We scale down total compliance costs by the conversion factor of .631745 to obtain
total compliance costs after gradual designation.

       Estimating the Relationship Between the Compliance Costs to be Paid by AFO Owners and AFO
       Revenues

       The 1997 Census of Agriculture provides information on annual revenues by farms of different
sizes raising each of the different sorts of animals. Thus, for example, the Census indicates the total
number of farms in the country with anjnventory of 50 - 99 swine and the number of these small hog
farms with annual revenues falling within each of 13 ranges (i.e, less than $1,000, between $1,000 and
$2,500, etc.). In effect, the Census provides sufficient information to ascertain the distribution of annual
revenues for the farms in each category of AFO (species/size) within which designations may potentially
occur.

       Exhibit 111-1 shows the initial calculations in the small business impact analysis for AFOs that
will receive no cost share funds. The first set of columns shows the annualized compliance costs for an
average farm in each AFO category.  Compliance costs include the annualized cost of the management
measures assumed to be implemented following designation, plus the farm's annualized administrative
cost of being designated.  The cost of the management measures increases with AFO size, while the
administrative cost was estimated in subsection # 8 at an annualized $257.17 for each designated farm.
The management measures costs shown in this exhibit represent the average cost estimated across for all
AFOs in a given category, averaging across the different costs for AFOs in different rainfall regions and
in northern vs. southern states (see subsection # 5).

       The next set of columns in Exhibit III-7 scales typical annualized compliance costs down by the
factor of .631745 to reflect gradual designation of AFOs over the period 2000 through 2015.

       The next set of columns shows the annual revenues that an AFO would need  to have if
annualized compliance costs were to constitute 1 % and 3 % of revenues. For example, a typical swine
farm with 50 - 99 pigs will incur annualized compliance costs of $581  if designated.  These compliance
costs will exceed 1 %  of the farm's revenues if such a farm has annual revenues of less than $58,104.
These compliance costs will exceed 3 % of the farm's annual revenues if the farm has.annual revenues of
less than  $19,368.

       Using data from the Census of Agriculture, we then estimate in the next set of columns the
fraction of farms in each category that have annual revenues less than these 1 % and 3 % break points.
Among the small hog  farms, for example, 61.3 % have annual revenues small enough (< $58,104) so that
compliance costs will  exceed 1 % of revenues, while 29.5 % have revenues small enough (<$19,368) so
that compliance costs  will exceed 3 % of revenues.

       The next column  of Exhibit III-7 shows, again  using Census of Agriculture Data, the percentage
of AFOs in each category that has revenues small enough to meet the definition of small business
                                                                                           119

-------
applicable for the category.88

        The final set of columns in Exhibit III-7 provides, for each type and size category, the percentage
of farms that both 1) meet the definition of small business applicable for the category, and 2) have annual
revenues that fall short of the 1% and 3% breakpoints. We determined this percentage by selecting the
lower of the small business fraction of farms or the sub-breakpoint fraction of farms, depending on the
size of the breakpoint under consideration. For example, 98.0 % of swine farms in the 100 - 199 head
size category have annual revenues less than $500,000, and therefore are defined as small businesses.
According to Census of Agriculture revenue size distribution data, 44.4 % of all farms in this type and
size category have annual revenues less than the 1% breakpoint - $64,340.  Because all farms that have
revenues less than $64,340 must also have revenues less than $500,000, while the reverse is not true, we
select the sub-breakpoint percentage. If we look at dairy farms in the 200 - 499 head category, 26 %
have annual revenues less than $500,000, while 51.0 % of the farms have revenues less than the 1%
breakpoint — $738,060. In this case, we select the small business fraction of farms, because farms that
have revenues less than $500,000 must also have revenues less than $738,060.

        Exhibit III - 8 shows a parallel set of calculations for AFOs that receive cost share funds.  For
these AFOs, we assumed that EQIP-like cost share funds will be available for 75 % of the cost of the
management measures, up to a maximum of $50,000 over ten years.89 We then also added the
administrative costs to the AFO owner of being designated.  We assumed, however, that the AFO owner
would be provided with a nutrient management plan or similar pollution prevention plan at no cost in
connection with the cost sharing (financed either by the State, the AFO owner's integrator if the owner is
a contract grower, or by the cost share agency). Thus, the administrative costs that we added included
only the costs to the AFO owner of hosting the designation site visit and filing the necessary NOIs and
updates for coverage under the presumed AFO general permits (see subsection # 7). The administrative
costs added were $51 annualized for each AFO receiving cost share funds.  Again, after calculating total
compliance cost after cost share, we scale down our estimates to reflect gradual designation of AFOs.
        88 In some cases, the Census of Agriculture does not provide sufficient information to determine exactly
the percentage of AFOs in a category that exceed a compliance cost break point or a small business threshold. For
example, for dairy cow farms, the Census provides information on the distribution of income for farms with greater
than 500 dairy cows.  The Census groups farms with 500 - 999 dairy cows (a size range we are interested in because
some AFOs of this  size will be designated) together with farms with 1000 or more dairy cows (a size range within
which no AFOs will be designated), and presents revenue distribution information for the two sets of farms together
as a category of dairy farms with 500 or more cows. The distribution of revenues specifically for the 500 - 999
dairy cow farms is likely to be generally lower than the distribution of revenues provided in the Census for the
broader group of 500 or more dairy cow farms, but we don't know exactly how much lower. In such cases where
the Census provided information for a category that was aggregated into broader groupings than we wanted, we
used several other sets of Census information to estimate roughly the percentage of AFOs in each category that fell
below the break points arid the small business ceiling.


        89 We annualized this S50,000 limit over 10 years at a 10 % discount rate, to be consistent with the
discount rate inherent in DPRA's cost estimates. $8,137 is the  annualized equivalent of the $50,000 limit over 10
years. Thus, the value of the cost share funds to be received by an AFO (represented in Exhibit III - 8 by the
decrement in moving from column 2 to column 3) is the lesser of: a) 1/4 of the annualized cost of the management
measures; and b) $8,137.

120

-------
                            EXHIBIT III-7
Initial Small Business Impact Calculations for Farms Receiving no Cost-Share
Type/Size
Dairy Farms
20-49
50-99
100-199
200-499
500;999
3ig/hog (aims
50-99
100-199
200-499
500-999
1000-1999
2000-4999
layer (arms
3200-9999
10000-19999
20000-49999
50000-99999
lopooof
Jroiler farms
60000-99999
100000-199999
200000-299999'
30QOOO-499999
500000-749999
Jeef cotv farms
20-49
50-99
100-199
200-499
500-999
Typical Annuallzed
Management
Measure Cost
$T,593
$"2~,83"3
$"5,50i
$11,426
$23,877


$663
$761
$1,209
	 $2',i86
$"4,927
$7.246
368
' $1,106
$3.119
$9.312
$12,951 _;
$1,867
$3,173
$5,996
$11,584
$16.161
$20.377 "
. ..
"$j,40i
$1,367
$3,200
$6.627
' $1~4.263
Typical Annuallzed
Total Compliance Cost
(mgt measure • admin cost]


	 $1,850"
$3.090
$5,759
ffi.683
$24,134


$120
$1,018
$1,466
$2,437
$5,184
$"7,504
$625
$1,363
$3,376
$9,569
_'"' $13.208

$2,124
$3,430
$6,253
$11.841
$16,418
	 $20,634 "


$1.658
$1,624
$3,457
	 $6.884
$14,520 N
Total Compliance Cost
After Gradual
Designation
— 	 - -
$1.169
$1.952
$3,636
$7,381
$15,247


$581
$643
{926
$1,539
$3,275
$4,740
	
$395
$861
"" ""$2,133
	 $6,045
$8.344 	
$i,342"~
$2,167
$1,950
$7,481
$10,372
$13,036

$1,047
$1,026
$1,184
$4.349
$9,173 •
Break Pol
lfcc«1%of
Revenues



$116,861
$195,216
$363,798
$738,060
$1,524,661


$58,104
$64,340
$92,638
$153,939
$327,498
$474,032
	
$39,495
$86,118
"$213,277 "'
$604,528
" "$834,420


$134,193
$216,699
$395.041
$748,060
""$1,037,210
$1.303.553


1104,749
$ToW~
$218,395
$434,907
$917,298
it Revenues
~]fcc^3%of
•Revenues



$38,954
$65,072
$121,266
$246,020
$508,220


$19,368
$21,447
$30.879
$51,313
$109,166
$"158.011
$13,165'
$"28.706
$71.092
$201,509
$278,140


$44,731
$72,233
$131,680
$249,353
"1345,737 ""
"1434.518


$34,916
J3T20T 	
$72,798
$144,969 	
•/. of ail AFOs of (his type/size
with Revenues < Breakpoint
If cc=1% of Revenues



56.0%
43.2%
55.1%
51.0%
16.0%


61.3%
44.4%
31.0%
38.0%
59.0%
.. _....- 	 ._
15.0%
27.4%
34.7%
28.1%
7.3%


24.8%
28.0%
50.0%
60.0%
59.0%
35.0%


75.0%
53.0%
60.0%
55.0%
76.0%
If cc°3% of Revenues



15.0%
5.7%
4.7%
.1.7%
2.1%

1
29.5%
13.9%
5.1%
8.0%
5.9%
5.0%
-— 	 	
0.7%
0.7%
5.6%
7.1%
1.1%


0.1%
0.9%
6.0%
15.0%
13.0%
8.0%


50.0%
40.0%
34.0% „
25.0%
36.0%
Percentage
of farms of
this type/size
which are small
businesses

99.6%
98.7%
88.0%
26.0%
2.0%


98.6%
98.0%
95.0%
77.0%
61.0%
60.0%
	
99.9%
99.0%
98.0%
80.0%
60.0%


98.0%
80.0%
60.0%
40.0%
20.0%
10.0%


99.0%
	 98.0% 	
97.0%
92.0%
64.0%
Percentage of farms that are both
small businesses and have
revenues < breakpoint
If cc=1% of Revenues


56.0%
43.2%
55.1%
26.0% .
2.0%


61.3%
44.4%
31.0%
38.0%
40.0%
59.0%
	
15.0%
27.4%
34.7%
28.1%
7.3%


24.6%
28.0%
50.0%
40.0%
20.0%
10.0%

"" ""755% 	
53.0%
60.0%
55.0%
76.0%
If cc°3% of Revenues


15.0%
5.7%
4.7%
1.7%
2.0%


29.5%
13.9%
5.1%
8.0%
5.9%
5.0%
	
0.7%
0.7%
5.6%
7.1%
1.1%


0.1%
0.9%
6.0%
15.0%
13.0%
8.0%


50.0%
40.0%
34.0%
25.0%
36.0%

-------
                           EXHIBIT III • 8
Initial Small Business Impact Calculations (or Farms Receiving Cost-Share
TypelSiio
Dairy Farms
2049
50-99
lob-199
200499
500-999
I'igfliog hums
00-99
i 00- 199
200-499
500-999
1000-1999
2000-4999
Layer latms
3200-9999
10000-19999
20000-49999
50000-99999
100000'
Oroitor (arms
60000-99999
100000-199999
500000-299999
300000-499999
500000-749999
750000.
Deel cow larms
20-49"
50-99
100-199
200-499
500-999
Typical Annuallted
Tolal Compliance Cos!
(mgl measure • admin cost)
~ $7.850
	 $3.090
$5.759
$11.683
$24.134
$920
$1.018
$1.466
$2,437
I5.J84
$7.504
S625
$1.363
$3,370
$9.569
$13.208

$2.124
$3,430
$6.253
$11,841
$16.418
$20,634

$1,658
	 SJ.624 	
$3,457
$6.884
$14,520
Total Compliance Cost
After Cost Share
$462
$773
St.440
$3,546
$15.997
$230
$255
$367
$609
$i. 296
$i.076


$156
$341
$844
$2.392
$5,071


$531
$858~
$1.563
$3,704
S8.2Bi
$12,497
- ••
$415
$406
$861
\ S1.721
$6,383
Total Cost After
Gradual Designation

$292
$488
$909
$2.240
sio.ioe
$145
$161
$232
$385
$819
SI. 185
$99
$215
$533
$1.511
$3.204

S335
$542
$988
$2.340
$5,232
' $7.895


$262
$257
$546
Si~087
$4,032
Break Point Revenues
llcc=1%of
Revenues

$29,215
$48.804
$90,949
$224,009
$1,010,610


$14,526
$16,085
523.160
$38,485
$81.875
S1 18.508


$9,874
$21,529
$53,319
$151.132
$320.369

$33*548
554,175
$98.760
$234,009
$523,159
$789.502


$26,187
$25.657
$54,599
$108.727
$403,247
llcc=3%ot
Revenues

$9,738
$16.268
$30,316
"$74!670
$336.870


S4.B42
$5,362
$7,720
$12,828
$27,292
$39.503


$3,291
...J7.176
"$17.773
$50,377
$106,790


$11,183
{16,058
$32.920
$78,003
$174,386
$263.167


$8,729
$8.552
$18,200
$36,242
""$134.416 ""
'/.of all AFOi of (hit type/size
with Revenues < Breakpoint
If cc«1% of Revenues

10.0%
1.3%
0.7%
1.4%
8.0%


21.7%
9.0%
2.6%
5.0%
4.4%
2.3%


0.4%
0.2%
2.0%
3.9%
1.6%


0.1%
0.5%
4.0%
14.0%
21.0%
28.0%


42.0%
33.0%
29.0%
20.0%
43.0%
If cc'3% of Revenues

2.0%
0.0%
0.0%
0.0%
1.2%


4.4%
1.1%
0.3%
0.3%
0.4%
0.5%


0.2%
_rj.2%
0.6%
0.4%
0.4%


0.1%
0.0%
0.1%
0.6%
4.0%
4.0%


22.0%
13.0%
15.0%
10.0%
18.0%
Percentage of
farms of this
type/sit* which
are (mall
businesses

99.6%
98.7%
88.0%
26.0%
2.0%


98.6%
98.0%
95.0%
77.0%
61.0%
60.0%

99.9%
_ 99.0%
98.0%
80.0%
60.0%


98.0%
60.0%
60.0%
40.0%
20.0%
10.0%


99.0%
98.0%
97.0%
92.0%
84.0%
Percentage of firms that tre both
small businesses and have
revenuei < breakpoint
II cc°1% of Revenues

10.0%
1.3%
0.7%
1.4%
2.0%


21.7%
9.0%
2.6%
5.0%
4.4%
2.3%


0.4%
_0.2%
2.0%
3.9%
1.6%


0.1%
0.5%
4.0%
14.0%
20.0%
10.0%


42.0%
33.0%
29.0%
20.0%
43.0%
If cc'3% of Revenues


2.0%
0.0%
0.0%
0.0%
1.2%


4.4%
1.1%
0.3%
0.3%
0.4%
0.5%


0.2%
0.2% 	
0.6%
"o".4%
0.4%


0.1% 	
6.0%
0.1%
0.6%
4.0%
4.0%


22.0%
13.0%
15.0%
10.0%
18.0%

-------
       In Exhibit III - 9, we examine the impacts on small entity AFOs that do not receive cost share
funds, under the assumption that EPA will use the proposed designation authority relatively infrequently
(see subsection #  7 for details on the "low use" assumptions). The second column shows the number of
all designated farms for each size category, scaled up by our 1.57 scaling factor to account for farms in
States that did not use 305(b) source codes for AFOs in reporting water body impairments.  The third
column reflects our assumption that 1/3 of these designated farms will not receive cost share funds.  In
the fourth column, we then calculated the number of these designated farms that are small businesses,
based on the small business percentages for each type/size category from Exhibit III - 7.  The set of
columns  in the center of the table contains the percentage of farms that are both small businesses and
whose revenues are less than the 1 % and 3% breakpoints, as calculated in Exhibit III - 7. By applying
these percentages to the scaled up number of designated non-cost share farms, we obtain the number of
these siriall businesses (for each size and type of farm) that would incur compliance costs exceeding 1 %
of their revenues, and the number that would incur compliance costs exceeding 3 % of their revenues.
The final set of columns in Exhibit III - 9 shows the percentage of all affected small businesses in each
type and  size category that have costs exceeding 1% of their revenues, and the percentage that have costs
exceeding 3% of revenues.

       Similar tables follow for AFOs that do receive cost share funds under the "low use" assumption
(Exhibit III -  10, which uses the break point calculations in Exhibit III - 8), for AFOs that do not receive
cost share funds under the "high use" assumption (Exhibit III - 11), and for AFOs that do receive cost
share funds under the "high use" assumption (Exhibit III - 12).
                                                                                            123

-------
                                                                                                       EXHIBIT III -9
                                             SMALL ENTITY IMPACTS ASSUMING LOW USE OF PROPOSED DESIGNATION AUTHORITY - IMPACTS ON FARMS NOT RECEIVING COST-SHARE FUNDS
Type/Size
Dairy Farms
20-49
50-99
100-199
200-499
500-999

Plgmog (arms
50-99~
100199
200499
500-999
roOO-1999
2000-4999

Layer [aims
3200-9999
10000-19999
20000-49999
50006-99999
ioooop*

Orollei hums
600W-99999 _
100000-199999
200000-299999
300000-499999
500000-749999
750000*

Beel cow (aims
20-49
5q:99_
100-199
200-499
500-999
flol designated
farms In
Ihls type/size
category
(scaled up)
4
4
2
13
1
1
1
2
20
20
2
2
• M
4
1
i
is
32
24
28
13
6
4
....
1
13
3
Ftaclion ol
(arms assumed
NOT lo receive
cosl-share
hinds
33.3%
33.3%
33.3%
33.3%
33.3%
313%
33.3%
33.3%
33.3%
33.3%
33.3%
33.3%
33.3%
33.3%
33.3%
	 313%

33.3%
33.3%
".'.33.3%
33.3%
33.3%
33.3%
33.3%
33.3%
33.3%
33.3%
33.3%
Percentage
ol (arms
wtilch are
small
businesses

99.6%
98.7%
88.0%
26.0%
2.0%

"" 98T%"
98.0%
95.0%
77.0%
<>7.o% 	
60.0%
99.9%
99.0%
98.0%
80.0%
60.0%
98.6%
80.0%
60.0%
40.0%
20.6%
10.0%
99.0%
98.0%
97.0%
92.0%
84.0%
tt ol designated
non-CS (arms which are
small businesses

1.5
1.4
0.5
1.2
0.0


0.4
0.5
0.6
5.1
4.0
0.4
0.7
4.6
1.3
0.2
0.2
5.0
8.6
4.8
3.8
0.9
0.2
1.2
0.6
0.3
3.8
6.9
Percenlage ol la
small buslnes
revenues <
II cc=1%of revenues

56.0%
43.2%
55.1%
26.0%
2.0%


61.3%
44.4%
31.0%
38.0%
40.0%
59.0%


15.0%
27.4%
34.7%
28.1%
7.3%

24.8%
28.0%
50.0%
40.0%
20.0%
10.0%


75.0%
53.0%
60.0%
55.0%
76.0%
rms lhal am both
ses and have
breakpoint
II cc=3%of revenues

15.0%
5.7%
4.7%
1.7%
2.0%


29.5%
13.9%
5.1%
8.0%
5.9%
5.0%


0.7%
0.7%
5.6%
7.1%
1.1%


0.1%
0.9%
6.0%
15.0%
13.0%
8.0%


50.0%
40.0%
34.0%
25.0%
36.0%
ft ol designated
non-CosI Share (arms
(scaled up lor nonreporling
delegated slates) whose costs
exceed the 1% breakpoint
and are small businesses

0.8
0.6
0.3
1.2
0.0


0.2
0.2
0.2
2.5
2.6
0.4


0.1
1.3
0.5
0.1
0.0


1.3
3.0
4.0
3.8
0.9
0.2

'
0.9
0.3
0.2
2.3
0.8
#ol designated
non-Cost Share (arms
(scaled up lor nonreporling
delegated slates) whose costs
exceed the 3% breakpoint
and are small businesses

0.2
0.1
0.0
0.1
0.0


0.1
0.1 '
0.0
0.5
0.4
0.0


0.0
0.0
0.1
0.0
0.0


0.0
0.1
0.5
1.4
0.6 S
0.2


0.8
0.2
0.1
1.0
0.4
Percenlage ol all Impacted
small businesses In Ihls type/
size category lhal have costs
exceeding the 1% breakpoint

56.2%
43.8%
62.6%
100.0%
100.0%


62.2%
45.3%
32.6%
49.4%
65.6%
98.3%


15.0%
27.7%
35.4%
35.1%
12.2%


25.3%
35.0%
83.3%
100.0%
100.0%
100.0%


75.8%
54.1%
61.9%
59.8%
90.5%
Percenlage ol all Impacted
small businesses In this type/
size category lhal have costs
exceeding the 3% breakpoint

15.1%
5.8%
5.3%
6.5%
100.0%


29.9%
14.2%
5.4%
10.4%
9.7%
8.3%


0.7%
0.7%
5.7%
8.9%
1.8%


0.1%
1.1%
10.0%
37.5%
65.0%
80.0%


50.5% '
40.8%
35.1%
27.21/.
42.9%
TOTAL
                   234
                                                                 52.5
r    .'."_5i5*'..'."..j

-------
                                                                                                          EXHIBIT III-10
                                               SMAIL ENTITY IMPACTS ASSUMING LOW USE OF PROPOSED DESIGNATION AUTHORITY »IMPACTS ON FARMS THAT DO RECEIVE COST-SHARE FUNDS
  Type/Stie
 Dairy Farms
 20-49
 50-99
 100-199
 200-499'
 566-999

 Pig/hog (amis
 50-99
 100-199
 200-499
 500~-999_7
 1000-1999
 20004999

 layer lams
 3200-9999
 (0000-19999
 20000-49999
 50000-9999~9
 iooooo>".

 3roiler (arms
 60000J9999
 l6oo60-199999
 300000-299939
 .lobboO-499999
 5000od->«999
 750000.

 3ecl cow (aims
 JO-49_
 50-99
 100-199
 200-499
 500-999
| TOTAL
ttol designated
(arms in
Ihis fype/slie
category
(scaled up)
..
4
2 '"
13
i
" i 	
. .
2
20
20
2
2
14
4
1
	 i
is
32
J4
28
1]
6
4
2
1
J3
....
234
Fractal ol
(arms assumed
NOT to receive
coslshare
funds

66.7%
667%
667%
667%
667%

667%
667%
667%
667%
66.7%
66.7%
66.7%
66.7%
66.7%
66.7%
66.7%
66.7%
667%
66.7%
66.7%
66.7%
667%
667V.
66.7%
66.7%
66.7%
66.7%
— 	 -
Peicenlage
ollarms
which are
small
businesses
_99.6%
987%
88.0%
26.0%
2.0%

	 98.6%
98.0%
950%
77.0%
61.0%
600%
999%
99.0%
980%
80.0%
600%
380%
80.0%
600%
40.6%
20.0%
100%

99.0%
98.0%
97.0%
"" 92.6%
84.0%
Kol designated
non-CS (arms which are
small businesses
2.9
2.9
0.9
2.3
0.0

0.8
0.9
1.2
10.1
6.0
08
1.5
91
2.6
0.4"
_os
99
17.3
9.6 '"
7.6
1.7
0.4

2.4" 	
1.2
0.7 .
7.7
17
Percentage o( (aims lhal are bolh
small businesses end have
revenues < breakpoint
II cc=1%ol revenues

10.0%
1.3%
	 0.7% 	
1.4%
20%


21.7%
9.0%
2.6%
5.0%
44%
23%


0.4%
02%
2.0%
3.9% _
1.6%
01%
0.5%
"" 4".0%
tuv." _
200%
10.0%


42.0%
330%
29.0% 	
20.0% 	
13.0%
II cc=3%of revenues

0.2%
0.0%
" 0.6%
0.0%
1.2%


4.4%
1.1%
0.3%
03%
_0.4%
05%


02%
0.2%
0.6%
0.4%
0.4%
"o.i%"
0.0%
"" Jj%
0.6V.
4.0%
4.0%


22.0%
13.0%
15.0% _
10.0%
IB.0%
Dof designated
non Cosl Share (arms
(scaled up Im nonreport'ng
delegated slates) whose costs
exceed Die 1% breakpoint
and are small businesses

0.3
0.0
'" 0.0"
0.1
0.0


0.2
0.1
0.0
0.7
....£«. .
00


0.0
0.0
o.t
0.0
0.6 	

_- .
g.t
6.6
	 !.6
T.7
0.4
	 	 -
1.0
0.4
"02
1.7
0.9
« of designated
non-Cost Share farms
(scaled up for nonreport'ng
delegated stales) whose costs
exceed the 3% breakpoint
and are smaH businesses

0.0
0.0
0.0
0.0
0.0


0.0
0.0
0.0
0.0
0.1
00

00
0.0
0.0
0.0
0.0

0.6" 	
0.0
00
6.1 	
0.3
0.2 	
0.5
0.2
o.V
0.8
0.4
Percentage ol all Impacted
small businesses In Ihls type/
slie category that have costs
exceeding the t% breakpoint

10.0%
	 1.3% 	
0.8%
5.4% .
100.0%


22.0%
9.2%
2.7%
6.5%
7.2%
3.8%

0.4%
0.2%
20%
4.9%
2.7%
0.1%
06%
67%
35.0% 	
io6.0%
100.0%
	
42.4%
33.7%
'~29.9%
}\'.h 	
51.2%
Percentage of all Impacted
small businesses In (Ms type/
slie category that have costs
exceeding Ihe 3% breakpoint

0.2%
0.0%
0.0%"
0.0%
60.0%


4.5%
1.1%
0.3%
0.4%
0.7%
08%

0.2%
0.2%
0.6%
0.5%
	 0.7%
~ ~~ "o.iv. "~
0.0%
0.2%
1.5%
20.0%
40.0%
	 	 	
22.2%
13.3%
_" " !»%'" 	
10.9V.
21.4%


105.?


11.9
2.8 1 11.3% | 2.7%

-------
                                                  EXHIBIT III -11
SMALL ENTITY IMPACTS ASSUMING HIGH USE OF PROPOSED DESIGNATION AUTHORITY •• IMPACTS ON FARMS NOT RECEIVING COST-SHARE FUNDS
Type/Size
Oaiiy Farms
20-49~
50-99
iod-199
200-499
500999

Pig/hog farms
50-99
100-199
200-499
500-999
1000-1999
2000-4999

Layer (arms
3200-9999
10000-19999
20000-49999
50000-99999
1 00000 •
Broiler (arms
60000-99999
100000-199999
200000-299999
300000-499999
500000-749999
750000*

Bed cow farms
20-49
50-99
100-199
266-499
500-999
TOTAL
flol designated
(arms in
Ihis type/size
category
(scaled up)
14
14
.... ..
67
4

5
5
8
98
99
10
11
69
20
4
6
~ 76
162
120
H2
65
30


14
7
4
62
15
1136
Fraction ol
(arms assumed
NOT lo receive
cast-share
(unds
33.3%
33.3%
33.3%
33.3%
33.3%


33.3%
33.3%
33.3%
33.3%
33.3%
33.3%


33.3%
33.3%
33.3%
33.3%
3_3.3%

33.3%
33.3%
33.3%
33.3%
33.3%
33.3%


33.3%
33.3%
33.3%
33.3%
33.3%

Percentage
ol (arms
which are
small
businesses

99.6%
98.7%
88.0%
26.0%
2.0%


98.6%
98.0%
95.0%
77.0%
61.0%
60.0%
	
99.9%
99.0%
98.0%
80.0%
60.0V"
98.0% 	
80.0%
60.0%
40.0%
20.0%
io.o%


99.0%
98.0%
97.0%
92.0%
84.0%
-----
#of designated
non-CS (arms which are
small businesses

4.8
4.5
1.6
5.8
0.0


1.5
1.8
2.4
"25.1 	
20.1
1.9

3.7
22.8
6.5
1.1
1.2
24.8
43.1
24.0
18.9
4.3
1.0

4.5
2.1
1.2
19.1
4.3
Percentage of (arms that are both
small businesses and have
revenues < breakpoint
II cc=1%ol revenues

56.0%
43.2%
55.1%
26.0%
2.0%


61.3%
44.4%
31.0%
38.0%
40.0%
59.0%


15,0%
27.4%
34.7%
28.1%
7.3%

24.8%
28.0%
50.0%
40.0%
20.0%
10.0%

75.0%
53.0%
60.0%
55.0%
76.0%
II ce=3%ol revenues

15.0%
5.7%
4.7%
1.7%
2.0%


29.5%
13.9%
5.1%
8.0%
5.9%
5.0%


0.7%
0.7% 	
5.6%
7.1%
1.1%


0.1%
0.9%
6.0%
15.0%
13.0%
8.0%


50.0%
40.0%
34.0%
25.0%
36.0%

252.3

._ _ ...
dot designated
non-Cost Share (arms
(scaled up lor nonreporting
delegated slates) whose costs
exceed the 1% breakpoint
and are small businesses

2.7
2.0
1.0
5.8
0.0


0.9
0.6
0.8
12.4
13.2
1.9


0.6
6.3
2.3
0.4
0.1


6.3
15.1 . .
20.0
18.9
4.3
1.0


3.4
1.2
0.8
11.4
3.9
Sol designated
. non-Cost Share (arms
(scaled up (or nonreporting
delegated stales) whose costs
exceed the 3% breakpoint
and are small businesses

0.7
0.3
0.1
0.4
0.0


0.5
0.3
0.1
2.6
T.9"
0.2


0.0
0.2
0.4
0.1
0.0


0.0
0.5
2.4
7.1
2.8
0.8


2.3
0.9
0.4
5.2
1.8
137.5 | 	 31.9 	
Percentage ol all Impacted
small businesses In this type'
size category that have costs
exceeding the 1% breakpoint

56.2%
43.8%
62.6%
100.0%
100.0%


62.2%
45.3%
32.6%
49.4%
"" 	 65.6%"
98.3%


15.0%
27.7%
35.4%
35.1%
12.2%


25.3%
35.0%
83.3%
100.0%
100.0%
100.0%


75.8%
54.1%
61.9%
59.8%
90.5%
Percentage ol all Impacted
small businesses In this type/
size category that have costs
exceeding the 3% breakpoint

15.1%
5.8%
5.3%
6.5%
100.0%


29.9%
14.2%
5.4%
10.4%
9.7%
8.3%


0.7%
0.7%
5.7%
8.9%
1.8%


0.1%
1.1%
10.0%
37.5%
65.0%
80,0%


50.5%
40.8%
35.1%
27.2V,
42.9%

54.5% | 1J.71/.

-------
                                                  EXHIBITIIM2
SMALL ENTITY IMPACTS ASSUMING HIGH USE OF PROPOSED DESIGNATION AUTHORITY •• IMPACTS ON FARMS THAT DO RECEIVE COST-SHARE FUNDS
fype/Si;o
Oaiiy Farms
20-49
50-99
iiJO-'J99
200-499
500999
Pig/hog (aims
50-99~
100-199
200-499
500J99"
ioOO-1999
7000.1999
layer (aims
3200-9999
10000-19999
20000-49999
50000-99999
100000*
Droiler (arms
60000-99999
100000-199999
200000-299999
300000-499999
500000.749999
750000^
Jeelcowiarms
20-49
5099
100-199
200.499
500999
TOTAL
0 of designated
farms In
Inis type/slie
category
(scaled up)
"»'
14 	
6
V
4
5
5
8
93
99
io
..
69
20
4
6
76
162
120
i«
65
30
14
7
' 4
62"
IS
tiJS'
Fracfion ol
(arms assumed
NOT io receive
cost-share
funds
66.7%
' 66.7%
66.7%
66.7V.
66.7V.

66.7%
66.7V.
66.7%
66.7%
66.7%
6C7V.
66.7% ""
66.7%
"66.7% 	
667V.
66.7%

66.7%
"66.7%
66.7V.
60 7%
co.)%
66.7%
.. . ...
66.7%
66.7%
66.7%
66.7%
66.7%
Percenlage
ol farms
which are
small
businesses
99.6%
98.7%
890%
26.0%
2.0%

98.6%
98.0%
95.0%
77.0%
61.0%
60.0%
99.9%
99.0%
98.0%
eo.ov.
60.0%

99.0%
80.0%
600V.
400%
20.0%
100%

99.0%
980%
97.0%
920%
84.0%
ffol designated
non-OS farms which are
small businesses
...........
9.1
3.3
11.6
O.I

30
3.5
4.9
50.4
402
	 3.9
7.4
45.7
130
2.1
2.4


49.6
86.4
4B.1
37.0
e.7
2.0

9.0
4.3
U_ _
39.3
8.6
505.4
Percentage of farms lhal are both
small businesses and have
revenues < breakpoint
If cc=1V. of revenues
10.0%
1.3%
0.7%
1.4V.
2.0%

21.7%
9.0% *
2.6%
5.0%
4.4%'
2.3%'" _"

0.4%
0.2%
2.0%
3.9%
1.6%


0.1%
0.5V.
4.0%
140%
20.0%
10.0%


42.0%
33.0V.
29.0%
20.0%
43.0%
If cc=3% ol revenues

0.2% "1
0.0%
0.0%
0.0%
1.2%


4.4%
1.1%
0.3%
0.3%
0.4%
0.5%


02%
0.2%
0.6%
~'6.4%"
"OM


0.1V.
00%
0.1%
0.6%
4.0%
4.0%


22.0%
13.0%
15.0%
16.0%
180%
Sol designated
non-Cost Share lams
(scaled up for nonreporting
delegated stales) whose cosls
exceed the 1% breakpoint
and are small businesses

1.0
0.1
00 — (
0.6
O.I


0.7
0.3
0.1
3.3
29
O.I


00
0.1
0.3
0.1
O.I


0.1
0.5
3.2
13.2
8.7
2.0


38
1.4
0.7
8.3
4.4
It of designated
non-Cost Share (amis
(scaled up for nonreporting
delegated stales) whose cosls
exceed the 3% breakpoint
and are small businesses

0.0
0.0
0.0
0.0
0.0


0.1
0.0
0.0
0.2
0.3
0.0


0.0
0.1
01
0.0
0.0 "~


0.1
0.0
0.1
08
1.7
0.6


2.0
0.6
	 (M 	
4.2
1.8
Percenlage ol all Impacted
small businesses In this type/
slie category that have cosls
exceeding (he IV.Neakpolnl

10.0%
1.3%
0.6%
5.4%
100.0%


22.0%
9.2%
2.7%
6.5%
7.2%
3.8%


0.4%
02%
3.0%
4.9%
2./V.


0.1%
0.6%
6.7%
35.0%
100.0%
100.0%


42.4%
33.7%
29.9%
21.7%
512%
Percenlage ol all Impacted
small businesses In this type/
siie category lhal have costs
exceeding the 3% breakpoint

0.2%
0.0%
0.0%
0.0%
60.0%


4.5%
1.1%
0.3%
0.4%
0.7%
	 0.8% 	

0.2%
0.2%
0.6%
[_ 0.5%
0.7%


0.1%
0.0%
0.2%
1.5%
20.0%
40.0%


22.2%
13.3%
15.5%
" 10.9%""
21.4%



56.3
11* | 11.1% | 2.6%

-------
                                                 Exhibit III • 13: Small Entity Impacts -• Summary
Assuming low use of
proposed designation
authority


Total number
of designated
(arms (scaled up for
nonreporting
delegated states)

234
Number of
designated
farms that are
small businesses
with costs
exceeding 1%
of revenues

40.5


% of designated
small business
farms that
have costs >1%
of revenue

25.7%
Number of
designated
farms that are
small businesses
with costs
exceeding 3%
of revenues

9.6


% of designated
small business
farms that
have costs > 3%
of revenue

6.1%
Assuming high use of
proposed designation
authority

1136 | 193.8 25.6%
                                                                                            45.1
                                                                                     6.0%
# ol small businesses
assuming low use of
proposed designation
authority

# of small businesses
Assuming high use of
proposed designation
authority
157.7
757.7

-------
        Exhibit III - 13 combines and summarizes the results from the four previous tables, covering
AFOs that do and do not receive cost shares under both the "low use" and the "high use" assumptions.
Exhibit III - 13 shows the numbers and percentages of small businesses that are expected to be affected
by the designation provision at levels exceeding 1 % and 3 % of their revenues. These figures provide
the basis for judging whether the rule is presumed to have a significant economic impact on a substantial
number of small entities, and whether EPA is recommended or required to undertake a regulatory
flexibility analysis to consider ways to minimize the proposal's economic impacts on small businesses.

        We project that:

•       Assuming low use of the proposed designation authority, 40.5 AFOs (25.7 % of all affected
        small entities) will eventually bear compliance costs exceeding 1 % of their revenues.  9.6 AFOs
        (6.1 % of all affected small entities) will eventually bear compliance costs exceeding 3 % of
        their revenues. These impacts will eventually occur by 2015. Each year until then, an average of
        2.5 AFOs annually will be affected at the 1 % level and 0.6 AFOs annually will be affected at the
        3 % level.

•       Assuming high use of the proposed designation authority, '193.8 AFOs (25.6 % of all affected
        small entities) will eventually bear compliance costs exceeding 1 % of their revenues.  45.1
        AFOs (6.0 % of all affected small entities) will eventually bear compliance costs exceeding 3 %
        of their revenues. These impacts will eventually occur by 2015.  Each year until then, an average
        of 12.1 AFOs annually will be affected at the 1 % level and 2.8 AFOs annually will be affected
        at the 3 % level.

        These projected impacts on small entities are highly uncertain. Most importantly, there is great
uncertainty in  any estimate of how often EPA will need to use the proposed designation authority and
which particular AFOs EPA  will use it for. As discussed earlier, the proposed designation authority does
not compel any actions by EPA, States or AFOs; it simply authorizes EPA to act at the Agency's
discretion in the (expected to be) rare instances when a State does not develop and implement a TMDL
for an AFO-impaired water.

        On balance, however, we believe that most of the cost-estimating steps in this analysis have been
conservative, in the sense of aiming not to underestimate what the true costs are likely to be. The actual
costs eventually incurred as a result of the proposed Federal designation authority are quite likely to be
less than we have estimated here, for a variety of reasons:

•       We believe that most AFOs, if designated and permitted, will incur compliance costs
        substantially less than what we have projected. Many AFOs already have in place part of all of
        the management measures that we projected as necessary  to meet likely permit requirements.
        Most AFOs now manage their manure as a liquid, and most of them already have a manure
        lagoon and equipment to spread the manure as fertilizer.  We did not estimate any "credit" or
        reduction in incremental compliance costs to reflect this extensive treatment that is already in
        place.

•       We believe that many States, between now and when the bulk of TMDLs will be developed, will
        significantly improve their regulatory and voluntary programs for AFOs. The baseline of State
        AFO programs that we  assumed for costing purposes is the set of State programs as they

                                                                                           129

-------
       currently exist. In fact, States are rapidly improving their AFO programs. Within the last year,
       for example, Maryland and Virginia both adopted major new regulatory programs and financial
       and technical assistance for poultry waste management. In practice, with constantly improving
       State AFO programs over the coming years, EPA will likely need to use the proposed Federal
       designation authority much less frequently than we project in this cost analysis.

•      For the past 20 years or so, the number of small AFOs has been declining and the number of very
       large AFOs has been increasing. We use the 1997 inventory of AFOs as the baseline for
       analysis, assuming, in effect, that the number of small AFOs potentially subject to designation
       during 2000 - 2015 will be the same as the number existing now. However, if recent trends
       continue and the number of small AFOs shrinks in the future, the number of small AFOs
       potentially needing designation in future years will be smaller than what we have assumed.

We also believe that in our small entity impact analysis we have likely understated AFO income, and
thereby overestimated the fraction of AFOs that  will occur compliance cost impacts exceeding 1 % or 3
% of revenues. This stems from shortcomings in the Census of Agriculture data on farm income
distribution.

•      Many AFOs are owned by firms rather than individuals. A firm that owns multiple AFOs or an
       AFO as well as other lines of business will have additional revenues from sources other than the
       single AFO that may have been designated. Our analysis makes a worst case assumption that the
       entity owning the AFO has no revenues  other than those from the AFO. The Census does not
       provide sufficient information to allow us to determine  who the AFO's owner is and how much
       the owner's annual revenues are.

•      The Census reports as income for AFOs only the market value of agricultural products sold by
       the AFO. Several common sources of additional farm revenue to AFO owners are thereby
       excluded, including: revenues from selling animal manure  as fertilizer or feed to nearby crop
       farmers, to manure brokers, or to value-added manure processors; revenues from leasing farm
       land or equipment to others;  revenues from cost share payments, incentive payments, price
       support payments, crop insurance payments, etc..

SUMMARY OF IMPACTS OF PROPOSED AFO DESIGNATION PROVISION

       Exhibit III - 14 summarizes the estimated impacts of the proposed AFO designation provision.
These estimates are uncertain primarily because the provisions  authorize EPA to take designation action
under certain circumstances, but do not mandate specific action. It is very difficult to project how often
States will be unwilling or unable to  develop and implement TMDLs for AFO-impaired waters, and thus
how often EPA will need to consider using the proposed designation authority. It is then further difficult
to predict what EPA will choose to do.

       A large number of assumptions have been made in order to develop low and high estimates of
how often the proposed designation authority is  likely to be used.. For the most part, we believe these
assumptions have been conservative, and the estimated impacts shown below are likely overestimates.
130

-------
                       Exhibit HI- 14
Summary Estimated Impacts of Proposed AFO Designation Provision

Costs of AFO mgmt measures (million $/yr)
Administrative costs to AFOs (million $/yr)
Administrative costs to EPA/States (million S/yr)
Total costs of provision (million $/yr)
Low Assumed
Use
$0.929
$.038
$.034
$1.001
High
Assumed Use
$4.601
$.185
$.157
$4.943

# AFOs designated eventually
# AFOs designated annually
234
15
1,136
71

Small entity AFOs designated eventually
Small entity AFOs w/compliance costs >1 % of
revenues
Small entity AFOs w/compliance costs >3% of
revenues
158
41
(25.7 %)
10
(6.1%)
758
194
(25.6 %)
45
(6.0%)
                                                                    131

-------
     IV. PROPOSED AUTHORITIES REGARDING AQUATIC ANIMAL PRODUCTION
                                       FACILITIES

PROPOSED PROVISIONS ALLOWING EPA TO DESIGNATE SOME AQUATIC ANIMAL PRODUCTION
FACILITIES (AAPFs) AS CONCENTRATED AQUATIC ANIMAL PRODUCTION FACILITIES (CAAPFs),
THUS MAKING THEM SUBJECT TO NPDES

       Under existing NPDES regulations, some AAPFs are defined as CAAPFs and hence point
sources subject to NPDES requirements. An AAPF is necessarily a CAAPF if it contains, grows or
holds:

•      cold water fish or animals in ponds, raceways or similar structures which discharge at least 30
       days per year, but does not include facilities that:

              Produce less than 20,000 pounds of aquatic animals per year; and
              Feed less than 5,000 pounds of food during the calendar month of maximum feeding

•      warm water fish or animals in ponds, raceways or similar structures which discharge at least 30
       days per year, but does not include:

              Closed ponds that discharge only during periods of excess runoff; or
              Facilities that produce less than 100,000 pounds of aquatic animals per year.

In essence, large AAPFs that discharge at least 30 days per year are CAAPFs subject to NPDES
requirements. In addition, under existing regulations, the NPDES Program Director (the State Director in
authorized NPDES program States, and the EPA Regional Administrator in non-authorized States) may
designate an AAPF that does not meet these criteria as a CAAPF on a case-by-case basis if it is
determined to be an actual or potential significant contributor of pollution to waters of the U.S. If such
an AAPF is designated on a case-by-case basis as a CAAPF, a NPDES permit application cannot be
required of the facility until the Director has conducted an on-site inspection of the facility and
determined that it should and could be regulated under NPDES.

       Under existing regulations, in authorized NPDES States, only the State Program Director has this
case-by-case designation authority for AAPFs. EPA now has no such designation .authority in authorized
States.

       The proposed regulation would extend the designation authority in  authorized States to the EPA
Regional Administrator also. EPA's use of the designation authority would be limited to only those
situations where EPA is  establishing a TMDL for 303(d) listed waters. It is intended that EPA will use
the designation authority only  in instances where a 3-03(d) listed water body is impaired in part by
AAPFs, and where for some reason the State has not developed an  approvable (e.g., including reasonable
assurance for implementation) TMDL for the water body.  When EPA then develops a TMDL for the
water body, EPA would  have designation authority over the AAPFs affecting the water body as a means
of enforceably achieving whatever reduction in AAPF loadings is necessary to meet water quality
standards. EPA's use of this authority is expected to be infrequent, for several reasons:

•      States are expected to  develop the great majority of TMDLs for their waters.  Instances where
       EPA must establish part or all of a TMDL in an authorized State have been substantially less

132

-------
       than 20 % of all such TMDLs that have been established thus far,

•      Under Federal NPDES regulations, an authorized State already has this case-by-case CAAPF
       designation authority. The proposed extension of this authority to EPA serves only as a backup,
       to ensure that authority over AAPFs is available to EPA when, for some reason, a State does not
       choose to exercise its authority.

This "backup" nature of the proposed designation authority over AAPFs is quite different from the
designation authority over some silvicultural activities proposed elsewhere in this regulation. The
silviculture provision extends to the States and EPA authority over these sources for the first time,
whereas the AABF provision only extends to EPA an authority that the States already have.

DOES THE PROPOSED REGULATION IMPOSE INCREMENTAL COSTS?

       One might argue as follows that the proposed designation provision imposes no incremental costs
beyond what could be expected in the baseline.  1) The States already have this designation authority
under existing NPDES regulations; 2) States are obligated to do what is necessary to achieve effective,
approvable TMDLs; 3) Therefore now giving this authority also to EPA simply gives EPA the ability to
ensure what the States are already required and able to do, and hence the provision adds no incremental
cost beyond what is already required.

       In order to be conservative in our regulatory cost estimates, however, we will reject this baseline
argument. We will estimate the costs that will be incurred when EPA invokes the proposed authority and
designates AAPFs as CAAPFs, independent of whether some or all of these costs would have been
incurred in the baseline nevertheless.  The remainder of this paper estimates these costs.

       Note that the proposed new designation authority for EPA is applicable only in States that are
authorized to conduct the NPDES program ("authorized" States). In non-authorized States ~ AK, AZ,
ID, ME, MA, NH, NM, and TX — EPA already has the designation authority and the proposed regulation
adds nothing further.90 Only in authorized States may the proposed regulation result in incremental costs.

SUMMARY OF TYPES OF COSTS

       In this paper, we will estimate the following costs attributable to the proposed AAPF designation
provisions:

•      Compliance costs for the AAPFs that may be designated, permitted and required to change their
       practices in various ways that will protect water quality. These compliance costs will include: 1)
       The additional capital and operating costs of the measures these operations will be required to
       implement; and 2) The administrative costs of obtaining and maintaining coverage under NPDES
       permits.
        90 Subsequent to completion of much of this cost analysis, Texas was authorized to conduct the relevant
portions of the NPDES program, and now there are only 7 non-authorized States. There has not been sufficient time
to revise the analysis to reflect this increase in the number of States in which the proposed designation authority will
be applicable.  The costs estimated for 42 States in this report would likely increase by less than 5 % if authorization
for Texas were to be reflected.

                                                                                            133

-------
        Administrative costs to EPA for designating these operations, and to EPA and the States for
        permitting the designated operations. (Note that even if it is EPA that designates an AAPF as a
        CAAPF, it may be the State that ultimately develops the NPDES permit for the CAAPF.)

 In addition to developing national cost estimates for all AAPFs that may be designated, we will also
 provide a rough estimate of the regulatory costs and impacts specifically for the small entity AAPFs that
 may be designated.

        The costs estimated in this paper represent indirect rather than direct costs attributable to the
 proposed AAPF designation provisions. The provisions do not require that any AAPFs be designated,
 nor do they directly require any AAPFs to implement any new practices. The provisions only authorize
 EPA to designate AAPFs as CAAPFs under certain circumstances. The circumstances may not arise, and
 even if they do arise, EPA may determine that it is not necessary to exercise the designation authority.
 Costs  will be incurred only if and when EPA invokes this authority.  This paper presents some relatively
 uncertain estimates of how often EPA may invoke the proposed designation authority, and, if so, the
 costs that will likely ensue.   .

 BRIEF BACKGROUND ON AQUATIC ANIMAL PRODUCTION FACILITIES IN THE U.S.

        The nature of aquatic animal production practices, the water pollution problems they can cause,
 and the measures that might be undertaken to abate this water pollution all vary greatly with the species
 of aquatic animal that is being raised. Exhibit IV - 1 provides some summary information.

        Catfish are the most important aquatic animal production species, accounting for 72 % of total
 U.S. aquatic animal production by pounds and for 55 % by value.  Most catfish is produced in the deep
 South in large (often several acres or more) constructed ponds typically of 3 - 6 feet in depth. Levee
 ponds (filled by pumping from a water source, and which receive only that precipitation which falls on
 the pond surface) are  now much more common than watershed ponds (which are often filled by runoff,
 and often discharge as a result of being overfilled by runoff from the pond's watershed). Under some
 circumstances, discharges from catfish ponds may have substantial quantities of nutrients (mostly from
 unconverted feed), BOD and suspended solids (soil particles, algae, detritus). Pollutant concentrations in
 discharges from the surface of a catfish pond may approximate those in  domestic sewage after
 undergoing primary treatment, but concentrations become much higher when the final 20 % or so of the
 water  in a pond is discharged.

        Several decades ago, catfish farming was conducted in a manner resulting in frequent discharges
 from ponds.  The common harvesting practice was to draw down and discharge much or all of the pond
 water, so  as to concentrate the fish and make them easier to harvest. Now, though, it is common to
 harvest catfish by seining without discharging and lowering the water level in the pond. In addition to
'reducing water pollution problems, full-pond harvest also conserves water, reduces pumping costs, and
 makes the harvest schedule more flexible. Many catfish ponds are now drained only at 5 - 10 year
 intervals, when it is necessary to renovate fish stocks, to repair wave damage to levees, or to remove
 accumulated sediment. Most catfish ponds now discharge infrequently - only when there is extensive
 precipitation or periodic maintenance is scheduled.

        Other species that are commonly raised in fresh water ponds include baitfish, striped bass,
 tilapia and carp. These pond species pose similar water pollution  issues as catfish, and are similarly

 134

-------
trending toward full-pond harvest and infrequent pond discharge. In total, pond-raised species constitute
roughly 65 - 75 % of both the volume and value of all U.S. aquatic animal production.

        Trout are typically raised in raceways — constructed areas in which large volumes of fresh water
flow through pens or holding areas that simulate the flow of a stream. In Idaho where much of the
nation's trout production occurs, water is obtained by diverting a portion of a river or by pumping ground
water. The trout need very clean water, and the effluent from raceways is typically quite clean. Although
the concentration of nutrients, BOD and suspended solids in raceway effluent is very low, the flow
volume is large and some downstream problems have resulted.

        Most other aquatic animal species are grown in situ — in the same waters in which the species
normally live, but either in zones in which the species are concentrated (e.g., concentrated beds of clams,
mussels and oysters) or in large netpens within which salmon, flounder or other fish may be confined.
Occasional water quality problems from these operations typically involve very local impacts from
enhanced feeding, but dilution is usually great and adverse impacts are unlikely even a short distance
away from the feeding operation. In many cases, production of aquatic species in situ involves gathering
naturally occurring species whose number or growth has been somehow enhanced. It is often difficult to
distinguish aquaculture from traditional fishing and shellfish harvesting.

        For several reasons, this paper estimates the costs for use of the proposed Federal CAAPF
designation authority only for pond-raised species.  As noted, pond-raised species constitute roughly 65 -
75 % of both the volume and value of all U.S. aquatic animal production. However, we expect that
virtually all of the costs of the proposed designation provision will be incurred for pond-raised species,
and virtually none will be incurred for non-pond-raised species.

•       There are very few water bodies reported as impaired by aquatic animal production facilities in
        areas where non-pond-raised species are  being produced. The few instances in which States have
        associated water body impairments with non-pond aquatic animal production involve trout, with
        most of these being in Idaho. We are aware of no water body impairments associated with
        species raised in  situ. We have no reason to believe that the proposed Federal designation
        authority would be used for any non-pond AAPFs, except for trout. For trout AAPFs, however,
        the great majority of reported impairments are in Idaho, which is a non-authorized NPDES State.
        EPA already has authority in Idaho to designate an AAPF as a CAAPF if necessary, so the
        proposed regulation will entail no incremental costs there.

•       Quantitative information is extremely limited on the number of non-pond species AAPFs, their
        location and their production.  We would need this information to estimate the potential costs of
        designating non-pond AAPFs. However, because of the smaller volume of production of non-
        pond species and the relatively few such farms in any geographic area, information on these
        farms collected in the Census of Agriculture is very often withheld to prevent disclosure of,
        proprietary information.

•       Even if the proposed Federal designation authority were for some reason used for a non-pond
        AAPF, the compliance measures required in the resulting NPDES permit would likely involve
        only better control of feeding practices and modest resulting costs.
                                                                                            135

-------
                 EXHIBIT IV -1: INFORMATION ON LEADING AQUATIC ANIMAL SPECIES IN U.S.
Species
Non-fond
Cnlfisli
Chiins
Crayfish
Freshwater Prawns
Mussels
Oysters
Salmon
Sin Imp
Trout
Oilier Species
TOTAL
Pounds (1,000) (1997)
19,000
569,579
8,100
46,900
250
COO
s
15,400
33,000
5.800
56,900
37,000
792.2S9
Vnluc ($1,000)
(1997)
56,000
426.800
18,000
27,900 ;.
1.000
1,200
46,700
75,000
6,500
79,800
34,000
772,900
Locution
Scattered
MS, AL,
deep South
New England, Pacific NW,
OnlfofMexico
LA
TX, FL. Ml
New Hnglnnd, Pacific NW,
OnlfofMexico
Pacific NW, New England,
OnlfofMexico. CA
WA.ME
TX, HI. SC
ID, scattered elsewhere
scattered

How Raised?
Ponds, tanks
Ponds
In Situ
Rice paddies, ponds
7
In Situ
In Silu, on bottom or on
suspended structures
Net pens
Flow-through saline ponds
Raceways
Ponds

Control Measures
Limit discharge and provide
settling for discharge that
docs occur
Limit discharge and provide
settling for discharge that
does occur
7
7
7
7
7
Control feeding, prevent
escape of non-native species
Construct ponds above the
100-yenr flood plain, lower
risk of escape of exotic
species, control feeding and
walcr flow
Control feeding
Limit discharge and provide
settling for discharge that
does occur

136

-------
OVERVIEW OF PROCEDURES FOR ESTIMATING COSTS

       We will estimate the compliance and administrative costs of the proposed AAPF designation
authority through a sequence often steps:

1.      Determine the number, location, size and characteristics of AAPFs in the U.S.. This will provide
       the data base from which to begin the analysis.

2.      Focus the analysis on only those "authorized" States that manage the NPDES program
       themselves, including permitting CAAPFs. In non-authorized States, the EPA Regional
       Administrator already has designation authority, and the proposed regulation will have no
       impact.

3.      Determine the locations where AAPFs are likely to need further controls in order for water
       quality standards to be met. This will involve locating the water bodies impaired by AAPFs.

4.      Determine the number, size and characteristics of the AAPFs that are likely to need further
       controls in these areas and for which the designation authority could provide the means of
       requiring the further controls. This will involve extracting from the national data base on AAPFs
       (#1) information on the specific AAPFs in the geographic areas identified in (#3).  We will
       further consider only those AAPFs that can be designated as CAAPFs, excluding those that are
       already CAAPFs.

5.      Determine the performance standards that designated and permitted AAPFs will typically be
       required to meet, the management measures they will respond with, and the per unit costs for
       these measures.

6.      Estimate the maximum possible total national compliance costs by multiplying the number of
       AAPFs that could be designated as CAAPFs (#4) by the estimated unit costs for compliance (#5).

7.      Estimate the frequency with which EPA will likely need to invoke the proposed Federal
       designation authority, the number of AAPFs that may be designated, and the costs attributable to
       the proposed regulation that will result from these  designations. We expect that instances in
       which the further controls on AAPFs needed to meet WQS will be obtained by EPA exercising
       the proposed Federal authority will be relatively infrequent. Most often, States will use their
       own authorities (either independent State authorities, or the State designation authority under
       existing NPDES regulations) to mandate the needed controls over AAPFs. We will assume that
       EPA will need to use the proposed Federal designation authority for 2 - 10 % of the TMDLs
       involving water bodies impaired by aquatic animal production facilities.

8.      Estimate the administrative costs for designated AAPFs to obtain and maintain coverage under
       NPDES permits.  This will be estimated by multiplying the number of operations that may be
       expected to be designated (#7) by an estimate of the administrative cost for a typical AAPF.

9.      Estimate the administrative costs to Federal and State governments for designating and
       permitting these AAPFs. These costs will be estimated by multiplying the number of AAPFs to
       be designated and permitted by an estimate of the  unit cost to perform these activities.


                                                                                          137

-------
10.    Develop estimates of compliance costs and administrative costs for small entities specifically.
       This will be done by disaggregating the estimated costs for all entities developed under #7, #8
       and #9.

Most of this paper will consist of describing the data, calculations, results and uncertainties in each of
these ten steps.

KEY ASSUMPTIONS IN ESTIMATING COSTS

       Estimating the costs of the proposed Federal designation provision is particularly difficult
because the provision authorizes EPA to act, but does not mandate specific actions. No duties are
established for EPA, States, or AAPFs. There is no certain answer to fundamental questions about what
will happen if the proposed regulations are finalized:  What will the States do in developing TMDLs for
AAPF-impaired waters? How often will EPA need to step into accomplish what the States will not have
done? What permit conditions will EPA or the States impose on designated AAPFs?

       As the starting point for estimating the costs of the proposed provisions, we assume that the
designation authority will be implemented as follows:

•      The designation authority will be invoked bv EPA only when EPA must develop a TMDL in an
       authorized  State for a 303(d) water body that is impaired or threatened bv AAPFs. This
       limitation is explicit in the proposed regulation.

•      In the great majority of instances, the State will develop an effective, approvable TMDL for
       SOSCdt waters impaired bv AAPFs. and EPA will not need to use the proposed Federal
       designation authority.  The TMDL program is intended to be implemented by the States, with
       EPA developing TMDLs only in cases of State non-performance. All States have authorities
       sufficient to address AAPFs as necessary for TMDLs — most States have existing authorities and
       programs under State law, and all authorized States have the further option of using the NPDES
       designation authority available under current Federal regulations. EPA will use the proposed
       Federal designation authority only when an authorized State chooses for some reason not to
       perform the TMDL tasks that the State is normally able and expected to perform. Some of the
       likely infrequent reasons why a State might not develop a TMDL and EPA would could include:
       a resource shortfall in the State TMDL program; lack of State authority over AAPFs and
       unwillingness by the State to use the NPDES designation authority available under Federal
       regulations; State unwillingness to take on the political issues  attendant to a controversial
       TMDL; or  perhaps some intractable policy difference between the State and EPA in which EPA
       contends that a State-developed TMDL is inadequate.  In program experience thus far, EPA has
       needed to establish the TMDL in less than 20 % of the instances in which a TMDL has been
       undertaken. Based on this experience and the fact that all authorized States already have
       authority under NPDES regulations to designate AAPFs as CAAPFs as necessary, we will
       assume that EPA will need to establish a TMDL for 2 - 10 % of the water bodies impaired bv
       AAPFs.
138

-------
•      In a geographic area where EPA must develop the TMDL and will use the proposed designation
       authority, we assume that EPA will designate as point sources subject to NPDES all the AAPFs
       in that area.

ESTIMATED COSTS OF THE PROPOSED AAPF DESIGNATION PROVISION

       In this subsection, we describe the ten steps resulting in estimates for the different varieties of
costs attributable to the proposed regulation.

1. Determine the number, location, size and characteristics of AAPFs in the U.S.

       The Census of Agriculture provides the most comprehensive data available on AAPFs. The most
recent Census, from 1997, provides data at the national, State and county level. The Census provides
information on six categories of aquatic animal, including: catfish, trout, hybrid striped bass, crawfish,
"other fish" (a residual category for all fish not addressed separately), and "other aquaculture products"
(another residual category consisting largely of aquatic plants, but also including alligators, eels, etc.).
For this analysis focusing on pond-raised species, we used Census information on catfish, hybrid striped
bass, and other fish and assumed conservatively that all such fish are raised only in ponds.

       Census data for these three categories offish that were assumed to be pond-raised include: the
number of pounds offish raised, the number offish raised, the dollar value of annual sales, and the
number of farms that provided responses for each of these data categories. Unfortunately, there are many
gaps in the Census aquatic animal production data, in large part because quantitative data is often
withheld to prevent disclosure of proprietary information. Based on the Census data that was provided,
we used the following decision rules to impute missing data at the county and State levels:

County level:

•      Number of farms: We used the number of farms that reported sales data (in all cases, an equal or
       greater number of farms reported sales data than reported the pounds offish they produced).

       Pounds offish: We scaled up the number of pounds offish reported by the farms that reported
       production in proportion to the ratio of the number of farms reporting sales data to the number of
       farms reporting pounds of fish.                                       .

•      The Census often reported data for an entire State, and then county-level data for only the most
       important counties in the State.  When we sought data on a non-reported county, we assumed that
       it included less aquatic animal production than any county in the State for which data were
       reported. We assumed that any non-reported  county included fish production and farms equal to
       half of that in the county in the State with the lowest reported production and farms,   s

       When the Census provided no county-level data at all for a State, we assumed that each affected
       county had 1 farm (i.e. the minimum number of farms that could cause a water quality
       impairment), and would scale the State's total number of pounds to a proportional amount for
       one farm, based on the total number of farms in the State.
                                                                                           139

-------
State level:

       As with counties, it appeared that all farms reported sales data while many fewer reported
       production data. We therefore assumed that the number of farms reporting sales data was the
       total number of farms in the State.  The total pounds produced in the State was estimated by
       scaling the pounds reported by those farms reporting pounds. The scaling factor was the ratio
       between the number of farms reporting sales and the number of farms reporting production.

       Bass: If the Census reported the number of bass farms for a State, but not pounds, we used the
       average number of pounds per farm found in the "other States" category (i.e. States which were
       grouped together and not listed individually in the Agriculture Census U.S.-level table). Where
       the Census did not provide the number of bass farms for a given State, we determined the number
       of states in the "other states" category (37), and assigned each State l/37th (38,000 pounds) of
       the number of pounds for those "other states". The exceptions to this rule were Alaska and New
       England States (cold weather states which are unlikely to have much, if any, bass production),
       "which together shared a single State portion (again, 38,000 pounds). In each case we assumed
       that each State lacking bass data had only one farm, given the small number of pounds each was
       assigned.

•      Catfish: For those States that provided no farm data, we determined the number of States in the
       "other states" category (19, only one of which was a warm weather State and was therefore more
       likely to have catfish farms), and assigned each State lacking data an equal fraction of one states'
       portion of the number of pounds offish raised in those "other states". The exception to this rule
       was New Mexico (the one warm weather state that lacked data), which received a full "other
       state" share of 52,000 pounds. In each case we assumed that these states each had only one farm,
       given the small number of pounds each was assigned.

       Other fish: Departing from our general rule of thumb,  we did not scale up the number of pounds
       of fish in the "other state" grouping according to the number of farms reporting sales.  The
       "other states" figure for number of pounds, even unsealed, makes up a far larger percentage of
       total U.S.  pounds than the other states' percentage of total sales.  Therefore, for states not
       providing data on number of pounds, we simply used the data provided for pounds in "other
       states" (i.e. without scaling the data by the ratio of the number of farms reporting sales data to
       the number of farms reporting pounds offish), and divided that by the number of other states (7)
       to assign pounds (1,484,000 each). We assigned 8 farms to each of these states - the average of
       the number of farms reported by States that did not report their pounds figures in this category.

2. Focus the analysis on only those "authorized" States that manage the NPDES program
themselves, including permitting CAAFFs

       In non-authorized NPDES States, EPA already has the designation authority, and the proposed
new authority will have no  impact. These non-authorized States where EPA manages the NPDES
program are Alaska, Arizona, Idaho, Maine, Massachusetts, New Hampshire, New Mexico and Texas.
These States were dropped from further analysis.  If one of these States gains authority to conduct the
NPDES program,  the applicability of the proposed EPA designation authority will expand, and the costs
140

-------
of the proposed rule will potentially increase beyond what we have estimated."

       Our exclusive focus on authorized States is important with respect to our being unable to cost out
non-pond species.  The one place we know there is substantial impairment of water bodies due to non-
pond aquatic animal production is Idaho, where many large trout farms are located. Idaho is a non-
authorized State, so the provision will entail no costs there. A substantial amount of the production of
other non-pond species also takes place in non-authorized States: salmon in Maine, shrimp in Texas, and
clams/mussels/oysters in Maine and Massachusetts.  So, our omitting non-pond species misses perhaps
25 - 35 % of total U.S. aquatic animal production,  but much of this omitted production occurs in non-
authorized States and will not be affected by the regulation anyway.

3. Determine the locations where AAPFs are likely to need further controls in order for water
quality standards to be met.

       This involved locating the water bodies impaired by AAPFs that will need to have TMDLs
developed for them. We then assumed (in the absence of any more precise method for identifying the
spc•;•  c AAPFs that will need further controls) that all AAPFs in any county (or counties) in which an
AAi'j -impaired water body is located would need further controls in order to achieve WQS.  Most of the
identified AAPF-impaired water bodies are relatively small (typical length of the impaired river segments
was 5 to25 miles), and our assumption that the entire surrounding county's worth of AAPFs would need
to be controlled will usually result in a substantial  overestimate of compliance costs. However,  some
water bodies are larger and seriously impaired; for them it is possible that more than the surrounding
county's worth of AAPFs would need to be controlled.

       We  began with a search of the 1996 and 1998 data bases of 303(d) waters for water bodies with
aquatic animal production facilities cited as the cause of impairment. The search was performed by Tetra
Tech, Inc. Such waters were reported for only 3 States in the partly completed 1998 data base, and for 6
States in the 1996 data base.  We judged that coverage in these 303(d) data bases was substantially
incomplete — some States are not included in the data bases at all, and many States provide either no
information  on the source types responsible for impairment of the listed waters or information organized
at too high a level of aggregation to allow identification of AAPFs (e.g., many States cite impairment as
due generally to "agricultural nonpoint sources" without further information that allows us to determine
whether the  cause more specifically might be row  crops or grazing or AAPFs, for example.)

       Seemingly more complete and detailed information on sources of impairment is available from
State reporting under section 305(b) than from reporting for 303(d).  In order to identify and locate water
bodies that were impaired specifically by AAPFs,  we consulted EPA's 1996 National Assessment
Database. The database was created to provide a uniform and consistent method for States to report
water quality information in their 305(b) reports.  In this database, 9  States reported having water bodies
that were impaired by sources related to AAPFs. About twice as many water bodies were reported as
impaired by AAPFs in the 305(b) data base as were reported in the 303(d) data bases. Based on the
apparently more complete reporting of water bodies impaired by AAPFs in the 305(b) data base relative
to the 303(d) data base, we decided to use the National Assessment Database as  our point of departure.
This choice  of the 305(b) data base may result in some upward bias in our cost estimates, as States'
303(d) lists  of the waters that will need to have TMDLs developed for them (and thus the waters where
        91 As has occurred recently for Texas, a development that is not reflected in this analysis.

                                                                                           141

-------
the designation authority might potentially be used) are significantly shorter than their lists of threatened
or impaired waters reported under 305(b).

       For each water body listed as impaired by AAPFs, the 1996 National Assessment Database
contains a data field indicating the source magnitude — the extent to which AAPFs contributed to the
impairment of the water body. For each affected water body, AAPFs may be listed as a "major",
"moderate", or "minor" source of impairment  We decided to analyze only water bodies for which
AAPFs constitute a major or moderate source of impairment. Water bodies where AAPFs are only a
minor source of impairment are impacted more significantly by one or several other sources of
impairment. For these water bodies, addressing the AAPF contribution to impairment would not fully
eliminate the impairment, and would likely be of much lower priority in a TMDL than addressing the
major or moderate sources of impairment. For these waters, we assumed that AAPFs (if they need to be
addressed at all) are likely to be addressed by means that are less onerous than designating them as point
sources; perhaps they will be addressed sufficiently through voluntary programs or non-permit State
regulatory programs. Accordingly, we assumed that the proposed designation provision is unlikely to be
used in connection with these "minor AAPF impairment" water bodies, and excluded these water bodies
from our initial list of affected water bodies.

       A total of 24 water bodies in 8 States were listed as having major or moderate impairments due
to AAPFs.  We then reviewed our States to exclude water bodies from any non-authorized States among
the 8. None of the States were non-authorized, so all 24 waterbodies were included in our list of AAPF-
impaired waters in authorized States for which the proposed regulation might have some impact.

       For each of these 24 water bodies, we wanted to identify the AAPFs that were responsible for .
their impairment and that would need to be controlled in order to meet WQS.  There were two significant
difficulties in doing so, however. First, the finest level of geographic resolution at which AAPF activity
data (see Step #1) is reported is the county level. Second, the finest level of geographic resolution at
which we were able to locate identify individual water bodies in the time available was similarly at the
county level.92 With more time, we might be able to develop a more precise location for each water body
and then delineate the boundaries of the sub-watershed within which AAPF-related loadings contribute to
impairment of the water body. However, the difficulty of obtaining AAPF activity data on a sub-county
level would likely still present an obstacle to pinpointing the exact amount of AAPF activity that
contributes to impairment. For this analysis, the best we could attempt to do is identify a single county
(or counties) within which the AAPF-impaired water body is located, and assume that all the AAPFs
within that county (or counties) would need to be controlled in order to eliminate the impairment. We
assumed that the maximum possible use of designation authority would be to address all the AAPFs
located within every county in which there is an AAPF-impaired water body.

       We identified the county in which each AAPF-impaired water body is located by using 3 sources
of information. First, for each waterbody in the database, there is an associated 8-digit "HUC". or
       92 The data base printout we received identified the geographic location of an AAPF-impaired water body
only by indicating the "HUC" (watershed code) within which the AAPF-impaired water body is located and the
counties covered by the HUC.  HUCs often span 10 or more counties. Thus, the data base printout allowed us to
locate an impaired water body  only to the extent of knowing that it is somewhere within a group of counties. EPA
is currently developing a database that will map the latitude and longitude of impaired water bodies, allowing the
user to identify the location of these water bodies more precisely.

142

-------
watershed code. Using other information provided by EPA and the Research Triangle Institute, we
identified the counties covered by each watershed, thereby narrowing the range of possible counties in
which each water body might be.93 Second, we compared the set of possible counties for each water
body with results from searching the U.S. Geological Survey's National Atlas of the U.S.
(http://www-atlas.usgs.gov/scripts/start.htmn forthe nara e of the w aterbody. If thess StBps d±i notiESolt
in identifying the single county within which the particular impaired water body was located, we took the
third step of consulting State maps.  In cases where these three steps did not narrow down the choice of
possible counties in which a water body was located to one, we included all possible counties for that
waterbody in our initial list of counties to consider.

        Out of the 24 AAPF-impaired water bodies, we identified a single individual county associated
with each of 20 water bodies. For an additional 4 water bodies, we were able to narrow the list of
possible counties to two. In total, we identified 23 counties as being associated with the 24 water bodies.

                                         Exhibit IV-2
   Number of Reported AAPF-impaired Water Bodies and Corresponding Counties  (Authorized
                                             States)

Water bodies
Water bodies
in one county
in two counties
Total
Water
Bodies
20
4
24
Number of
Corresponding
Counties
17
6 more
23
        By including all counties in our final list, including those counties associated with "2-county"
waterbodies, we may be overestimating the number of counties that will be affected by the provision.
While it is possible that AAPFs from adjoining counties may impact a particular water body if it lies near
their shared county border or if the impaired water body crosses a county line, thereby implicating
AAPFs in two counties, we believe this to be the exception rather than the rule.  Generally speaking, we
would expect for the large majority of cases that those AAPFs that are actually causing impairment of a
particular listed water body are located within a single county. If so, the actual number of counties in
which AAPF-impaired water bodies are located is roughly equal to the 17 "1-county" counties, plus one-
half of the 6 "2-county" counties. This rough estimate approach yields a total of 20, and probably should
be inflated somewhat to account for those exceptional cases where AAPFs in more than one county
impact a given water body. Therefore, our final estimate of 23 affected counties probably represents a
slight overestimate.

        In addition to the 9 authorized  States reporting AAPF-impaired waters (8 of which have AAPF-
impaired waters showing major or moderate impairment), there are 33 authorized States which did not
        93 William Cooter of RTI provided us with a copy of the database and performed the initial-queries to
identify AAPF-impaired water bodies, the watersheds in which they were located, and the counties included in these
watersheds.
                                                                                             143

-------
report any AAPF-impaired waters in the 305(b) data base.  We do not believe this is because there
actually are no AAPF-impaired waters in these 33 States. Instead, in many cases, we believe this is
because these States did not report the cause of impairment in their waters at all or did not report causes,
of impairment in a manner that afforded any possibility of citing "AAPFs" as a cause of impairment.94
We believe the costs we estimate associated with use of the proposed Federal CAAPF designation
authority in the 9 authorized States for which we have data should be scaled up to account for likely costs
in the additional 33 authorized States for which we do not have data on AAPF-related impairments. We
will assume conservatively that the 33 States that did not report any impairments actually do have them
but simply did not report. This is an additional assumption that likely causes us to overestimate the costs
of the proposed designation provision.

        We calculated a scale-up factor to extrapolate from authorized States that reported AAPF-
impaired water bodies to the authorized States that did not. We began by obtaining (and, where
necessary, estimating — see step # 1) data for each of the 9 reporting authorized States and each of the 8
non-authorized States on the number of pounds of catfish, bass and "other fish" raised.  Using the
Agriculture Census' national-level figures for the total number of pounds of catfish, bass, and "other
fish" raised in the U.S., we calculated the number of pounds raised for the remaining States ~ i.e. the 33
non-reporting authorized States. The 33 authorized non-reporting States accounted for 47 % as many
pounds of catfish, bass, and "other fish" as the 9 States that reported.  We therefore used a factor of 1.47
to scale up the costs associated with controls on AAPFs in the 9 reporting States to obtain a national cost
estimate covering all 42 authorized States.  We apply this scale-up at the end of our cost analysis.
        94 Many States use EPA's recommended system of "major source category codes" in reporting causes of
impaired water for their 305(b) reports, but to varying degrees. In EPA's system, "agriculture" (cause code 1000) is
a major category that is used by 38 States. Aquatic animal production facilities (cause code 1700) is a more detailed
subcategory within the agriculture category.  Only 12 States used cause code 1700, and it was this code for which
the 305(b) data base was searched for aquatic animal production-related impairments.  Some States did not report
causes of impairment at all, some reported based on only very gross categories (e.g., point sources, nonpoint
sources), and some used substantially different classification systems.

144

-------
                                          Exhibit IV-3
   Calculations to Determine Scale-Up Factor to Account for Non-reporting Authorized States
                                                  Catfish       Bass
                                                  Pounds   |   Pounds
                                                   (1000)    I   (1000)
                        Other fish
                         Pounds
                          (1000)
                        All fish
                        Pounds
                         (1000)
Authorized states that reported
AL
72,040
 38
  66
72,144
CA
 2,565
 212
10,957
 13,734
IL
 1,094
 38
 374
 1,506
LA
51,864
 38
 188
 52,090
MS
583,861
 318
 873
385,053
MT
               38
             1,484
             1,531
OH
  177
 265
  56
  498
TN
  377
 38
 1,352
 1,767
WV
  238
 38
 312
  588
Total
512,225
1,023
15,662
528,910
Non-authorized states
AK
               10
             1,484
             1,502
AZ
  225
 38
 1,484
 1,747
ID
               38
              990
             1,037
ME
               10
             13,502   I    13,520
MA
               10
             1,484    !    1,502
NH
               10
             1,484
             1,502
NM
  52
 38
 1,484
 1,574
TX
 4,408
 212
 1,451
 6,071
Total
 4,728
 364
23,363
 28,455
US Total
682,694
11,104
109,698
803,495
Totals for 9 authorized states that reported
             528,910
Totals for 8 non-authorized states
             28,455
Remainder (total for 33 non-reporting authorized
states)    	
             246,130
Totals from all 42 authorized states (reporting and
non-reporting)   	
             775,040
Scale-up factor from 9 authorized states that
reported to all 42 authorized states
               1.47
                                                                                               145

-------
4. Determine the number, size and characteristics of the AAPFs that are likely to need further
controls in these areas and for which the designation authority could provide the means of
requiring the further controls.

       This involved two steps. First, we extracted from the Census of Agriculture (Step #1)
information on all the AAPFs located in the counties within which there are AAPF-impaired water
bodies (Step #3). Exhibit IV-4 provides, for each reporting authorized State, the number of catfish, bass
and "other fish" farms, and the number of pounds offish raised on those farms, in counties with
waterbodies that have been impaired to a major or moderate extent by aquatic animal production
facilities. It should be noted that, due to the paucity of county-level data reported in the Census of
Agriculture, nearly all county-level figures on the number of farms and the number of pounds have been
estimated using the data imputation procedures discussed in step # 1 (the only exceptions are data on
catfish farms and pounds associated with the counties in Mississippi and with  Lawrence County,
Tennessee). We then applied the 1.47 scale-up factor calculated in Exhibit IV-3 to estimate the total
number of farms (259) and pounds of pond-raised fish (332,113,000) to which the designation authority
might be applied in all the authorized States.
                ..                       ExhibitIV-4
             AAPFs That Potentially Could Be Designated, and Pounds of Fish Raised
• • • 1
State County
AL Marshall
CA Mono
IL Johnson
Williamson
LA Iberville .
LaFourche
Tanqipahoa
MS Bolivar
Coahoma
Leflore
Sunflower
Washington
Yazoo
OH Paulding
TN : Davidson
Fayette
Hardeman
Lauderdale
Lawrence
: McNairy
Shelby
WV Fayette
Pendleton
r-,«-,c.h • Catfish
Catfish i Pounds
Farms i (1000)
1 Bass
Bass • Pounds
Farms i (1000)
2 ! 1 i
2 i 12
1 : 61
1 ! 61
1 : 147
1 : 147
1 I 147
8 16,104
7 1,512
15 : 54,681
59 i 90,395
16 i 44,452
8 16,850
1 ; 15
1 : 1
1 : 1
1 1
1 .1
3 3
1 1
T 1
1 . 34
1 34
1 : 313
1 40
1 i 40
1 I 64
1 64
1 ! 64
1 ! 40
1 40
1 ; 40
1 ! 40
' 1 • 40
1 ! 40
1 15


•




1 2
1 2
Other j Other fish
fish ! Pounds
Farms ! (1000)
1 : 2
1 72
All fish
Farms
3
j
All fish
Pounds
(1000)
3
397
1 34 !
1 ; 34
1 I 14
1 • 14
1 i 14
6


9
270


675
2 ! 16 !
2 ' 16
2 ' 16
2 i 16
2 i 16 .
2 I 16
1 i 1
1 ! 20
1 : 20
1 i 20
1 ; 20
1 ' 20
1 • 20
1 ' 20


Total for reporting authorized States
Total for all authorized States




131
o






16

4
176
. 259




224,330
31






149

72
225,927
332.113
146

-------
       Second, we considered which of these AAPFs would not potentially be subject to being
designated as CAAPFs because they are so large as to be CAAPFs already. As for AFOs, some large
AAPFs may already be CAAPFs under existing regulatory definitions, and the proposed designation
authority will have no impact on them.  As for AFOs again, though, there are important definitional
uncertainties. Under existing regulations, an AAPF for warm water aquatic animals (e.g., catfish, bass,
minnows, tilapia) is a CAAPF if the animals are held in ponds or other structures which discharge at
least 30 days per year, except that CAAPFs do not include:

       1. Closed ponds which discharge only during periods of excess runoff; or

       2. Facilities which produce less than approximately 100,000 pounds of aquatic animals per year.

       We will assume that none of the pond AAPFs that we are analyzing are already CAAPFs, and
hence the proposed designation authority could potentially be applied to all of them. Many catfish farms
produce well over 100,000 pounds per year, and thus could be CAAPFs already (i.e., they do not avoid
being CAAPFs by meeting criterion #2). These catfish farms, however, generally use ponds from which
there are intentional discharges typically at most once a year when the water may be partially drawn
down for seining the fish, and once every 3-10 years when the pond is drained for maintenance.  If
criterion #1 is interpreted literally, the one intentional discharge per year or so is sufficient for them also
to fail to meet this possible means of not being a CAAPF.  If a catfish farm fails to meet both criterion #1
and criterion #2, the question then becomes whether the farm discharges at least 30 days per year. We
believe that most of them do not.  Except for the once-a-year drawdown, they will only discharge due to
overflows from storms. Most catfish ponds have some freeboard (i.e., they can accommodate some
rainfall without overflowing), and many will often not discharge to waters of the U.S. even when they
overflow (e.g., their overflow will be channeled to an unused pond, or the overflow will be pumped to
fields and used for irrigation). So, we believe that very few catfish farms will discharge at least 30 days
per year, and so very few are already CAAPFs.

5. Determine the performance standards that designated and permitted AAPFs will typically be
required to meet, the management measures they will respond with, and the per unit costs for these
measures.

       There is  currently no effluent guideline or other national standard specifying the minimum level
of performance expected of CAAPFs. EPA is now beginning studies to characterize the aquatic animal
production industry,  its impacts on water quality and potentially applicable control measures.  These
studies will inform future decisions on the need to regulate wastewater discharges from the industry.
Until these studies are completed, there is a great deal of uncertainty about what performance might be
expected to be required of permitted aquatic animal production facilities.

       Assumed control measures:

       We will  assume that the performance standard that designated and permitted CAAPFs will need
to meet is to provide 6 hours of detention time for all overflow or drainage from fish ponds.95  This will
        95 This goal is recommended in: Claude E. Boyd and Craig S. Tucker. Pond Aquacuhure Water Quality:
Management. 1997. Page 574.
                                                                                           147

-------
provide for thorough settling of TSS. The supernatant from the detention basin can then be pumped to
nearby agricultural fields for irrigation or returned to a nearby water body without impairing water
quality. The collected sediment should be cleaned out of the detention basin periodically and land-
spread or disposed.  The detention basin should be sized to handle a 25 year/6 hour storm.

        Further assumptions:

        We make a series of assumptions in order to estimate the costs for a typical pond-based AAPF to
meet this performance standard of providing at least 6 hours of detention time for all overflow (up to and
including a 25 year storm) and drainage from the pond.
                                                                                           X
•       We assume the average fish pond size to be 2 hectares. Annual fish production from such a pond
        might be about 26,400 pounds.  (A typical harvest rate is approximately 5.000 - 6,000 kg/hectare.
        Source: Boyd and Tucker, ibid.)

•       We assume that a detention basin will be sited to serve an average of two fish ponds.

•       We assume the farm is near the Gulf Coast, where the 25 year/6 hour storm is about 6 inches of
        rainfall. (This provides a nearly worst case nationally, though it is typical for most of the
        nation's catfish production, which occurs predominantly in the deep South.)

•       We assume that the fish ponds at the AAPF are levee ponds rather than watershed ponds, as is
        typical of most catfish production.  The ponds thus receive no runoff from surrounding land,
        only precipitation directly on their surface.

We next calculate the size of the detention basin needed to serve the two 2-hectare fish ponds producing
about 26,400 pounds offish annually.

•       The detention basin must be large enough to hold the greater of: 1) the 25 year/6 hour storm; or
        2) 6 hours worth of pond drawdown when the ponds are being drained.

•       (1) is greater than (2), and thus the storm establishes the basin design criterion.

               1. The 25 year/6 hour storm would discharge to the detention basin in a 6-hour period
               the equivalent of 6 inches of water covering an area of 4 hectares (the two ponds, each
               with surface area of two hectares).

               2. Typical drawdown of a pond will occur over 3-4  days. Assuming a typical pond
               depth of 58 inches, the rate of drawdown will be approximately 58 inches over 72 - 96
               hours from one 2-hectare pond (assume only one pond will be drained at a time). This
               rate is 3.6 - 4.8 inches of water drawn from the 2 hectare pond over six hours, or the
               equivalent of 1.8 - 2.4 inches of water if drawn from a 4 hectare area over six hours.

        The rate of discharge from draining a pond  to the detention basin is thus only 30 - 40 %  as great
        as the discharge rate associated with the storm event.

•       To hold the 25  year/6 hour storm, the basin volume must be 4 hectares x 6 inches, or about 21.5

148

-------
       x!04cuft.

       Cost estimate to construct operate and maintain such a detention basin:

       We have adapted figures provided in: DPRA, Incorporated. Economic Impact Analysis of
Coastal Zone Management Measures Affecting Confined Animal Facilities. October 7, 1992. This work
was performed for EPA/OW's Nonpoint Source Control Branch in the course of developing the
Agency's estimate for the costs of the CZARA management measures. We also used this reference in
estimating the costs of the proposed AFO  designation authority in the previous section of this report.

       The DPRA study developed cost estimates for a lined retention pond, associated pumping and
irrigation equipment, and annual O&M expenses for the basin and irrigation system. DPRA  estimated
these costs for many different sized ponds, none of which are as large as 21.5 x 10* cu ft.  We used
regression analysis on DPRA's data to establish relationships between pond volume and cost for both
construction and the irrigation equipment, and used these relationships to estimate the capital cost of the
25 x 10" cu ft pond. We estimate this pond will cost:

               pond cost                    $70,359
               equipment cost               $ 6,798
               total capital cost              $77,157

DPRA estimated the annual O&M costs for a pond and irrigation equipment to be 5 % of capital costs.
O&M costs for the 21.5 x 10" cu ft pond would thus be $3,858 annually.

       DPRA estimated a pond and equipment lifetime of 20 years. At a 7 % discount rate, the
corresponding capital recovery factor is .09439.  Thus the annualized capital cost for the 21.5 x 104 cu ft
pond would be $7,283.  The annualized total cost would be $11,141.

       DPRA's cost estimates are expressed in 1991 dollars.  Inflating to 1999 dollars (using the
producer price index) we get an annualized cost of $12,043.

Expressing the cost for these management measures on a production basis:

       The detention pond and associated equipment costing an annualized $12,043 •will serve two 2-
hectare fish ponds. The two fish ponds will yield approximately 52,800 pounds offish annually. The
cost for the management measures is thus $12,043/52,800 Ibs, or $0.2281  per pound offish produced.
We will apply this cost to the amount offish production from each farm producing pond-based species in
order to estimate the cost of the management measures for each farm.

6. Estimate the maximum possible total national costs for management measures by multiplying
the number of AAPFs that could be designated as CAAPFs (#4) by the estimated unit costs for the
management measures (#5).

       In step #4, we identified 176 AAPFs for pond-based species in the 23 counties within which
were located the 24 water bodies that authorized States (8 of them) reported as major or moderately
AAPF-impaired.  These 176 AAPFs raised a total of 225,927,000 pounds per year of pond-raised fish.
In order to account for the 33 authorized States which did not report any aquatic animal production-

                                                                                          149

-------
related impairment, we scaled up our estimates by a factor of 1.47 to obtain a national estimate of 259
AAPFs that might potentially be designated, producing 332,113,000 pounds of pond-raised fish annually.
In step #5, we developed a unit cost for management measures for pond-raised species — $0.2281 per
pound offish production. To calculate the cost of management measures if all potentially designatable
AAPFs were designated, we multiplied the unit cost by the number of pounds of pond-raised fish
produced in all authorized States — both reporting and non-reporting — and obtained a maximum
possible annualized cost for management measures of $75.7 million per year.

7. Estimate the frequency with which EPA will likely need to invoke the proposed Federal
designation authority, the number of AAPFs that may be designated, and the management
measure costs that will result from these designations.

       The management measure cost estimates in step #6 reflect costs that would be incurred if EPA
invoked the proposed Federal designation authority for all AAPFs that produce pond-raised fish in
authorized States. However, we expect that EPA will need to use the proposed authority for only a small
fraction of these AAPFs. Most often, States will develop their own TMDLs and use their own authorities
(either independent State authorities, or the State designation authority under existing NPDES
regulations) to mandate any controls over AAPFs needed to implement the TMDLs.  As noted earlier,
EPA has needed to establish the TMDL in less than 20 % of the instances in which a TMDL has been
undertaken.  Based on this experience and the fact that all authorized States already have authority under
NPDES regulations to designate AAPFs as CAAPFs as necessary to implement TMDLs, we will assume
that EPA will need to establish a TMDL and use the proposed designation authority for only 2 - 10 % of
the water bodies impaired by AAPFs.

       In step #6, we estimated that there are 259 potentially designatable AAPFs. If all of these
AAPFs were designated, the management measures they would  implement would cost $75.7  million
annualized. Based on our assumption that EPA eventually will develop 2 -10 % of the TMDLs for
AAPF-impaired waters, we estimate that 5 to 26 AAPFs might be designated by EPA, resulting in costs
for management measures of $1.51 - $7.57 million annually.

       This $1.51 - $7.57 million per year estimate reflects the costs that would occur if all EPA
designations of AAPFs were to occur in  the first year following  promulgation of the proposed
designation authority (assumed to be the year 2000). Alternatively, this estimate represents the
annualized management measure costs once all the AAPFs that will be designated by EPA have been
designated. However, all EPA designations under the proposed authority will not be completed
immediately in the year 2000. We expect that the TMDLs for AAPF-impaired waters will be developed
at an even pace over the years from 2000 through 2015, and some portion of these TMDLs will require
the proposed Federal designation authority to be used in order to implement the TMDLs. A best estimate
is that the 5 to 26 expected Federal AAPF designations will also be spread evenly over these  16 years.
The present value costs for  most designated AAPFs to implement the required management measures
will be reduced because designation of many of them will occur later than 2000.

       The cost impact of this gradual rather than immediate designation is discussed at length in the
silviculture section of this report (see  Section II, Subsection 5).  In the earlier section, we developed a
scaling factor of 0.631745 to reflect gradual designation of sources over 16 years rather than
immediately. We apply this scaling factor to the S 1.51 - $7.57 million per year estimated cost to AAPFs
assuming immediate designation, and arrive at a final annualized cost estimate of $0.95 - $4.78 million

150

-------
per year.
                                         Exhibit IV - 5
          Annualized Cost of Management Measures Resulting from Federal Designation
                          (With Gradual Designation Over 2000 - 2015)

Cost of management measures for designated AAPFs
(million $/yr)
Number of AAPFs that
ultimately will be designated
Low
Estimate
0.95
5
High
Estimate
4.78
26
8. Estimate the administrative costs for designated AAPFs to obtain and maintain coverage under
NPDES permits.

       This will be estimated by multiplying the number of AAPFs that may be expected to be
designated (5 to 26, as estimated in #7) by an estimate of the administrative cost for a typical AAPF.

       AAPFs that are candidates for designation will incur the administrative costs of being considered
for designation and then, if they are designated, will incur the costs to seek and maintain coverage under
NPD.ES permits. We assume that an AAPF that is a candidate for designation will need to:

•      Host a site visit by an EPA representative to help determine if the operation is an AAPF and
       whether it meets the criteria for designation.  As for AFOs, we estimate that roughly 2 hours will
       be sufficient for the AAPF owner or manager to assemble the information and materials that will
       be requested during the site visit, and that another 2 hours will be necessary to host the actual site
       visit.

Assuming that the AAPF is then designated as a CAAPF, the owner or his representative will then need
to:

•      Apply for an NPDES permit. We assume that the permit application form for an AAPF will be
       substantially streamlined relative to the existing NPDES application forms for industrial
       dischargers, and that the AAPF application will require limited information similar to that
       typically required by a Notice of Intent (NOI) for coverage under a general permit.  The required
       information would likely include identification of the applicant, a description of the aquatic
       animal production activity and its location and volume, site information, an estimate of water use
       and water discharges, information on potentially affected water bodies, and perhaps more.  VVe
       will assume that the effort required to prepare such an application will be the same as that for the
       storm water NOI — estimated at 2 hours per NOI in EPA's ICRfor Notice of Intent and
       Stormwater Pollution Prevention Plan for Discharges Associated with Industrial Activity
       (March, 1999).

•      Prepare some sort of pollution prevention plan, indicating how the performance standards and
       management measures required by the permit will be translated into specific facilities.
                                                                                           151

-------
        equipment and BMPs that will be employed in the applicant's operations.  We have not
        researched further details on what such a plan might contain. Some indications could probably
        be obtained by reviewing the requirements in States that already implement permit programs for
        AAPFs.  Absent further information, we will assume that the burden associated with preparing a
        pollution prevention plan would be similar to that for the Storm water Pollution Prevention Plans
        required under the stormwater general permit program. This one-time burden is estimated in the
        relevant ICR at 80 hours per plan.

•       Provide additional updated information whenever there are substantial changes in the AAPF's
        operations (e.g., change in ownership). Also, some sort of reporting whenever there are large or
        exceptional discharges may be required under the pejmit.  We assume that the burden for these
        additional filings or reports will average 2 hours annually.

        Assuming that the pollution prevention plan remains valid for an average of 10 years, the
administrative burden  over a  10-year period to a designated and permitted AAPF would thus be:

        86 hours in the first year, consisting of 4 hours to host the site visit, 2 hours to file the NOI, and
        80 hours to prepare the pollution prevention plan;

•       2 hours in each of the remaining nine years for filing new information and discharge reports.

This burden is equivalent to 84 hours in the first year, plus 2 hours annually over each of the 10 years.
Annualizing the first year burden of 84 hours (by applying a capital recovery factor of .1424 for a 10-
year asset life at a 7 % discount rate), we obtain 11.96 hours annually.  The total annualized burden
estimated for a designated and permitted AAPF is thus 13.96 hours.

        To these burden estimates, we apply a costing factor of $36.12 per hour for fully loaded private
sector labor costs, drawn from the recent ICR for Applications for the NPDES Discharge Permit and the
Sewage Sludge Management Permit (December, 1998).

        The total estimated annualized administrative cost to a designated and permitted AAPF is thus
$504.24. Applying this per-AAPF figure to the 5 to 26 AAPFs to be designated (step #7), we estimate
annualized administrative costs for these designated sources of $2,521 - $13,110 per year. This is the
estimated cost if all AAPFs were designated immediately.  To reflect .gradual designation over 2000 -
2015, we multiply by the scaling factor of 0.631745.  The result is an estimated annualized
administrative cost to designated AAPFs of $1,593 - $8,282 per year.

9. Estimate the administrative costs to Federal and State governments for designating and
permitting these AAPFs.
                                                /
        In using the proposed designation authority, EPA will incur administrative costs to make the
required site inspection for all potentially designatable AAPFs, and EPA or the States will then incur
costs subsequently to designate and permit the chosen AAPFs.

        We assume that a site inspection of an AAPF will  require an EPA regional office staff person to
spend '/z day, including travel time to the  site.  We assume a staff inspector can visit 2 AAPFs in a  day.
AAPFs will typically be some distance away from the EPA regional  office responsible for the

152

-------
inspections. We assume an average travel cost of $100 for transportation and $100 for lodging and per
diem for a typical day spent inspecting 2 AAPFs. To inspect 2 AAPFs will thus require 8 hours of staff
time (or $253.44, priced at $31.68/hr for fully loaded State or Federal labor costs, drawn from the recent
ICRfor Applications for the NPDES Discharge Permit and the Sewage Sludge Management Permit
[December, 1998]) plus $200 in travel costs, for a total of $453.44. The one-time inspection cost per
AAPF will be half of this, or $226.72 per AAPF.

       We assume that designation of an inspected AAPF that is to become a CAAPF, review of its
permit application and issuance of a NPDES permit will be managed in a coordinated manner by EPA
and the relevant State. We assume that these three steps together will require 80 hours per designated
and permitted CAAPF. After permits are issued to CAAPFs, we assume that 20 % of tfee permits will
need modification or reissuance annually, at a cost of 8 hours per modification.

       To these burden estimates, we apply a costing factor of $31.68 per hour for fully loaded State or
Federal labor costs, drawn from the recent ICRfor Applications for the NPDES Discharge Permit and
the Sewage Sludge Management Permit (December,  1998).

       The total estimated administrative cost for EPA and the States for inspection, designation and
permitting of CAAPFs as a result of the proposed authority is thus:

•      One-time costs for inspection                               $226.72 per CAAPF

•      One-time costs for designation, application review and NPDES permit issuance:
                                                                $2,534.40 per CAAPF

•      Annual costs for permit modifications and reissuance:          $50.69 per CAAPF

       Converting the one-time costs to annualized costs by assuming that a permitted CAAPF's status
is thoroughly reviewed every 10 years and applying a discount rate of 7 % (resulting in a capital recovery
factor of 0.1424), we obtain an annualized cost for inspection, designation and permitting of $393.18 per
CAAPF. The summed annualized costs for all activities are $443.87 per CAAPF.

        Applying the per-CAAPF cost for Federal and State activities to the estimated 5-26 AAPFs to
be designated gives an annualized cost of $2,219 - $11,541 per year. This is the-estimated cost if all
AAPFs were designated immediately. To reflect gradual designation over 2000 - 2015, we multiply by
the scaling factor of 0.631745. The result is an estimated annualized administrative cost to the Federal
and State governments of $1,402 - $7,291 per year.

10. Develop estimates of compliance costs and administrative costs for small entities specifically.
 X
       This will be done by disaggregating the estimated costs for the 5-26 AAPFs developed under
#7, #8 and #9.

       The Small  Business Administration has established the small business size standard for
aquaculture operations (SIC code 0273) at $500.000 in annual revenues, and the standard for fish
hatcheries (SIC 0921) at $3 million in annual revenues.  Of the AAPFs in the 23 counties within which
                                                                                         153

-------
AAPF-impaired water bodies were reported, approximately 61 % appear to be large businesses.96
Extrapolating this 39 % small business fraction of AAPFs in these States for which we have data to the 5
- 26 AAPFs projected to be designated nationwide, we estimate that 2 - 10 small entity AAPFs may be
affected by the proposed Federal designation provision.  The projected cost of the management measures
for an AAPF that is designated is $2218 per pound offish produced while AAPFs' revenues from fish
sales are typically one to several dollars per pound.  It appears that all (100 %) of the 2-10 affected
small entity AAPFs will incur compliance costs exceeding 1 % and also exceeding 3 % of their revenues.

SUMMARY OF IMPACTS OF PROPOSED AAPF DESIGNATION PROVISION

       Exhibit IV - 6 summarizes the estimated impacts of the proposed aquatic animal production
facility designation provision.  These estimates are uncertain primarily because the provisions authorize
EPA to take designation action under certain circumstances, but do not mandate specific action. It is
very difficult to project how often States will be unwilling or unable to develop and implement TMDLs
for AAPF-impaired waters, and thus how often EPA will need to consider using the proposed designation
authority. It is then further difficult to predict what EPA will choose to do.

       Further uncertainty results from our very limited information on aquaculture. A large portion of
the Census of Agriculture data on aquaculture farms, production and receipts is withheld to prevent
disclosure of proprietary information, thus requiring us to impute data extensively. Our choice of the
management measures that designated AAPFs will need to implement and our procedure for estimating
their costs are speculative.  EPA currently has very little information on aquatic animal production, and
in the limited time available we could obtain little additional information from published literature and
other sources. However, EPA will soon begin a study of the aquaculture industry to support a decision
on whether or not to regulate the industry further. On completing the study, EPA will know much more.

       A large number of assumptions have been made in order to develop low and high estimates of
how often the proposed designation authority is likely to be used and what costs will likely ensue. For
the most part, we believe these assumptions have been conservative, and the estimated impacts shown
below are likely overestimates.
       96 This estimate is derived from reviewing the data in Exhibit IV - 4 on the amount of production at the
AAPFs in the 23 counties. We assume that all of these AAPFs are classed in the aquaculture SIC rather than the
fish hatchery SIC, and the applicable small business size standard is thus $500,000 in annual revenues. Catfish
farms typically realize roughly SI per pound for their fish produced, while bass and other fish farms typically
realize more. A catfish farm thus will need annual production of 500,000 pounds or more to be a large business,
while other fish farms will need somewhat lesser production, in the range of 200,000 - 400,000 pounds. Exhibit IV
- 4 shows thgj the average catfish farms in five of the six Mississippi counties (all except Coahoma) have production
well in excess of that needed to be a large business. Some but not all of the catfish farms in Coahoma County, MS,
may be large businesses, and the bass farm in Mono County, CA may be a large business.  The remainder of the
farms appear to be small businesses. The number of large businesses is approximately 107 (106 among the five MS
counties, and 1  in CA), which represents 61 % of the 176 total farms.
       This estimate is quite rough. Some of the AAPFs in the MS counties where the average farm is a large
business may nevertheless be small businesses. On the other hand, some of the AAPFs that have insufficient
production to make them a large business by virtue of their own revenues are likely owned by corporate entities that
are large businesses by virtue of owning multiple farms or having other revenues.

154

-------
                                Exhibit IV - 6
Summary Estimated Impacts of Proposed Aquatic Animal Production Facility Provision

Management measure costs to sources (million $/yr)
Administrative costs to sources (million $/yr)
Administrative costs to EPA/States (million $/yr)
Total costs of provisions (million $/yr)
Low Estimate
$0.95
$.002
$.001
$0.95
High Estimate
$4.78
$.008
$.007
$4.80

# AAPFs designated eventually
# AAPPS designated annually
5
0.3
26
1.6

Small entity AAPFs designated eventually
Small entity AAPFs w/ compliance costs >1% of
revenues
Small entity AAPFs w/ compliance costs >3% of
revenues
2
2
(100%)
2
(100%)
10
10
(100%)
10
(100%)
                                                                              155

-------
ATTACHMENT 1: SURVEY OF OFFSET ACTIVITIES

-------
                                SUMMARY OF OFFSET ACTIVITIES
        This attachment summarizes nine instances around the nation in which a point source discharger has
sought or will seek to offset the plant's new or expanding discharge by achieving offsetting, pollution
reduction measures elsewhere.  The offset activities summarized here are distinguished from a more extensive
set of trading activities that generally involve a set of sources broader than those seeking to offset new or
expanded discharges. The offset experience is most directly relevant to the proposed NPDES  rule.1

        The nine activities summarized in this appendix have occurred or are being planned in six states and
five U.S. EPA regions.  The activities are in various stages of implementation: two have actually begun to
implement some of the offsetting activities; two have been issued permits but have not yet implemented the
offsets; one will be issued a draft permit in September 1999; one program (comprising three separate activities)
is in the final planning stages before permit issuance and implementation; and the remaining three are still
being studied and discussed.

        A broader list of effluent trading  programs underway or under development is included as part of this
appendix.  The broader list identifies 36 activities in twenty states (nine EPA regions). Of the 36 programs,
twelve have actually executed trades or offsets.

        The offset summaries in this appendix are grouped by U.S. EPA region.  The summary of each offset
includes brief notes covering: nature of the activity, environmental problem, pollutant(s) or pollution type(s),
trade type, stage of implementation, relation to TMDL, number of potential participants, trading ratios,
estimated cost savings, available written information, innovative aspects, obstacles, related web sites and
contact information.

                                          Table of Contents

Summary of 36 Trading Programs, Including Offsets
Wayland Business Center Treatment Plant Permit, MA
Town of Acton Municipal Treatment Plant. MA
Specialty Mineral Company in Town of Adams. MA
New York City Watershed Phosphorus Offset Pilot Program, NY
Delaware River Pilot Trading Program, PA
Rahr Malting Permit, MN
Southern Minnesota Beet Sugar Cooperative Trading Program, MN
San Francisco Bay Mercury Offset Program. CA
Truckee River Program Water Rights and Pollution Offset Program, NV
Page
2
3
4
5
'6
7
8
9
10
11
        1 This attachment is excerpted from a report summarizing a broader set of trading and offset programs. "A
Review of the Nation's Effluent Trading and Offset Experience." Prepared by Environomics for U.S. EPA Office
of Water, 1999 (forthcoming).
                                                                            Attachment 1, Page 1  of 11

-------
                      Summary of 36 Trading Activities, Including Offsets
Project
Long Island Sound
Acton (Assabet)
Adams (Hoosic)
Wayland (Sudbury)
Passaic Valley
NYC Watershed
Blue Plains WWTP
Maryland Program
Chesapeake Bay
Delaware River
Henry County (Smith)
Virginia Program
Tampa Bay
Neuse River Basin
Tar-Pamlico Basin
Illinois Program
Kalamazoo (Allegan)
Michigan Program
Rahr Malting
SMBC
Clermont County
Fox-Wolf Rivers
Red Cedar River
Rock River
Wisconsin Program
Cargill- Ajinomoto
Bear Creek
Boulder Creek
Cherry Creek
Clear Creek
Lake Dillon
Lake Chatfield
Grassland Basin
San Francisco Bay
Truckee River
Lower Boise
State
CT
MA
MA
MA
NJ
NY
DC
MD
muiti
PA
VA
VA
FL
NC
NC
IL
Ml
Ml
MN
MN
OH
Wl
Wl
Wl
Wl
IA
CO
CO
CO
CO
CO
CO
CA
CA
NV
ID
Region
1
1
1
1
2
2
3
3
3
3
3
3
4
4
4
5
5
5
5
5
5
5
5
5
5
7
8
8
8
8
8
8
9
9
9
Activity
Watershed-wide program
Offset for 1 discharger
Offset for 1 discharger
Offset for 1 discharger
Pretreatment program
3 Pilot programs
Interstate credit creation
State-wide program
Watershed-wide program
Basin pilot program
POTW permit flexibility
State-wide program
Regional cooperation
Basin-wide program
Basin-wide program
Pretreatment program
Watershed pilot program
State-wide program
Offset for 1 discharger
Offset for 1 discharger
Regional project
Watershed pilot program
Watershed pilot program
Watershed pilot program
State-wide pilots
NPDES permit flexibility
Basin-wide program
Area-wide offsets
Basin-wide program
Clean-up orphaned sites
Basin-wide program
Study & possible trading
Regional trading program
Regional offset program
Offset via water rights
10 Basin-wide program
Stage
Under development
Under development
Under development
Permit Written
Implementation
Final Planning
Near Implementation
Concept paper
initial development
Early Discussion
Trade Approved
Initial development
Implementation
Initial implementation
Implementation
Under development
Implementation
Under development
Implementation
Early Implementation
Early discussion
Implementation
Implementation
Under development
Implementation
Implementation
Early development
Implementation
Implementation
Discontinued .
Implementation
Study completed
Initial Implementation
Under development
Initial implementation
Under development
Trades/ Offsets Arranged?
Permits Written?
N
N
N
Y
Y
Y
N
N
N
N
Y
N
Cooperative efforts underway
N
Credits banked but not used
N
Y
N
Y
N
N
One trade identified
2 cities pursuing trades
N
Pilots underway
Y
N
Y
Y
N
Y
N
Y
N
Partially
N
Attachment 1. Pa°e 2 of 11

-------
                          WA\XAND BUSINESS CENTER TREATMENT PLANT PERMIT (MA)

Nature of Activity: The Wayland Business Center, an office building complex, sought to discharge into the Sudbury River.
The facility had been owned by the Raytheon Corporation, and the new owners originally sought to renew (or roll over) the
existing permit. The Massachusetts Department of Environmental Protection and the U.S. EPA ruled that the Raytheon
permit could not be rolled over to the new owners of the Wayland Business Center, and hence the facility's discharge was to
be construed as a new discharge.  As a condition for allowing the discharge, the NPDES permit specified that the facility
needed to obtain an offset for its discharge.  The facility will obtain offsets by linking faulty septic systems on 24
neighboring properties to the facility's WWTP.

Environmental Problem: General water quality and eutrophication of Sudbury River.

Pollutant(s) / Pollution ^Type(s): Phosphorous.

Trade Types: Point/ Nonpoint.

"'age of Implementation: The process began in early 1998; the permit was issued in September '98. The municipality is
 -.rently finalizing plans for the sewer hook-ups.

Relation to TMDL: Parts of the Sudbury River are 303(d) listed for metals; parts of neighboring rivers are listed for
nutrients.  The offset does not have a direct relation to a TMDL, but the actions will result in an 80% decrease in phosphorus
loadings to the Sudbury.

Number of Potential Participants: The Wayland Business Center, the Town of Wayland and 24 property owners with
septic tanks.

Trading Ratios: The  facility is permitted to discharge 0.125 pounds per day of phosphorous, and must reduce loadings via
septic tank connections by at least 0.375 ppd, thus the ratio is 3:1.

Estimated Cost Savings: The facility will save approximately $700,000. The municipality will avoid potential significant
future capital investments that would be necessary if the sewerage system would eventually need to be extended from the
municipal POTW to the 24 septic tanks.

Available Written Information: NPDES permit available.

Innovative Aspects: A clause in the NPDES permit specifying the offset provision and the septic tank sewerage is believed
to be the first of its kind.  The municipality will eventually assume responsibility for the business center's WWTP,
effectively making it a POTW. A contingency plan is specified if the sewer connection option cannot be implemented,
including the possibility of harvesting excess algal growth in the river.

Obstacles: The discharger did not have authority to perform sewerage work so the municipality had to play a role in
implementing the effort that had not been anticipated.

Web-sites: Concord: http:/Avww.epa.gov/surf2/hucs/01070005/
http://ww\v.sudburwallevtrusfees.org/

Contact: Jane Downing, EPA Region 1. (617)918-1571. downing.ianef2jepa.gov
                                                                                       Attachment 1, Page 3 of 11

-------
                             TOWN OF ACTON MUNICIPAL TREATMENT PLANT (MA)

Nature of Activity: The Town of Acton needs a new POTW, however new discharges into the Assabet River are normally
prohibited. A plan has been proposed that would allow the new discharge if the municipality were to obtain offsets via non
point source reductions. One possibility for obtaining the offsets is connecting failing septic systems.

Environmental Problem: Eutrophication in the Assabet River.

Pollutant(s) / Pollution Type(s): Phosphorous.

Trade Types: Point/ Nonpoint.

Stage of Implementation: Study and discussions are ongoing now.                        ^

Relation to TMDL: Portions of the Assabet River, including the segment from Acton to the confluence downstream with
the Sudbury and Concord Rivers are 303(d) listed for nutrients, organic enrichment / DO, and pathogens. A TMDL is under
development.

Number of Potential Participants: The Town of Acton,  multiple nonpomt sources and federal, state and local agencies.

Trading Ratios: A preliminary target of 3:1 has been  proposed.

Estimated Cost Savings: N/A

Available Written Information: Study underway.

Innovative Aspects: The Town is using a very detailed GIS to pinpoint the sources of phosphorus loadings within the
watershed that have the greatest impact on water quality.  Title V of the State of Massachusetts' Environmental Code has
stringent requirements regarding sales of homes with faulty septic systems, thus the septic system sewerage also provides the
opportunity to increase the real estate value of many homes.

Obstacles: The project has had difficulty in identifying sufficient nonpoint sources to offset the entire planned discharge
from the new POTW.

Web-sites: Concord: http://www.epa.gOv/surf2/hucs/01070005/
http://www.ultranet.com/~oar/

Contact: Jane Downing, EPA Region 1. (617)918-1571, down ing. janefSjepa.gov
Attachment 1, Page 4 of 11

-------
                            SPECIALTY MINERAL COMPANY IN TOWN OF ADAMS (MA)

Nature of Activity: The Specialty Minerals Company is seeking to increase its high temperature discharge to the Hoosic
River.  One possible plan is for the company to help finance in-stream temperature reduction measures upstream in an Army
Corps of Engineers flood control project.

Environmental Problem: There are cold water fish in the Hoosic River which would be at risk due to increased temperature
from an expanded discharge.

Pollutant(s) / Pollution Type(s): Temperature.

Trade Types: Point / Nonpoint

Stage of Implementation: A draft NPDES permit is expected in September 1999.

Relation to TMDL: Portions of the Hoosic are 303(d) listed for pathogens, siltation, suspended solids and priority organics,
however these are not related to the offset program.

Number of Potential Participants: Specialty Minerals, the Town of Adams, environmental groups, the Army Corps of
Engineers and other federal agencies.

Trading Ratios: 2:1

Estimated Cost Savings: The  company would avoid an estimated capital cost of S300,000 for installing effluent chillers at
its plant

Available Written Information: A draft NPDES permit is expected in September 1999.

Innovative Aspects: The project is possibly the first offset or trade involving the Army Corps of Engineers. The company
would contribute to a local matching share of costs to modify the upstream USAGE project so as to reduce background water
temperature. The municipality was committed to contributing a share of the funds before Specialty Minerals became
involved. Since the Town is not wealthy and since the plant is one of the major employers, the project can also be viewed as
a form of economic development.

Obstacles: There are multiple  responsible parties implementing various actions, which makes coordination difficult.

Web-sites: Hudson/Hoosic: http://www.epa.gov/surf2/hucs/02020003/

Contact: Jane Downing, EPA  Region 1 (617)918-1571, downins.iane@.epa.gov
                                                                                     Attachment 1, Page 5 of 11

-------
                   NEW YORK CITY WATERSHED PHOSPHOROUS OFFSET PILOT PROGRAMS (NY)

Nature of Activity: New York City is implementing a pilot program to allow new/expanding WWTPs to purchase effects
for increased phosphorus discharges in phosphorous restricted basins.

Environmental Problem: Eutrophication in watershed.

Pollutant(s) / Pollution Type(s): Phosphorus.

Trade Types: Point/ Nonpoint.

Stage of Implementation: The watershed protection agreement was signed in January of 1997; the guidance for
phosphorous offsets was published in March '97; the rules and regulations for watershed protection were published in May
'97. A maximum of three pilot projects East of the Hudson River Watershed, with a total volume of 150,000 gallons per day,
as well as three pilots West of the Hudson Watershed, with a total flow of 100,000 gpd may be included in the pilot program.
Three conceptual offset plans were accepted and are now in the final, detailed planning stages. Once implemented, these
projects will serve as the basis for deciding whether to implement a permanent program.

Relation to TMDL: There are phased TMDLs being developed for the East of Hudson and West of Hudson regions. The
TMDLs make expansion or new discharge very difficult without an offset.

Number of Potential Participants: The pilot stage is  limited to three  projects on each side of the Hudson and by total
gallons per year. There were five applicants, three of which were accepted.  One of the approvals was revoked when the
applicant failed to secure the necessary approval from the town or county. New York City, New York State, the U.S. EPA,
and approximately 80 localities were party to the watershed protection agreement

Trading Ratios: 3:1

Estimated Cost Savings: N/A.

Available Written Information: The published rule, guidance and participation selection criteria are available from NYC
DEP.

Innovative Aspects: Three offsets have been allowed: two retrofits of storm water controls and one street cleaning program.
The NYC DEP requested conceptual rather than detailed offset proposals so as to minimize the preparation costs and
encourage proposals.

Obstacles: The program was negotiated as part of a very broad watershed protection agreernent-between NYC, NYS, U.S.
EPA and dozens of localities.

Web-sites: Lower Hudson: http://www.epa.gov/surf2/hucs/02030101 /
htrp://www.ci.nvc.nv.us<1itml/dep/html/watershed.html

Contact: James Benson, New York City Department of Environmental Protection. (9141742-2034
Attachment 1, Page 6 of 11

-------
                                   DELAWARE RIVER PILOT TRADING PROGRAM

 Nature of Activity: U.S. EPA and the Pennsylvania Department of Environmental Protection are sponsoring the creation of
 a pilot trading program in Pennsylvania. A potential site for the trade would be along the Delaware River in West Fall
 Township, PA. West Fall is planning a significant expansion of its POTW. The Delaware River Basin Commission has
 designated the area as Special Protection Waters, thus any new or significantly expanding discharger cannot effect a
 "measurable change" in existing water quality. Moreover, new or significantly expanding POTWs are prohibited from
 discharging until they have exhausted "load reduction" options. The West Fall POTW may seek offsets from a POTW in
 Port Jervis, NY for projected increases in multiple pollutant discharges.

 Environmental Problem: The area has been classified by the DRBC as having "exceptionally high scenic,
 recreational, and ecological values." The Special Protection Waters Designation is aimed at protecting the area's water
 quality and places very stringent restrictions on new discharges.
                1
 Pollutant(s) / Pollution Type(s): multiple

' Trade Types: The trade could potentially involve point /point and point / nonpoint reductions.

 Stage of Implementation: The DRBC created the Special Protection Waters classification in 1992 and amended them in
 1994. The program is in the early stages of planning. The recently formed group has met several times during the Spring
 and Summer of 1999.

 Relation to TMDL: A phased TMDL is being developed for portions of the Delaware River Estuary for toxics. The trading
 program is not directly related.

 Number of Potential Participants: POTWs of West Fall and Port Jervis, NY DEP and PA DEP, U.S. EPA, DRBC.

 Trading Ratios: Ratios would be decided case by case and pollutant by pollutant.

 Estimated Cost Savings: -

 Available Written Information: The DRBC Water Quality Regulations are available from the DRBC.

 Innovative Aspects:  The trade would be designed to offset both the POTW's direct discharge, as well as any new, indirect
 discharge such as storm water. Moreover, the trade would be designed to offset a broad set of pollutants, rather than just one
 pollutant or class of pollutants. Finally, the trade would be an interstate trade.

 Obstacles: Coordinating between the various State, Federal, and Basin regulatory agencies over a' large set of pollutants may
 prove difficult.

 Web-sites: http:/Avww.epa.gov/surf2/hucs/02040104/
 http://www.epa.gov/surf2/ahr/lO/
 http://www.state.ni.us/drbc/drbc.htm

 Contact: Charles Marshall, Phifip Services Corporation. (215) 643-5466
                                                                                        Attachment 1, Page 7 of 11

-------
                                          RAHR MALTING PERMIT (MN)

 Nature of Activity: Rahr Malting, an existing facility on the Minnesota River, had been an indirect discharger to a local
 POTW, but calculated that there were business advantages and cost savings to be had from constructing and operating their
 own WWTP.  The Rahr plant is situated in a portion of the river for which there are assigned load allocations under a
 TMDL. The plant was permitted to establish a new discharge point in exchange for upstream offsetting reductions. The
 offset clause was written into the facility's NPDES permit. The facility was given a stringent discharge limit in addition to
 the offset provision. The plant has financed upstream reductions in phosphorous in exchange for CBOD discharge from
 their WWTP. Rahr established a trust fund with an oversight board to implement the trades. Rahr gives funds to the trust
 fund to implement the BMPs.  Five types of nonpoint source projects are allowed under the terms of the permit: soil erosion
 BMPs, livestock exclusion from  waterways, rotational grazing, set-aside of critical areas, and wetland treatment systems.

 Environmental Problem: Biological oxygen demand s-nd dissolved oxygen in the Minnesota River.

 Pollutant(s) / Pollution Type(s): The Rahr plant must reduce upstream phosphorous discharges to offset CBOD discharges
 from their WWTP.

 Trade Types:  Point/Nonpoint

 Stage of Implementation:  The TMDL load allocations were.assigned in 1988. The Rahr permit was issued  in January
 1997.  The plant has contracted for trades with three nonpoint sources thus far.

 Relation to TMDL: The permit was designed to comply with TMDL limits.

 Number of Potential Participants: Thus far, the trading participants include Rahr Malting plant and three NPS sources;
 Rahr may seek more NPSs in the future. Rahr works with the Coalition for Clean Minnesota River (a member of which sits
 on the  board of directors of the trust fund) and with American Rivers to identify trading opportunities.

 Trading Ratios:  The trading ratio is 2:1 nonpoint to point.  There is an additional cross-parameter ratio of 8 Ibs. CBOD :  1
 Ib. P.  The trading ratio accounts for uncertainty and provides a net environmental benefit  The cross parameter ratio reflects
 a scientific assessment of the relative impacts on chlorophyl from phosphorus runoff and from CBOD discharge.

 Estimated Cost Savings: The impetus for building the WWTP was to generate long term business cost savings and avoid
 uncertainty regarding industrial user fees to the POTW.

 Available Written Information: Several brief reports, as well as the NPDES  permit are available from the Minnesota
 Pollution Control Agency.

 Innovative Aspects: The trades  are cross pollutant trades (P for BOD). Five categories of BMPs were  specified as
 permissible under the terms of the permit. After Rahr has met its requirements, the company hopes  to build the trust fund
 into a separate corporate sponsorship for the river and attract money for additional projects.

 Obstacles: Negotiations between plant, environmental groups and State over setting trading ratios were difficult.
 Establishing an acceptable ratio between BOD and P was a significant scientific undertaking.
                                      /
 Web-sites: Minnesota River Basin: http://www.epa.gov/surf2/ahr/30/
 http://www.pca.state.mn.us/water/basins/mnriver/index.htTnl
 http://www.pca.state.mn.us/hot/es-mn-r.html

 Contact: Wayne Anderson, Minnesota Pollution Control Agency (651) 296-7323, wavne.p.andersonfa).pca.state.mn.us
Attachment 1, Page 8 of 11

-------
                    SOUTHERN MINNESOTA BEET SUGAR COOPERATIVE TRADING PROGRAM (MN)

Nature of Activity: In order to allow the Southern Minnesota Beet Sugar Cooperative's plant to increase its discharge to the
Minnesota River, the Minnesota Pollution Control Agency issued SMBSC a permit specifying that the facility must obtain
offsets. SMBSC will seek reductions via BMPs from farmers, some of whom are members of the cooperative.

Environmental Problem: Nutrients in Minnesota River.

PoIIutant(s) / Pollution Type(s): Phosphorous.

Trade Types: Point/ Nonpoint

Stage of Implementation: The permit was issued in April 1999, no trades have occur/ed to date.

Relation to TMDL: There is a TMDL in the lower Minnesota River for dissolved oxygen.  Excess phosphorous entering the
watershed leads to eutrophication and increased oxygen demand.

Number of Potential Participants: Souther Minnesota Beet Sugar Cooperative and multiple farmers

Trading Ratios: Includes 0.6 Ib for uncertainty and 1 Ib for environmental improvement = ratio of 2.6:1

Estimated Cost Savings: -

Available Written Information: The NPDES permit and a brief summary are available from the Minnesota Pollution
Control Agency.

Innovative Aspects: The fact that the SMBSC is a cooperative owned by farmers provides for an unusually direct
relationship between the point source discharger and the non-point sources.  There are very high penalties for non-compliance.

Obstacles: SMBSC had a sub-par environmental track record, which initially made the environmental community concerned
about allowing the plant to participate in a trading program. Negotiations over the environmental improvement component of
the trade ratio were difficult.

Web-sites: Minnesota River Basin: http://www.epa.gov/surf2/ahr/30/
http://www.pca.state.mn.us/water/basins/mnriver/index.html
http://www.pca.state.mn.us/news/mav99/nr51299.html

Contact: Wayne Anderson, Minnesota Pollution Control Agency (651) 296-7323, wavne.p.anderson@pca.state.mn.us
                                                                                       Attachment 1. Pa?e 9 of 11

-------
                              SAN FRANCISCO BAY MERCURY OFFSET PROGRAM (CA)

Nature of Activity: The San Francisco Bay Regional Water Quality Control Board of the California EPA is in the process of
developing a regional mercury offset program. The program would allow new or expanding point sources in the Bay to offset
their mercury loadings by paying for remediation efforts at abandoned mines.

Environmental Problem: Accumulating toxics in San Francisco Bay exceed standards in fish tissue and in the water.

Pollutant(s) / Pollution Type(s): Mercury.

Trade Types: Point/ Nonpoint

Stage of Implementation: The program has been under development since 1996. A draft of the trading plan was released in
June 1998 and first discussed by the full Board in December '98. The plan proposes a pilot project that would  last seven
years.  In June 1999 the RWQCB defined a set of trading issues which it is beginning to examine in mere specific detail.

Relation to TMDL: The trading program is being planned in coordination with a TMDL for the Bay.

Number of Potential Participants: There are twenty POTWs and thirteen industrial dischargers in the Bay, all of which
could potentially be involved in an offset program.  There are potentially hundreds of mines that could be included as well.

Trading Ratios: 3:1 or greater (5:1 has been discussed as well).

Estimated Cost Savings: —

Available Written Information: The draft report and various memoranda are available from the RWQCB.

Innovative Aspects: The Federal abandoned mine  clean-up effort in the Sierras is being linked to the offset program in the
Bay.  POTWs appear willing to make some effort to control Hg from mines even though POTWs contribute only 2 % of
loadings. This program is the first  to develop a trading scheme specifically for toxics. The Board is considering
implementing trades with ten-year  durations (the length of two NPDES permits) so as to minimize the uncertainty faced by
point sources and encourage their participation. The Board is seeking ways to involve the Air Quality Board, since
atmospheric deposition is a major factor in  the Bay's Hg loadings.

Obstacles: Point source dischargers are wary of buying into Superfund-like liability. It is difficult to model the loadings from
the mines, so the offsets may not measured in strictly quantitative terms. It is therefore anticipated that the permit language
may be especially difficult to write. There are jurisdictional complications that arise due to the fact that the Bay and the mines
are in different California EPA regions. The Board is concerned about putting adequate and clear.language in the permits
regarding anti-degradation and anti-backsliding.  The Board is concerned that only the same bio-available form of Hg be
traded.

Web-sites: San  Francisco Bay: http://www.epa.gov/surf2/hucs/180S0004/
http://www. swrcb.ca.gov/-rwqcb2/

Contact: Khalil Abu-Saba, California Environmental Protection Agency / San Francisco Bay Regional Water Quality Control
Board (510)622-2300, ABU@rb2.swrcb.ca.gov .
Attachment 1, Page 10 of 11

-------
                      TRUCKEE RIVER WATER RIGHTS AND POLLUTION OFFSET PROGRAM (NV)

Nature of Activity: The Reno-Sparks Joint Wastewater Treatment Facility (Truckee Meadows Water Reclamation Facility)
sought to increase its discharge to the Truckee River. The river has been the locus for decades of contentious water rights
disputes. The POTWs' need to expand its discharge became one of several related issues in the dispute. A number of
elements of the dispute were resolved by a water rights agreement that involves some facets of pollutant trading. The
agreement allows the plant to increase its discharge while assuring attainment of water quality standards during the dry (i.e.,
low flow)  season. The local communities and the Federal government are sharing the costs of purchasing upstream water
rights. The purchases will both reduce NPS pollution by precluding the use of the water for agricultural purposes, and will
also possibly mitigate the impact of increased POTW loading by increasing the river's flow during the dry season.

Environmental Problem: There are a variety of water quality problems as well as endangered species in the Truckee River,
due to a combination of very low flow and discharges from point and nonpoint sources.

Pollutant(s) / Pollution Type(s): Total Nitrogen, Total Phosphorous, Total Dissolved Solids, Dissolved Oxygen and
Temperature.

Trade Types: The water rights purchases do not fit easily into the categories of trading types employed in this report,
however a point/ nonpoint trade may be the most fitting category.

Stage of Implementation: Water rights have been a major issue on the Truckee River for the greater part of this century.
Negotiations for this project started in 1994 and an agreement was reached in June 1996. The 1996 agreement resolved
several lawsuits that had been ongoing since the early 1980's. Some water rights have  already been bought, but completion of
the project is not expected for another five years.

Relation to TMDL: A TMDL is in place for total nitrogen, total dissolved solids and dissolved phosphorous in order to
control dissolved oxygen. The municipalities had to  produce creative solutions in order to both expand the plant and comply
with the TMDL loading allocations and wasteload allocation.

Number of Potential Participants: The Federal government, the City  of Reno, the City of Sparks, Washoe County, the
Pyramid Lake Paiute Tribe and various farmers and nonpoint sources are all involved in this program.

Trading Ratios: Trading ratios per se were not calculated, but the amount of water rights to be purchased is expected to
offset the effects of increased loadings from the POTW.

Estimated Cost Savings: N/A

Available Written Information: A TMDL fact sheet is available from EPA Region 9 and a-PBS video "Healing the Water"
is available from Water Environment Federation and EPA Region 9. A chronology of events on the Truckee (pre 20th century
and 20* century) is available from the Nevada Division of Water Planning.

Innovative Aspects: This is the only trading project  to employ purchasing of water rights as a tool for improving water
quality. Another rare, if not unique aspect is the utilization of a mediation consulting firm to assist in dispute resolution.

Obstacles: Water rights and water pollution have been contentious issues in the region for many decades; negotiations
sponsored by Senator Reid of Nevada were facilitated by a mediation consulting firm over a two year period.

Web-sites: Pyramid - Winnemucca Lakes: http://www.epa.gov/surf2/hucs/16050103/
http://www.state.nv.us/cnr/ndwp/truckee/trchrono.htni

Contact: Cheryl McGovem. Environmental Protection Agency (415) 744-2013
                                                                                       Attachment 1, Page 11 of 11

-------
ATTACHMENT 2: CHARACTERISTICS OF OTHER STORM WATER FACILITIES

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 KOI Database, and Difference in Establishments, 1995-1996

                    SBA definition        % Ekery small
 Facilities in       Difference in         Expected £
NOI Database,  Establishments, 1995-  Owned by Large
    1996             1996              Entities
4581
5015
3273
5093
4512
2951
4215
4513
3732
1442
4212
^213
Pl1
11
4522
4231
3271
2421
1389
2653
3272
4952
1422
45
2048
^5
Airports, Flying Fields, and Airport Terminal Services
Motor Vehicle Parts. Used
Ready Mixed Concrete
Scrap and Waste Materials
Air Transportation, Scheduled
Asphalt Paving Mixtures and Blocks
Courier Services, Except by Air
Air Courier Services
Boat Building and Repairing
Construction Sand and Gravel
Local Trucking Without Storage Except Garbage and
Refuse Collection, Without Disposal
Trucking, Except Local
Railroads. Line-Haul Operating
United States Postal Service
Air Transportation, Nonscheduted PT, Offshore Marine
Air Transportation Services
Terminal and Joint Terminal Maintenance Facilities for
Motor Freight Transportation
Concrete Block and Brick
Sawmills and Planing Mills, General
Oil and Gas Field Services. N.E.C.
Corrugated and Solid Fiber Boxes
Concrete Products. Except Block and Brick
Sewerage Systems
Crushed and Broken Limestone
MAJOR GROUP 45 - TRANSPORTATION BY AIR
Prepared Feeds and Feed Ingredients for Animals and
Fowls, Except Dogs and Cats
General Warehousing and Storage
$5.0 Mil
100 ppl
500 ppl
100 ppl
1.500 ppl
500 ppl
$18.5 Mil
•1,500 ppl
500 ppl
500 ppl
$18.5 Mil
$18.5 Mil
1.500 ppl
-
1.500 ppl
$5.0 Mil
500 ppl
500 ppl
S5.0 Mil
500 ppl
500 ppl
$5.0 Mil
500 ppl

500 ppl
S18.5 Mil
92%
100%
98%
97%
Approximately 94%
92%
Between 98% and 99%
Approximately 99%
99%
97%
Between 98% and 99%
Between 94% and 98%
N/A
N/A
Approximately 98%
84%
95%
98%
97%
95%
98%
90%
92%

94%
Between 96% and 93%
715
576
397
273
263
213
202
199
148
136
125
121
99
82
79
76
70
' 68
63
61
61
54
52
44
42
40
433
192
220
415
233
49
1046
136
349
145
7912
2593
no data available
no data available
236
10
negative
negative
318
64
negative
166
0
no data available
negative
1325
*









V
1
N
N


h
I


I





                                                                                                                       Attachment 2, Page 1 of 16

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 HOI Database, and Difference in Establishments, 1995-1996

                    SBA definition        % likely small
 Facilities in       Difference in        Expected $
NOI Database,  Establishments, 1995-  Owned by Lap
    1996             1996              Entities
3731
5171
4151
4493
4200
4953
2851
1429
3479
39
4013
2491
3441
2893
3674
2086
2051
2842
4911
4131
2655
1311
^99
3398
Shipbuilding and Repair of Nuclear Propelled Ships
except Shipbuilding of Nonnudear Propelled Ships and
NonpropeOed Ships
Petroleum Bulk Stations and Terminals
School Buses
Marinas
MAJOR GROUP 42 - MOTOR FREIGHT
TRANSPORTATION AND WAREHOUSING
Refuse Systems
Paints, Varnishes, Lacquers, Enamels, and Allied
Products
Crushed and Broken Stone, N.E.C.
Coating, Engraving, and Allied Services, N.E.C.
Wood Products. N.E.C.
Railroad Switching and Terminal Establishments
Wood Preserving
Fabricated Structural Metal
Printing Ink
Semiconductors and Related Devices
Bottled and Canned Soft Drinks and Carbonated Waters
Bread and Other Bakery Products, Except Cookies and
Crackers
Specialty Cleaning, Polishing, and Sanitation
Preparations
Electric Services
Intercity and Rural Bus Transportation
Fiber Cans, Tubes. Drums, and Similar Products
Crude Petroleum and Natural Gas
Chemicals and Chemical Preparations. NEC
Metal Heat Treating
1.000 ppl
100 ppl
$5.0 Mil
$5.0 Mil

$6.0
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
/
4 Million MW hrs
S5.0 Mil
500 ppl
500 ppl
500
750 ppl
Approximately 97%
92%
97%
98%

Between 79% and 84%
95%
92%
97%
98%
N/A
96%
98%
92%
89%
87%
96%
95%
Approximately 82%
84%
89%
99%
90% -
Approximately 97%
40
33
32
32
31
31
31
30
29
27
26
26
26
24
23
"22
22
21
20
20
20
20
19
19
68
negative
94
460
4835
negative
109
39
42
negative
no data available
23
288
25
150
negative
440
85
528
negative
4
233
negative
51

\


r
t-



N
N




1



r-


t

                                                                                                                    Attachment 2, Page 2 of 16

-------
SIC  S\C definition
Facilities with SIC Information in the 1996 NC» Database, and Difference in Establishments, 1995-1996

                     SBA definition        % likely small
 Facilities in      Difference in
NO! Database, Establishments, 1995-  Owned
    1996             1996
2026 Fluid Milk
2891 Adhesives and Sealants
2013 Sausages and Other Prepared Meat Products
3499 Fabricated Metal Products, N.E.C.
500 ppl
500 ppl
500 ppl
500 ppl
2821 Plastics Materials, Synthetic Resins.and Nonvulcanizable 750 ppl
Elastomers
3089 Plastics Products. N.E.C.
3721 Aircraft
3321 Gray and Ductile Iron Foundries
2869 Industrial Organic Chemicals, N.E.C.
2841 Soap and Other Detergents. Except Specialty Cleaners
3471 Electroplating, Plating, Polishing, Anodizing, and
Coloring
2448 Wood Pallets and Skids
2621 Paper Mills
3449 Miscellaneous Structural Metal Work
4111 Local and Suburban Transit
3444 Sheet Metal Work
2411 Logging
2015 Poultry Slaughtering and Processing
2037 Frozen Fruits, Fruit Juices, and Vegetables
75 1 4 Passenger Car Rental
3086 Plastics Foam Products
3496 Miscellaneous Fabricated Wire Products
3999 Manufacturing Industries, N.E.C.
2992 Lubricating Oils and Greases
3812 Search, Detection, Navigation, Guidance, Aeronautical,
and Nautical Systems and Instruments
3411 Metal Cans
3365 Aluminum Foundries
4600 MAJOR GROUP 46 - PIPELINES, EXCEPT NATURAL
GAS
500 ppl
1,500 ppl
500 ppl
1,000 ppl
750 ppl
500 ppl
500 ppl
750 ppl
500 ppl
$5.0 Mil
500 ppl
500 ppl
500 ppl
500 ppl
S18.5 Mil
500 ppl
500 ppl
500 ppl
500 ppl
750 ppl
1,000 ppl
500 ppl

84%
89%
93%
97%
Approximately 91%
94%
Approximately 95%
90%
Approximately 91%
Approximately 98%
99%
100%
Approximately 85%
90%
90%
98%
100%
79%
80%
Between 93% and 97%
92%
96%
98%
90%
Approximately 93%
Approximately 95%
95%

17
17
17
17
16
16
16
15
15
15
15
15
14
14
14
14
14
13
13
13
12
12
12
11
11
11
11
10
negative
34
negative
107
72
negative
34
4
negative
78
164
247
27
negative
124
negative
188
negative
10
negative
48
182
negative
25
20
negative
19
negative
Nit

Nlf
*
c
N/A
2
0
N/A
2
2
T
4
N/A
12
N/A
0
N/A
2
N/A
4
7
N/A
3
1
^J/A
w>

                                                                                                                       Attachment 2, Page 3 of 16

-------
SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in Establishments, 1995-1996
                                                  «
                    SBA definition      .  % Ekety small
 Facilities in       Difference in        Expected*
NOI Database, Establishments, 1995-  Owned by Large
    1996             1996              Entities
1321
5169
3599
4226
3714
2631
4500
1221
3679
2844
2875
f-
.4
3357
3295
1423
2077
2493
2033
2434
3728
3531
2813
32
2436
1
,,11
Natural Gas Liquids
Chemical and Allied Products, N.E.C.
Industrial and Commercial Machinery and Equipment,
N.E.C.
Special Warehousing and Storage, NEC
Motor Vehicle Parts and Accessories
Paperboard Mills
MAJOR GROUP. 45 - TRANSPORTATION BY AIR
Bituminous Coal and Lignite Surface Mining .
Electronic Components, N.E.C.
Perfumes, Cosmetics, and Other Toilet
Fertilizers, Mixing Only
Fabricated Plate Work (Boiler Shops)
Marine Cargo Handling
Aircraft Engines and Engine Parts
Drawing and Insulating of Nonferrous Wire
Minerals and Earths, Ground or Otherwise Treated
Crushed and Broken Granite
Animal and Marine Fats and Oils
Reconstituted Wood Products
Canned Fruits, Vegetables, Preserves, Jams, and Jellies
Wood Kitchen Cabinets
Aircraft Parts and Auxiliary Equipment N.E.C.
Construction Machinery and Equipment
Industrial Gases
MAJOR GROUP 32 - STONE, CLAY. GLASS. AND
CONCRETE PRODUCTS
Softwood Veneer and Plywood
Terminal and Service Facilities for Motor Vehicle
Passenger Transportation
Leather Tanning and Finishing
500 ppl
100 ppl
500 ppl
$18.5 Mil
750 ppl
750 ppl

500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
$18.5 Mil
1.000 ppl
1,000 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
1.000 ppl
750 ppl
1,000 ppl

500 ppl
$5.0 Mil
500 pp!
60%
95%
99%
Betvreen 87% and 92%
Approximately 96%
Approximately 85%

92%
92%
92%
93%
95%
Between 71% and 82%
Approximately 94%
Approximately 91%
82%
84%
84%
85%
90%
99%
Approximately 97%
Approximately 95%
Approximately 93%

77%
88%
95%
10
10
10
9
9
9
9
9
9
9
9
9
8
8
8
8
8
8
8
8
• 8
7
7
7
7
7
7
7
7
591
1278
115
154
9
514
negative
negative
negative
negative
22
negative
negative
negative
40
6
negative
33
33
558
81
negative
7
no data available
negative
15
9
3
30
13
12
6
1
N/A
N/A
N/A
N/A
N/;
"
Nlf
Nit
Nil


N/,




N

N
N


                                                                                                                Attachment 2, Page 4 of 16

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in EstabEshments, 1995-1996

                     SBA definition        % liKely small
 Facilities in       Difference in
NOI Database,  Establishments, 1995-  Owned by
    1996             1996              Entities,
721
2752
4214
4222
2819
42
40
3600
2879
1499
3341
3533
.'.861
2099
3498
3281
4121
9711
3519
3663
3585
3241
3354
Crop Planting, Cultivating, and Protecting
Commercial Printing, Lithographic
Local Trucking With Storage
Refrigerated Warehousing and Storage
Industrial Inorganic Chemicals, N.E.C.
MAJOR GROUP 42 - MOTOR FREIGHT
TRANSPORTATION AND WAREHOUSING
MAJOR GROUP 40 - RAILROAD TRANSPORTATION
MAJOR GROUP 36 - ELECTRONIC AND OTHER
ELECTRICAL EQUIPMENT AND COMPONENTS,
EXCEPT COMPUTER EQUIPMENT
Pesticides and Agricultural Chemicals, NEC
Miscellaneous Nonmetallic Minerals, Except Fuels
Secondary Smelting and Refining of Nonferrous Metals
Oil and Gas Field Machinery and Equipment
Photographic Equipment and Supplies
Food Preparations, N.E.C.
Fabricated Pipe and Pipe Fittings
Cut Stone and Stone Products
Taxi cabs
National Security
Internal Combustion Engines. N.E.C.
Radio and Television Broadcasting and Communications
Equipment
Air-Conditioning and Warm Air Heating Equipment and
Commercial and Industrial Refrigeration Equipment
Cement Hydraulic
Aluminum Extruded Products
$5.0 Mil
500 ppl
$18.5 Mil
$18.5 Mil
1.000 ppl



500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
$5.0 Mil
N/A
1,000 ppl
750 ppl
750 ppl
750 ppl
750 ppl
98%
. 99%
Between 96% and 99%
Between 87% and 100%
Approximately 93% y



83%
88%
88%
94%
94%
95%
95%
99%
99%
N/A
Approximately 96%
Approximately 94%
Approximately 94%
Approximately 87%
Approximately 84%
7
7
6
6
6
6
6
6
6
6
6
6
6
6'
6
6
6
5
5
5
5
5
5
68
856
160
negative
negative
no data available
no data available
161
19
negative
negative
34
negative —
negative
63
129
10
no data available
negative
102
negative
6
7
1
9
4
N/A
N/A
N/A
N/A
N/A
"W

N/A
2
— N/A—
N/A-
3
1
0
N;A
N/A
6
N/A

1
                                                                                                                        Attachment 2, Page 5 of 16

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 KOI Database, and Difference in Estabfishmerrts, 1995-1996

                    SBA definition        % likely small
 Facilities in       Difference in        Expected ?
NOI Database,  Establishments, 1995-  Owned by La
    1996             1996              Entities
1021
2221
2023
4142
3589
3088
3672
2439
3366
7389
'2511
2759
7538
3011
3621
3315
3312
3221
3351
1400
4100
Copper Ores
Broadwoven Fabric Mills, Manmade Fiber and Silk
Dry, Condensed, and Evaporated Dairy Products
Bus Charter Service, Except Local
Service Industry Machinery, N.E.C.
Plastics Plumbing Fixtures
Printed Circuit Boards
Structural Wood Members, N.E.C. -
Copper Foundries
Business Services, N.E.C. FT, Map Drafting Services,
Mapmaking (Including Aerial) and Photogrammetnc
Mapping Services
Wood Household Furniture, Except Upholstered
Commercial Printing, N.E.C.
General Automotive Repair Shops
Tires and Inner Tubes
Motors and Generators
Steel Wiredrawing and Steel Nails and Spikes
Steel Works, Blast Furnaces (Including Coke Ovens),
and Rolling Mills
Glass Containers
Rolling, Drawing, and Extruding of Copper
MAJOR GROUP 14 - MINING AND QUARRYING OF
NONMETALLIC MINERALS, EXCEPT FUELS
MAJOR GROUP 41 - LOCAL AND SUBURBAN
TRANSIT AND INTERURBAN HIGHWAY PASSENGER
TRANSPORTATION
500 ppl
500 ppl
500 ppl
$5.0 Mil
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
S5.0 Mil
500 ppl
500 ppl
$5.0 Mil
1,000 ppl
1,000 ppl
1.000 ppl
1.000 ppl
750 ppl
750 ppl


77%
82%
82%
93%
95%
96%
96%
97%
97%
97%
98%
99%
99%
Approximately 92%
Approximately 91%
Approximately 89%
Approximately 88%
Approximately 88%
Approximately 86%


5
5
5
5
5
5
5
5
5
5
5
5
5
4
4
4
' ' 4
4
4
4
4
8
30
11
103
69
116
121
83
18
9979
368
negative
6506
14
22
negative
92
4
6 '
37
497
                                                                                                                     Attachment 2, Page 6 of 16

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in Establishments, 1995-1996

                     SBA definition        % likely small
 Facilities in       Difference in
NOI Database, Establishments, 1995-  OwnedT
    1996             19%
1600
41
7600
14
1459
2674
3324
2451
4789
3432
3255
3081
3494
2024
3561
2022
5172
4141
7359
2299
4499
1041
3087
MAJOR GROUP 16 - HEAVY CONSTRUCTION OTHER
THAN BUILDING CONSTRUCTION - CONTRACTORS
MAJOR GROUP 41 - LOCAL AND SUBURBAN
TRANSIT AND INTERURBAN HIGHWAY PASSENGER
TRANSPORTATION
MAJOR GROUP 76 - MISCELLANEOUS REPAIR
SERVICES
MAJOR GROUP 14 - MINING AND QUARRYING OF
NONMETALUC MINERALS. EXCEPT FUELS
Clay, Ceramic, and Refractor/Minerals, N.E.C.
Uncoated Paper and Multiwall Bags
Steel Investment Foundries
Mobile Homes
Transportation Services. N.E.C.
Plumbing Fixture Fittings and Trim
Clay Refractories
Unsupported Plastics Film and Sheet
Valves and Pipe Fittings, N.E.C.
Ice Cream and Frozen Desserts
Pumps and Pumping Equipment
Natural, Processed, and Imitation Cheese
Petroleum and Petroleum Products Wholesalers, Except
Bulk Stations and Terminals
Local Bus Charter Service
Equipment Rental and Leasing, N.E.C.
Textile Goods. N.E.C.
Water Transportation Services, N.E.C..
Gold Ores
Custom Compounding of Purchased Plastics Resins




500 ppl
500 ppl
500 ppl
500 pp)
S5.0 Mil
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
100 ppl
$5.0 Mil
$5.0 Mil
500 ppl
$5.0 Mil
500 ppl
500 ppl




76%
76%
82%
83%
85%
85%
85%
86%
88%
89%
89%
91%
92%
93%
93%
93%
94%
94%
94% •
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
878
no data available
negative
no data available
negative
3
1
15
191
negative
9
115
8
34
25
negative
133
31
16
negative
161
39
183

f
;•.
f.
N
•-


*
N;

1



N/
1


N//
1

                                                                                                                       Attachment 2. Past 7 of 16

-------
SIC definition
Facilities with SIC Information in the 1996 N01 Database, and Difference in Establishments, 1995-1996

                    SBA definition        'A likely small
 Facilities in       Difference in         Expected tt
NOI Database, Establishments, 1995-  Owned by Large
    1996             1996             Entities
4959
4119
3448
5085
3469
2011
3799
3446
1611
5511
3694
V
J7
2611
3317
1475
2297
3085
2671
3643
2269
3462
2231
1446
3363
3069
2679
'5
Sanitary Services, N.E.C.
Local Passenger Transportation, N.E.C.
Prefabricated Metal Buildings and Components
Industrial Supplies
Metal Stampings, N.E.C.
Meat Packing Plants
Transportation Equipment, Kl.E.C.
Architectural and Ornamental Metal Work
Highway and Street Constnjction.Except Elevated
Highways
Motor Vehicle Dealers (New and Used)
Electrical Equipment for Internal Combustion Engines
Synthetic Rubber (Vulcanizable Elastomers)
Nonday Refractories
Pulp Mills
Steel Pipe and Tubes
Phosphate Rock
Nonwoven Fabrics
Plastics Bottles
Packaging Paper and Plastics Film, Coated and
Laminated
Current-Carrying Wiring Devices
Finishers of Textiles, N.E.C.
Iron and Steel Forgings
Broadwoven Fabric Mills, Wool (Including Dyeing and
Finishing)
Industrial Sand
Aluminum Die-Castings
Fabricated Rubber Products, N.E.C.
Converted Paper and Paperboard Products, N.E.C.
Steel Foundries, N.E.C.
$5.0 Mil
^5.0 Mil
500 ppl
100 ppl
500 ppl
500 ppl
500 ppl
500 ppl
$17.0 Mil
$21.0 Mil
750 ppl
1,000 ppl
750 ppl
750 ppl
1,000 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
96%
96%
96%
96%
96%
96%
97%
98%
Between 86% and 96%
Between 37% and 76% .
Approximately 95%
Approximately 93%
Approximately 88%
Approximately 86%
Approximately 85%
50%
77%
82%
83%
87%
89%
89%
89%
89%
89%
90%
91%
91%
4
4
4
4
4
4
4
4
3
3
3
^

3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
349
604
51
915
negative
173
41
172
326
407
54
21
6
9
1
2
6
35
10
30
negative
28
s negative
7
4
negative
negative
13
14
24
2
37
N/A
7
1
3
29
177
3
1
1
1
0
1
•5

'
4
N/A
-
M

i
Nl,
Nl.

                                                                                                                 Attachment 2, Page 8 of 16

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in Establishments, 1995-1996
                          *

                    SBA definition        % likely small
 Facilities in      Difference in
NOI Database, Establishments, 1995-  Owned b
    1996             19%
ExpectP^^
2241
3429
5051
5032
3842
3442
2298
2426
5812
3944
3993
1411
391!;
2261
;641
572
511
557
537
573
i95
.56
15
52
Narrow Fabric and Other Smallwares Mills: Cotton, Wool,
Silk and Manmade
Hardware, N.E.C.
Metals Service Centers and Offices
Brick, Stone, and Related Construction Materials
/
Orthopedic, Prosthetic, and Surgical Appliances and
Supplies
Metal Doors, Sash, Frames, Molding, and Trim
Cordage and Twine
Hardwc-. : ^imension and Flooring Mills
Eating P&JKS EXCEPT, Food Service, Institutional
Games, Toys, and Children's Vehicles, Except Dote and
Bicycles
Signs and Advertising Specialties
Dimension Stone
Jewelers' Findings and Materials, and Lapidary Work
Finishers of Broadwoven Fabrics of Cotton
Electric Lamp Bulbs and Tubes
Computer Storage Devices
Steam, Gas, and Hydraulic Turbines, and Turbine
Generator Set Units
Folding Paperboard Boxes. Including Sanitary
Industrial Trucks, Tractors, Trailers, and Slackers
Nitrogenous Fertilizers
Magnetic and Optical Recording Media
Rolling, Drawing, and Extruding of Nonferrous Metals,
Except Copper and Aluminum
Cold-Rolled Steel Sheet. Strip, and Bars
Asphalt Fetts and Coatings
500 ppl
500 ppl
100 ppl
100 ppl
500 ppl
500 ppl
500 ppl
500 ppl
S5.0 Mil
500 ppl
500 ppl
500 ppl
500 ppl
1.000 ppl
1.000 ppl
1.000 ppl
1.000 ppl
750 ppl
750 ppl
1,000 ppl
1,000 ppl
750 ppl
1.000 ppl
750 ppl
92%
92%
94%
95%
95%
95%
95%
96%
98%
98%
99%
99%
100%
Approximately 98%
Approximately 98%
Approximately 97%
Approximately 96%
Approximately 96%
Approximately 95%
Approximately 95%
Approximately 94%
Approximately 93%
Approximately 93%
Approximately 93%
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
8
negative
354
234
negative
60
19
140
27309
negative
773
24
10
108
7
19
3
18 '
negative
13 .
5
6
20
5
1
N/A
21
12
N/A
3
1
6
546
•^
€
j
0
0
2
0
1
0
1
N/A
1
0
0

0
                                                                                                                    Attachment 2, Page 9 of 16

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in Estabfishmerrts, 1995-1996

                    SBA definition        % likely small
 Facilities in       Difference in        Expected*
NOI Database,  Establishments, 1995-  Owned by Larg
    1996              1996              Entities
3211
3743
2911
2892
3334
3331
2865
4400
4000
24
2895
'4
1479
2999
2254
2047
3491
8733
2672
3648
3554
3532
3253
3841
3291
•15
Flat Glass
Railroad Equipment
Petroleum Refining
Explosives
Primary Production of Aluminum
Primary Smelting and Refining of Copper
Cyclic Organic Crudes and Intermediates, and Organic
Dyes and Pigments
MAJOR GROUP 44 - WATER TRANSPORTATION
MAJOR GROUP 40 - RAILROAD TRANSPORTATION
MAJOR GROUP 24 - LUMBER AND WOOD
PRODUCTS. EXCEPT FURNITURE
Carbon Black
Lime
Chemical and Fertilizer Mineral Mining. N.E.C.
Products of Petroleum and Coal, N.E.C.
Knit Underwear and Nightwear Mills
Dog and Cat Food
Industrial Valves
Noncommercial Research Organizations
Coated and Laminated Paper, N.E.C.
Lighting Equipment N.E.C.
Paper Industries Machinery
Mining Machinery and Equipment Except Oil and Gas
Field Machinery and Equipment
Ceramic Wall and Floor Tile
Surgical and Medical Instruments and Apparatus
Abrasive Products
Truck Trailers
1,000 ppl
1.000 ppl
1,500 ppl
750 ppl
1,000 ppl
1,000 ppl
750 ppl



500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
$5.0 Mil
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
Approximately 92%
Approximately 91%
.Approximately 91%
Approximately 90%
Approximately 88%
Approximately 87%
Approximately 86%



56%
66%
72%
75%
80%
85%
86%
88%
89%
92%
93%
93%
93%
93%
93%
93% '
2
2
2
2
*
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
18
6
24
2
11
1
3
676
9810
no data available
.1
negative
negative
negative
4
18
45
negative
51
27 . '
29
11
22
180
negative
43







K
N
N

N
N
N



I






I

                                                                                                                   Attachment 2, Page 10 of 16

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in Establishments, 1995-1996

                     SBA definition        V, likely small
 Facilities in       Difference in        Expect
NOI Database,  Establishments, 1995-  Owned t
    1996             1996              Entt,
3536
3569
5141
3259
5551
2064
2512
3231
3952
2521
3423
8999
5091
3523
7699
3544
5995
7622
5083
7999
3949
9221
9621
9224
9199
Overhead Traveling Cranes, Hoists, and Monorail
Systems
General Industrial Machinery and Equipment N.E.C.
Groceries, General Line
Structural Clay Products, N.E.C.
Boat Dealers
Candy and Other Confectionery Products
Wood Household Furniture, Upholstered
Glass Products, Made of Purchased Glass'
Lead Pencils, Crayons, and Artists' Materials
Wood Office Furniture
Hand and Edge Tools, Except Machine Tools and
Handsaws
Services. N.E.C.
Sporting and Recreational Goods and Supplies
Farm Machinery and Equipment
Repair Shops and Related Services, NEC
Special Dies and Tools, Die Sets, Jigs and Fixtures, and
Industrial Molds
Optical Goods Stores
Radio and Television Repair Shops
Farm and Garden Machinery and Equipment
Amusement and Recreation Services, NEC
Sporting and Athletic Goods. N.E.C.
Police Protection
Regulation and Administration of Transportation
Programs
Fire Protection
General Government. N.E.C.
500 ppl
500 ppl
100 ppl
500 ppl
$5.0 Mil
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
$5.0 Mil
100 ppl
500 ppl
$5.0(9)
500 ppl
S5.0 Mil
S5.0 Mil
100 ppl
S5.0 Mil
500 ppl
-
N/A
N/A
-
93%
93%
94%
94%
94%
95%
96%
96%
96%
97%
97%
97%
97%
97%
98%
98%
98%
98%
98%
98%
. 98%
N/A
N/A
N/A •
N/A
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
34
127
564 '
negative
285
negative
57
93
negative
75 ^^
116 ^P
2342
299
negative
negative
negative
347
1016
263
*
501
249
no data available
no data available
no data available
no data available
                                                                                                                      Attachment 2, Page 11 of 16

-------
       SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in Establishments, 1995-1996
                        •
                    SBA definition        V. likely small
 Facilities in      Difference in        Expected #
NOI Database, Establishments, 1995-  Owned by Large
    1996             1996              Entities
1799   Special Trade Contractors, N.E.C.EXCEPT.  Base      $7.0 Mil
       Housing Maintenance
                                         Between 98% and 99%
                             2902
                                                                                                                       44
4731   Arrangement of Transportation of Freight and Cargo     $18.5 Mil
                                         Between 97% and 99%
                             1317
26
8711   Engineering Services PT, Military and Aerospace       $2.5 Mil
       Equipment and Military Weapons EXCEPT, Contracts
       and Subcontracts for Engineering Services Awarded
       Underthe National Energy Policy Act of 1992. EXCEPT,
       Marine Engineering and Naval Architecture
                                         Between 96% and 98%
                             5209
                                                                                                                      156
1623
fcr
1
3229
3671
3577
3339
3571
3296
3484
3711
2211
3353
3669
2295
J661
P"15
A
Water, Sewer, Pipeline, and Communications and Power $17.0 Mil
Line Construction
Gasoline Service Stations
Farm Product Warehousing and Storage
Pressed and Blown Glass and Glassware. NEC
Electron Tubes
Computer Peripheral Equipment N.E.C.
Primary Smelting and Refining of Nonferrous metals,
.Except Copper and Aluminum
Electronic Computers
Mineral Wool
Small Arms
Motor Vehicles and Passenger Car Bodies
Broadwoven Fabric Mills. Cotton
Aluminum Sheet Plate, and Foil
Communications Equipment N.E.C.
Coated Fabrics, Not Rubberized
Telephone and Telegraph Apparatus
Distilled and Blended Liquors
Pharmaceutical Preparations
$6.5 Mil
$18.5 Mil
750 ppl
750 ppl
1,000 ppl
750 ppl
1.000 ppl
750 ppl
1,000 ppl
1,000 ppl
1,000 ppl
750 ppl
750 ppl
1,000 ppl
1,000 ppl
750 ppl
750 ppl
Between 92% and 98%
Between 92% and 94%
Between 85% and 90%
Approximately 98%
Approximately 98%
Approximately 97%
Approximately 97%
Approximately 97%
Approximately 97%
Approximately 97%
Approximately 96%
Approximately 96%
Approximately 96%
Approximately 96%
Approximately 93%
Approximately 93%
Approximately 93%
Approximately 93%
1 1403
1 negative
1 11
1 56
1 5
1 132
1 4
1 34
1 29
1 15
1 negative
/
1 65
1 5
1 52
1 34
1 48
1 1
1 90
70
N/A
1
1
0
4
0
1
1
0
N/A
-
C
£
<

(

                                                                                                                       Attachment 2, Page 12 of 16

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in Establishments, 1995-1996

                    SBA definition         % likely small
 Facilities in       Difference in
NOI Database,  Establishments, 1995-  Owned t
    1996             1996              Entith..
3399
3631
3624
3635
4612
2111
2300
28
3100
20
2600
34
4700
3400
4931
4922
8744
2079
1031
2843
Primary Metal Products, N.E.C. 750 ppl Approximately 92%
Household Cooking Equipment 750 ppl Approximately 92%
Carbon and Graphite Products 750 ppl Approximately 90%
Household Vacuum Cleaners 750 ppl Approximately 88%
Crude Petroleum Pipelines 1 .500 ppl Approximately 87%
s
Cigarettes 1,000 ppl Approximately 80%
MAJOR GROUP 23 - APPAREL AND OTHER
FINISHED PRODUCTS MADE FROM ICS AND
SIMILAR MATERIALS
MAJOR GROUP 28 - CHEMICALS AND ALLIED
PRODUCTS
MAJOR GROUP 31 - LEATHER AND LEATHER
PRODUCTS
MAJOR GROUP 20 - FOOD AND KINDRED
PRODUCTS
MAJOR GROUP 26 - PAPER AND ALLIED PRODUCTS -
MAJOR GROUP 34 - FABRICATED METAL
PRODUCTS, EXCEPT MACHINERY AND
TRANSPORTATION EQUIPMENT
MAJOR GROUP 47 -TRANSPORTATION SERVICES -
MAJOR GROUP 34 - FABRICATED METAL - ' -
PRODUCTS. EXCEPT MACHINERY AND
TRANSPORTATION EQUIPMENT
Electric and Other Services Combined $5.0 Mil 27%
Natural Gas Transmission $5.0 Mil ' 40%
Facilities Support Management Services EXCEPT, Base $5.0 Mil 63%
Maintenance EXCEPT, Environmental Remediation
Services
Shortening, Table Oils, Margarine, and Other Edible Fats 500 ppl 66%
and Oils, N.E.C.
Lead and Zinc Ores 500 ppl 70%
Surface Active Agents, Finishing Agents. Sulfonated 500 ppl 81 %
Oils, and Assistants
1 negative
1 4
1 6
1 1
1 4
1 2
1 35
1 no data available
1 negative
1 no data available
1 17
1 no data available
1 2204
1- 358
1 negative
1 negative
1 162
1 negative
1 4
1 11
N/A
0
1
0
1
0
N/A
N/A
N/A
N/A
^^
N/A
N/A
N/A
N/A
N/A
60
N/A


                                                                                                                      Attachment 2, Page  13 of 16

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in Estabfishments, 1995-1996

                    SBA definition        % likely small
 Facilities in       Difference in         Expected #
NOI Database,  Establishments, 1995-  Owned by Large
    1995             1996              Entities
3052 Rubber and Plastics Hose and Belting
2835 In Vitro and In Vivo Diagnostic Substances
3563 Air and Gas Compressors
3691 Storage Batteries
3678 Electronic Connectors
3489 Ordnance and Accessories, N.E.C.
2021 Creamery Butter
3465 Automotive Stampings
2251 Women's Full-Lenglh and Knee-Length Hosiery, Except
Socks
3084 Plastics Pipe
2038 Frozen Specialties, N.E.C.
2074 Cottonseed Oil Milts
2095 Roasted Coffee
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
• 500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
P785 Fixed Facilities and Inspection and Weighing Services for $5.0 Mil
Motor Vehicle Transportation
3629 Electrical Industrial Apparatus, NEC
3251 Brick and Structural Clay Tile
2321 Men's and Boys1 Shirts, Except Work Shirts
1044 Silver Ores
3825 Instruments for Measuring and Testing of Electricity and
Electrical Signals
3568 Mechanical Power Transmission Equipment, N.E.C.
2087 Flavoring Extracts and Flavoring Syrups, N.E.C.
2435 Hardwood Veneer and Plywood
3827 Optical Instruments and Lenses
3053 Gaskets, Packing, and Sealing Devices
8734 Testing Laboratories
^951 Pens, Mechanical Pencils, and Parts
'3 Plastics, Foil, and Coated Paper Bags
. A2 Packaged Frozen Foods
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl •
500 ppl
500 ppl
500 ppl
500 ppl
S5.0 Mil
500 ppl
500 ppl
100 ppl
83%
83%
85%
85%
85%
85%
87%
87%
88%
88%
88%
90%
90%
90%
90%
90%
90%
91%
91%
91%
91%
/
92%
92%
92%
92%
92%
92% '
92%
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
12
2
27
3
40
1
3
31
7
35
62
1
4
61
68
6
negative N.
3
negative N
12
3
22
45
41
negative N
5
negative f
52
                                                                                                                     Attachment 2, Page 14 of 16

-------
SIC  SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in Establishments, 1995-1996

                     SBA definition        % likely small
 Facilities in       Difference in
NOI Database,  Establishments, 1995-  Owned I
    1996              1996
3082
2096
3364
3955
5144
2392
3559
5031
5149
3299
2429.
3369
2754
3556
5013
2092
5039
2599
1381
5084
5191
3545
2741
2082
2396
3451
2431
Unsupported Plastics Profile Shapes
Potato Chips, Com Chips, and Similar Snacks
Nonferrous Die-Castings, Except Aluminum
Carbon Paper and Inked Ribbons
Poultry and Poultry Products
Housefumishmgs, Except Curtains and Draperies
Special Industry Machinery, N.E.C.
Lumber, Plywood, Miilwork, and Wood Panels
Groceries and Related Products, N.E.C.
Nonmetallic Mineral Products, N.E.C.
Special Product Sawmills, N.E.C.
Nonferrous Foundnes, Except Aluminum and Copper
Commercial Printing, Gravure
Food Products Machinery
Motor Vehicle Supplies and New Parrs
Prepared Fresh or Frozen Fish andSeafoods
Construction Materials. N.E.C.
Furniture and Fixtures, N.E.C.
Drilling Oil and Gas Wells
Industrial Machinery and Equipment
Farm Supplies
Cutting Tools, Machine Tool Accessories, and
Machinists' Precision Measuring Devices
Miscellaneous Publishing
Malt Beverages
Automotive Trimmings, Apparel Findings, and Related
Products
Screw Machine Products
Millwork
500 ppl
500 ppl
500 ppl
500 ppl
100 ppl
500 ppl
500 ppl
100 ppl
100 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
100 ppl
500 pp!
100 ppl
500 ppl
500 ppl
100 ppl
100 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
93%
93%
93%
94%
94%
94%
94%
95%
95%
95%
96%
96%
96%
96%
96%
96%
96%
97%
97%
97%
97%
97%
97%
98%
98%
98%, '
98%
1 208
1 7
1 14
1 2
1 negative N
1 negative N
1 13
1 373
1 542 :
1 negative N,
1 negative ^^* N,
1 16 ^
1 34
1 56
1 643 2
1 20
1 257 1!
1 negative N//
1 32 '
1 1669 • 5C
1 601 U
1 negative N/A
1 negative WA
1 137 3
1 566 1"
1 60
1 negative N/A
                                                                                                                       Attachment 2. Page 15 of 16

-------
SIC definition
Facilities with SIC Information in the 1996 NOI Database, and Difference in Establishments, 1995-1996

                     SBA definition        % likely small
 Facilities hi       Difference in        Expected 9
NO! Database, Establishments, 1995-  Owned by Large
    1996             1996              Entities
2795
2711
3914
2097
3199
7335
3961
7342
783
3911
782
Platemaking and Related Services
Newspapere: Publishing, or Publishing and Printing
Silverware, Plated Ware, and Stainless Steel Ware
s
Manufactured Ice
Leather Goods. N.E.C.
Commercial Photography
Costume Jewelry and Costume Novelties, Except
Precious Metal
Disinfecting and Pest Control Services
Ornamental Shrub and Tree Services
Jewelry, Precious Metal
Lawn and Garden Services
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
$5.0 Mil
500 ppl
$5.0 Mil
$5.0 Mil
500 ppl
$5.0 Mil
98%
98%
98%
99%
99%
99%
99%
99%
99%
100%
100%
1 46
1 516
1 negative
1 58
1 17
1 288
1 31
1 338
1 477
1 79
1 5550
1
10
N/A
1
0
3
0
3
5
0
0
                                                                                                                Attachment 2. Page 16 of 16

-------
ATTACHMENT 3: CHARACTERISTICS OF NEW PROCESS WATER DISCHARGERS
                          FROM 1995-98

-------
                             SICs of 4,345 new Facilities in PCS Database, 1995-1998
#of
^k facilities
0 904
4952
4941
5541
1442
1221
8211
1422
4953
9999
5171
5172
4911
6552
3273
5093
^21
4491
7542
8999
3731
6515
9711
3089
4959
1455
921
4231
5015
7215
5812
^931
556
166
155
147
140
83
79
69
68
60
53
53
52
51
37
37
35
33
29
29
26
24
23
23
22
22
22
21
20
19
18
18
20.81%
12.80%
3.82%
3.57%
3.38%
3.22%
1.91%
1.82%
1.59%
1.57%
1.38%
1.22%
1.22%
1.20%
1.17%
0.85%
0.85%
0.81%
0.76%
0.67%
0.67%
0.60%
0.55%
0.53%
0.53%
0.51%
0.51%
0.51%
0.48%
0.46%
0.44%
0.41%
0.41%
SIC not listed
Sewerage Systems
Water Supply
Gasoline Service Stations
Construction Sand and Gravel
Bituminous Coal and Lignite Surface Mining
Elementary and Secondary Schools
Crushed and Broken Limestone
Refuse Systems
Nondassifiabie Establishments
Petroleum Bulk Stations and Terminals
Petroleum and Petroleum Products Wholesalers, Except Bulk
Stations and Terminals
Electric Services
Land Subdivides and Developers, Except Cemeteries
Ready Mixed Concrete
Scrap and Waste Materials
Sawmills and Planing Mills, General
Marine Cargo Handling
Carwashes
Services, N.E.C.
Shipbuilding and Repair of Nuclear Propelled Ships except
Shipbuilding of Nonnuclear Propelled Ships and Nonpropelled
Ships
Operators of Residential Mobile Home Sites, PT, Leasing of
Building Space to Federal
National Security
Plastics Products, N.E.C.
Sanitary Services, N.E.C.
Kaolin and Ball Clay
Fish Hatcheries and Preserves
Terminal and Joint Terminal Maintenance Facilities for Motor
Freight Transportation
Motor Vehicle Parts, Used
Coin-Operated Laundries and Drycleaning
Eating Places EXCEPT, Food Service, Institutional
Logging
Electric and Other Services Combined
N/A
$5.0 Mil
$5.0 Mil
$6.5 Mil
500 ppl
500 ppl
$5.0 Mil
500 ppl
$6.0
$5.0 Mil
100 ppl
100 ppl
4 Million MWhrs
$5.0 Mil
500 ppl
100 ppl
500 ppl
$18.5 Mil
$5.0 Mil
$5.0 Mil
1,000 ppl
$5.0 Mil
N/A
500 ppl
SS.OMil
500 ppl
$3.0 Mil
$5.0 Mil
100 ppl
$5.0 Mil
$5.0 Mil
500 ppl
SS.OMil
N/A
90%
97%



Between 92% and 94%
97%
92%
90%
92%
Between 79% and
100%
92%
92%




84%



Approximately 82%
94%
98%
97%
98%
Between 71% and
99%
97%




82%


Approximately 97%
99%
N/A
94%
96%
52%
87%
84%
100%
99%
98%
100%
27%












                                                                                         Attachment 3, page 1 of 13

-------
                                  SICs of 4,345 new Facilities in PCS Database, 1995-1998
I facilities
            % of total
SIC definition
SBA definition
% likely small
4225
2952
7011
7032
3714
3399
5169
8361 x
3599
1499
1321
4922
2499
4932
2011
3069
20tf
20
4151
279
4215
8221
3295
1629
2435
7033
2813
2077
211
3479
4492
4961
3316
6514
4226
9
8
8
8
8
8
8
8
8
8
7
7
7
7
7
7
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
5
5
5
5
5
0.21%
0.18%
0.18%
0.18%
0.18%
0.18%
0.18%
0.18%
0.18%
0.18%
0.16%
0.16%
0.16%
0.16%
0.16%
0.16%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.14%
0.12%
0.12%
0.12%
0.12%
0.12%
General Warehousing and Storage
Asphalt Felts and Coatings
Hotels and Motels
Sporting and Recreational Camps
Motor Vehicle Parts and Accessories
Primary Metal Products, N.E.C.
Chemical and Allied Products, N.E.C.
Residential Care
Industrial and Commercial Machinery and Equipment, N.E.C.
Miscellaneous Nonmetallic Minerals, Except Fuels
Natural Gas Liquids
Natural Gas Transmission
Wood Products, N.E.C.
Gas and Other Services Combined
Meat Packing Plants
Fabricated Rubber Products, N.E.C.
Poultry Slaughtering and Processing
Perfumes, Cosmetics, and Other Toilet
School Buses
Animal Specialties, N.E.C.
Courier Services, Except by Air
Colleges, Universities, and Professional Schools
Minerals and Earths, Ground or Otherwise Treated
Heavy Construction, N.E.C.
Hardwood Veneer and Plywood
Recreational Vehicle Parks and Campsites
Industrial Gases
Animal and Marine Fats and Oils
Beef Cattle Feedlots (Custom)
Coating, Engraving, and Allied Services, N.E.C.
Towing and Tugboat Services
Steam and Air-Conditioning Supply
Cold-Rolled Steel Sheet, Strip, and Bars
Operators of Dwellings Other Than Apartment Buildings
Special Warehousing and Storage, NEC
Air and Water Resource and Solid Waste Management
$18.5 Mil
750 ppl
$5.0 Mil
$5.0 Mil
750 ppl
750 ppl
100 ppl
$5.0 Mil
500 pp!
500 ppl
500 ppl
$5.0 Mil
500 ppl
$5.0 Mil
500 ppl
500 ppl
500 ppl
500 ppl
$5.0 Mil
$0.5 Mil
$18.5 Mil
$5.0 Mil
500 ppl
$17.0 Mil
500 ppl
$5.0 Mil
1,000 ppl
500 ppl
$1.5 Mil
500 ppl
$5.0 Mil
$9.0 Mil
1,000 ppl
$5.0 Mil
$18.5 Mil
N/A
Between 96% and 98%
Approximately 93%
95%
96%
Approximately 96%
Approximately 92%
95%
91%
99%
88%
60%
40%
98%
62%
96%
90%
79%
92%
97%
N/A



















Between 98% and 99%
46%
82%


Between 94% and 98%
92%
99%
Approximately 93%
84%
N/A
97%
83%







Between 73% and 85%
Approximately 93%
98%


Between 87% and 92%
N/A

                                                                                                Attachment 3, page 3 of 13

-------
SICs of 4,345 new Facilities in PCS Database, 1995-1998
SIC
2891
2048
273
8811
3499
3663
1446
7359
2086
2873
2842
2436
4499
3674
9621
3498
1799
2033
4221
3357
8733
8062
5082
1423
7389
7996
7997
3241
facilities
5
5
5
5
5
5
5
5
4
4
. 4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
% of total
0.12%
0.12%
0.12%
0.12%
0.12%
0.12%
0.12%
0.12%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
0.09%
SIC definition
Adhesives and Sealants
Prepared Feeds and Feed Ingredients for Animals and Fowls,
Except Dogs and Cats
Animal Aquaculture
Private Households
Fabricated Metal Products, N.E.C.
Radio and Television Broadcasting and Communications
Equipment
Industrial Sand
Equipment Rental and Leasing, N.E.C.
Bottled and Canned Soft Drinks and Carbonated Waters
Nitrogenous Fertilizers
Specialty Cleaning, Polishing, and Sanitation Preparations
Softwood Veneer and Plywood
Water Transportation Services, N.E.C.,
Semiconductors and Related Devices
Regulation and Administration of Transportation Programs
Fabricated Pipe and Pipe Fittings
Special Trade Contractors, N.E.C.EXCEPT, Base Housing
Maintenance
Canned Fruits, Vegetables, Preserves, Jams, and Jellies
Farm Product Warehousing and Storage
Drawing and Insulating of Nonferrous Wire
Noncommercial Research Organizations
General Medical and Surgical Hospitals
Construction and Mining (Except Petroleum) Machinery and
Equipment
Crushed and Broken Granite
Business Services, N.E.C. PT, Map Drafting Services,
Mapmaking (Including Aerial) and Photogrammetric Mapping
Services
Amusement Parks
Membership Sports and Recreation Clubs
CemenL Hydraulic
SBA definition
500 ppl
500 ppl
$0.5 Mil
N/A
500 ppl
750 ppl
500 ppl
$5.0 Mil
500 ppl
1,000 ppl
500 ppl
500 ppl
$5.0 Mil
500 ppl
N/A
500 ppl
$7.0 Mil
500 ppl
$18.5 Mil
1,000 ppl
$5.0 Mil
$5.0 Mil
100 ppl
500 ppl
S5.0 Mil
$5.0 Mil
$5.0 Mil
750 ppl
% likely sma'
89%
94%
N/A
N/A
97%
Approximately 94% /
89%
93%
87%
Approximately 95%
95%
77%
94%
89%
N/A
95%
Between 98% and 99%
90%
Between 85% and 90%
Approximately 91%
88%
18%
93%
84%
97%
92%
97%
Approximately 87%
                                                            Attachment 3. page 4 of 13

-------
SICs of 4,345 new Facilities in PCS Database, 1995-1998
*
723
2631
7353
2875
3496
1411
2841
8351
3743
3082
4449
3271
7513
3261
3443
^B79
2041
2999
3675
241
7539
3229
5511
3444
9224
2087
4939
3315
2621
4522
4789
#of
facilities
4
4
4
4
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
% of total
0.09%
0.09%
0.09%
0.09%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
SIC definition
Crop Preparation Service for MarkeLExcept Cotton Ginning
Paperboard Mills
Heavy Construction Equipment Rental and Leasing
Fertilizers, Mixing Only
Miscellaneous Fabricated Wire Products
Dimension Stone
Soap and Other Detergents, Except Specialty Cleaners
Child Day Care Services
Railroad Equipment
Unsupported Plastics Profile Shapes
Water Transportation of Freight, NEC
Concrete Block and Brick
Truck Rental and Leasing, Without Drivers
Vitreous China Plumbing Fixtures and China and Earthenware
Fittings and Bathroom Accessories
Fabricated Plate Work (Boiler Shops)
Electronic Components, N.E.C.
Flour and Other Grain Mill Products
Products of Petroleum and Coal, N.E.C.
Electronic Capacitors
Dairy Farms
Automotive Repair Shops, N.E.C.
Pressed and Blown Glass and Glassware, NEC
Motor Vehicle Dealers (New and Used)
Sheet Metal Work
Fire Protection
Flavoring Extracts and Flavoring Syrups, N.E.C.
Combination Utilities. N.E.C.
Steel Wiredrawing and Steel Nails and Spikes
Paper Mills
Air Transportation, Nonscheduled PT, Offshore Marine Air
Transportation Services
Transportation Services, N.E.C.
SBA definition
$5.0 Mil
750 ppl
$5.0 Mil
500 ppl
500 ppl
500 ppl
750 ppl
$5.0 Mil
1,000 ppl
500 ppl
500 ppl
500 ppl
$18.5 Mil
750 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
S0.5 Mil
$5.0 Mil
750 ppl
$21.0 Mil
500 ppl
N/A
500 ppl
$5.0 Mil
1,000 ppl
750 ppl
1,500 ppl
$5.0 Mil
% likely small
75%
Approximately 85%
90%
93%
96%
99%
Approximately 98%
99%
Approximately 91%
93%
89%
95%
Between 89% and 93%
Approximately 94%
95%
92%
92%
75%
86%
N/A
99% '
Approximately 98%
Between 37% and 76%
98%
N/A
91%
72%
Approximately 89%
Approximately 85%
Approximately 98%
85%
                                                           Attachment 3, page 5 of 13

-------
SICs of 4,345 new Facilities in PCS Database, 1995-1998
SIC
8731
4924
5032
'4311
4214
1611
4011
3339
2671
3321
2816
3471
2269
3341
3491
2879
2493
3462
3087
2754
2099
3084
3274
3353
2611
3494
2431
2448
3264
3412
•#of
facilities
3
3
3
3
3
3
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
% of total
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.07%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05% .
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0,05%
0.05%
0.05%
0.05%
SIC definition
Commercial Physical and Biological Research EXCEPT, Aircraft,
Aircraft Parts, and Auxiliary Equipment, and Aircraft Engine Parts,
EXCEPT, Space Vehicles and Guided Missiles, their Propulsion
Units, their Auxiliary Equipment and Parts
Natural Gas Distribution
Brick, Stone, and Related Construction Materials
United States Postal Service
Local Trucking With Storage
Highway and Street Construction.Except Elevated Highways
Railroads, Line-Haul Operating
Primary Smelting and Refining of Nonferrous metals, Except
Copper and Aluminum
Packaging Paper and Plastics Film, Coated and Laminated
Gray and Ductile Iron Foundries
Inorganic Pigments
Electroplating, Plating, Polishing, Anodizing, and Coloring
Finishers of Textiles, N.E.C.
Secondary Smelting and Refining of Nonferrous Metals
Industrial Valves
Pesticides and Agricultural Chemicals, NEC
Reconstituted Wood Products
Iron and Steel Forgings
Custom Compounding of Purchased Plastics Resins
Commercial Printing, Gravure
Food Preparations, N.E.C.
Plastics Pipe
Lime
Aluminum Sheet, Plate, and Foil
Pulp Mills
Valves and Pipe Fittings, N.E.C.
Millwork
Wood Pallets and Skids
Porcelain Electrical Supplies
Metal Shipping Barrels. Drums. Kegs, and Pails
SBA definition
500 ppl
500 ppl
100 ppl
-
$18.5 Mil
$17.0 Mil
1,500 ppl
750 ppl
500 ppl
500 ppl
1,000 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
750 ppl
750 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 DDl
% likely srm
97%
84%
95%
N/A
Between 96% and 99%
Between 86% and 96%
N/A
Approximately 97%
83%
90%
Approximately 85% ^^
99%
89%
88%
86%
83%
85%
89%
94%
96%
95%
88%
66%
Approximately 96%
Approximately 86%
88%
98%
100%
86%
89%
                                                            Attachment 3, page 6 of 13

-------
SICs of 4,345 new Facilities in PCS Database, 1995-1998
g^ facilities
sW
2893
2096
3231
3251
2865
2822
2672
3363
2426
2451
2514
3365
3533
5191
5153
5149
*
5flS§
1794
5075
5441
4785
4731
1481
4512
1541
1622
4111
7532
9223
9199
m
SW9

2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
% of total
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
SIC definition
Steel Pipe and Tubes
Printing Ink
Potato Chips, Com Chips, and Similar Snacks
Glass Products, Made of Purchased Glass
Brick and Structural Clay Tile
Cyclic Organic Crudes and Intermediates, and Organic Dyes and
Pigments
Synthetic Rubber (Vulcanizable Elastomers)
Coated and Laminated Paper, N.E.C.
Aluminum Die-Castings
Hardwood Dimension and Flooring Mills
Mobile Homes
Metal Household Furniture
Aluminum Foundries
Oil and Gas Field Machinery and Equipment
Farm Supplies
Grain and Reid Beans
Groceries and Related Products, N.E.C.
Drilling Oil and Gas Wells
Durable Goods, N.E.C.
Excavation Work
Warm Air Heating and Air-Conditioning Equipment and Supplies
Candy, Nut, and Confectionery Stores
Fixed Facilities and inspection and Weighing Services for Motor
Vehicle Transportation
Arrangement of Transportation of Freight and Cargo
Nonmetallic Minerals Services, Except Fuels
Air Transportation, Scheduled
General Contractors - Industrial Buildings and Warehouses
Bridge, Tunnel, and Elevated Highway Construction
Local and Suburban Transit
Top, Body,,and Upholstery Repair Shops and Paint Shops
Correctional Institutions
General Government, N.E.C.
Testing Laboratories
Membership Organizations. N.E.C.
SBA definition
1,000 ppl
500 ppl
500 ppl
500 ppl
500 ppl
750 ppl
1,000 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
100 ppl
100 ppl
100 ppl
500 ppl
100 ppl
$7.0 Mil
100 ppl
$5.0 Mil
$5.0 Mil
$18.5 Mil
$5.0 Mil
1,500 ppl
$17.0 Mil
$17.0 Mil
$5.0 Mil
$5.0 Mil
N/A
-
$5.0 Mil
$5.0 Mil
% likely small
Approximately 85%
92%
93%
96%
90%
Approximately 86%
Approximately 93%
89%
89%
96%
83%
94%
95%
94%
97%
96%
95%
97%
99%
99%
96%
98%
90%





/

















Between 97% and 99%
85%

Approximately 94%
Between 88% and
Between 71% and
90%
99%
N/A
N/A
92%
97%
97%
90%






                                                            Attachment 3, page 7 of 13

-------
SICs of 4,345 new Facilities in PCS Database, 1995-1998
SIC
8422
7948
5211
7549
5399
7216
1044
6531
5999
5983
4925
7694
2022
3625
2051
3592
3612
3841
3694
3699
2076
2032
3728
5084
3732
3812
3825
3537
3826
2046
913
1021
2047
2281
#of
facilities
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
% of total
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.05%
0.02%
0.02%
0.02%
0.02%
SIC definition
Arboreta and Botanical or Zoological Gardens
Racing, Including Track Operation
Lumber and Other Building Materials Dealers
Automotive Services, Except Repair and Carwashes
Miscellaneous General Merchandise Stores
Drydeaning Plants, Except Rug Cleaning
Silver Ores
Real Estate Agents and Managers
Miscellaneous Retail Stores, N.E.C.
Fuel Oil Dealers
Mixed, Manufactured, or Liquefied Petroleum Gas Production
and/or Distribution
Armature Rewinding Shops
Natural, Processed, and Imitation Cheese
Relays and Industrial Controls
Bread and Other Bakery Products, Except Cookies and Crackers
Carburetors, Pistons, Piston Rings, and Valves
Power, Distribution, and Specialty Transformers
Surgical and Medical Instruments and Apparatus
Electrical Equipment for Internal Combustion Engines
Electrical Machinery, Equipment, and Supplies
Vegetable Oil Mills, Except Com, Cottonseed, and Soybean
Canned Specialties
Aircraft Parts and Auxiliary Equipment, N.E.C.
Industrial Machinery and Equipment
Boat Building and Repairing
Search, Detection, Navigation, Guidance, Aeronautical, and
Nautical Systems and Instruments
Instruments for Measuring and Testing of Electricity and Electrical
Signals
Industrial Trucks, Tractors, Trailers, and Stackers
Laboratory Analytical Instruments
Wet Com Milling
Shellfish
Copper Ores
Dog and Cat Food
Yarn Spinning Mills
SBA definition
$5.0 Mil
$5.0 Mil
$5.0 Mil
$5.0 Mil
$5.0 Mil
$3.5 Mil
500 ppl
$1.5(6)
$5.0 Mil
$9.0 Mil
$5.0 Mil
$5.0 Mil
500 ppl
750 ppl
500 ppl
500 ppl
750 ppl
500 ppl
750 ppl
750 ppl
1,000 ppl
1,000 ppl
1,000 ppl
100 ppl
500 ppl
750 ppl
500 ppl
750 ppl
500 ppl
750 ppl
S3.0 Mil
500 ppl
500 ppl
500 ppl
% likely srm
92%
93%
87%
98%
93%
Approaching 100%
91%
Between 87% and 96%
98%
Between 93% and 98% .
50%
95%
91%
Approximately 98%
96%
88%
Approximately 95%
93%
Approximately 95%
Approximately 95%
Approximately 1 1 %
Approximately 95%
Approximately 97%
97%
99%
Approximately 93%
91%
Approximately 95%
91%
Approximately 88%
95%
77%
85%
75%
                                                             Attachment 3, page 8 of 13

-------
SICs of 4,345 new Facilities in PCS Database, 1995-1998
•
2061
1011
2064
912
851
782
752
721
2075
2085
253
2095
181
2824
722
2037
1795
1771
^61
2021
1479
1475
2023
2024
2026
2035
1094
2673
1061
2711
2731
2038
1231
1222
2741
1099
B284
2771
#of
facilities
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
% of total
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
SIC definition
Cane Sugar, Except Refining
Iron Ores
Candy and Other Confectionery Products
Finfish
Forestry Services
Lawn and Garden Services
Animal Specialty Services, Except Veterinary
Crop Planting, Cultivating, and Protecting
Soybean Oil Mills
Distilled and Blended Liquors
Turkeys and Turkey Eggs
Roasted Coffee
Ornamental Floriculture Nursery Products
Manmade Organic Fibers, Except Cellulosic
Crop Harvesting, Primarily by Machine
Frozen Fruits, Fruit Juices, and Vegetables
Wrecking and Demolition Work
Concrete Work
Roofing, Siding, and Sheet Metal Work
Creamery Butter
Chemical and Fertilizer Mineral Mining, N.E.C.
Phosphate Rock
Dry, Condensed, and Evaporated Dairy Products
Ice Cream and Frozen Desserts
Fluid Milk
Pickled Fruits and Vegetables, Vegetable Sauces and
Seasonings, and Salad Dressings
Uranium-Radium-Vanadium Ores
Plastics, Foil, and Coated Paper Bags
Ferroalloy Ores, Except Vanadium
Newspapers: Publishing, or Publishing and Printing
Books: Publishing, or Publishing and Printing
Frozen Specialties, N.E.C.
Anthracite Mining
Bituminous Coal Underground Mining
Miscellaneous Publishing
Miscellaneous Metal Ores, N.E.C.
Thread Mills
Greeting Cards
SBA definition
500 ppl
500 ppl
500 ppl
$3.0 Mil
$5.0 Mil
$5.0 Mil
$5.0 Mil
$5.0 Mil
500 ppl
750 ppl
$0.5 Mil
500 ppl
$0.5 Mil
1,000 ppl
$5.0 Mil
500 ppl
$7.0 Mil
$7.0 Mil
$7.0 Mil
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
% likely small
72%
87%
95%
96%
96%
100%
100%
98%
81%









Approximately 93%
N/A
90%
N/A



Approximately 87%
93%
80%
Between
Between
Between
87%
72%
50%
82%
89%
84%
92%
79%
92%
69%
98%
97%
88%
97%
92%
97%
76%
89%
94%


93% and 96%
98% and 99%
98% and 99%



















                                                             Attachment 3, page 9 of 13

-------
SICs of 4,345 new Facilities in PCS Database, 1995-1998
SIC
1081
2429
4222
3511
5148
5146
5141
5085
5039
5014
5012
4971
4923
4813
4812
5251
4612
5311
4141
4131
4119
3999
3996
3965
3955
3861
3842
3827
3823
3792
4725
7941
9661
9229
9221
#of
facilities
1
1
1
1
1
1
x 1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
% of total
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0^02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
SIC definition
Metal Mining Services
Special Product Sawmills, N.E.C.
Refrigerated Warehousing and Storage
Steam, Gas, and Hydraulic Turbines, and Turbine Generator Set
Units
Fresh Fruits and Vegetables
Fish and Seafood
Groceries, General Line
Industrial Supplies
Construction Materials, N.E.C.
Tires and Tubes
Automobiles and Other Motor Vehicles
Irrigation "Systems
Gas Transmission and Distribution
Telephone Communications, Except Radiotelephone
Radiotelephone Communications
Hardware Stores
Crude Petroleum Pipelines
Department Stores
Local Bus Charter Service
Intercity and Rural Bus Transportation
Local Passenger Transportation, N.E.C.
Manufacturing Industries, N.E.C.
Linoleum, Asphalted-Feit-Base, and Other Hard Surface Floor
Coverings, NEC
Fasteners, Buttons, Needles, and Pins
Carbon Paper and Inked Ribbons
Photographic Equipment and Supplies
Orthopedic, Prosthetic, and Surgical Appliances and Supplies
Optical Instruments and Lenses
Industrial Instruments for Measurement. Display, and Control of
Process Variables; and Related Products
Travel Trailers and Cancers
Tour Operators
Professional Sports Clues and Promoters
Space and Research and Technology
Public Order and Safety. N.E.C
Police Protection
SBA definition
$5.0 Mil
500 ppl
$18.5 Mil
1,000 ppl
100 ppl
100 ppl
100 ppl
100 ppl
100 ppl
100 ppl
100 ppl
$5.0 Mil
$5.0 Mil
1,500 ppl
1,500 ppl
$5.0 Mil
1,500 ppl
$20.0 Mil
$5.0 Mil
$5.0 Mil
$5.0 Mil
500 ppl
750 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
55.0 Mil
S5.0 Mil
-
-
-
% likely sm j^b
76%
96%
Between 87% and 100%
Approximately 96%
96%
98%
94%
96%
96%
60%
96%
98%
23%
Approximately 99%
Approximately 99%
97%
Approximately 87% .^^
Between 25% and 42',. ^^
93%
84%
96%
98%
Approximately 82%
98%
94%
94%
95%
92%
92%
95%
96%
88%
N/A
	 I
N/A
N/A
                                                           Attachment 3, page 10 of 13

-------
SIC                % of total
^fe   facilities
SICs of 4,345 new Facilities in PCS Database, 1995-1938



          SIC definition                         SBA definition
% likely small
™1
9121
8744
8741
8249
8222
8099
8059
8051
5181
7991
3691
7692
7349
7261
A
7041
6211
6162
5989
5651
5599
5551
5531
7992
3275
3448
3442
3429
3423
3398
•_
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.02%
0.02%
0.02%
0.02%
4.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
Courts
Legislative Bodies
Facilities Support Management Services EXCEPT, Base
Maintenance EXCEPT, Environmental Remediation Services
Management Services EXCEPT, Conference Management
Services
Vocational Schools, N.E.C.
Junior Colleges and Technical Institutes
Health and Allied Services, N.E.C.
Nursing and Personal Care Facilities, NEC
Skilled Nursing Care Facilities
Beer and Ale
Physical Fitness Facilities
Storage Batteries
Welding Repair
Building Cleaning and Maintenance Services, N.E.C.
Funeral Service and Crematories
Beauty Shops
Organization Hotels and Lodging Houses, on Membership Basis
Security Brokers, Dealers and Flotation Companies
Mortgage Bankers and Loan Correspondents
Fuel Dealers, N.E.C.
Family Clothing Stores
Automotive Dealers, N.E.C., EXCEPT, Aircraft Dealers, Retail
Boat Dealers
Auto and Home Supply Stores
Public Golf Courses
Gypsum Products
Prefabricated Metal Buildings and Components
Metal Doors, Sash, Frames, Molding, and Trim
Hardware, N.E.C.
Hand and Edge Tools, Except Machine Tools and Handsaws
Metal Heat Treating
Copper Foundries
-
N/A
$5.0 Mil
$5.0 Mil
$5.0 Mil
$5.0 Mil
$5.0 Mil
$5.0 Mil
$5.0 Mil
100 ppl
$5.0 Mil
500 ppl
$5.0 Mil
$12.0 Mil
$5.0 Mil
$5.0 Mil
N/A
$5.0 Mil
$5.0 Mil
$5.0 Mil
$6.5 Mil
$5.0 Mil
$5.0 Mil
$5.0 Mil
N/A
1,000 ppl
500 ppl
500 ppl
500 ppl
500 ppl
750 ppl
500 ppl
N/A
N/A
63%
92%
93%
65%
85%
83%
64%
90%
98%
85%
99%
Approaching 100%
99%
100%
N/A
86%
86%
99%
Between 93% and 95%
97%
94%
97%
N/A
Approximately 94%
96%
95%
92%
97%
Approximately 97%
97%
                                                                                                     Attachment 3, page 11 of 13

-------
SIC    . *°!     % of total
       facilities
SICs of 4,345 new Facilities in PCS Database, 1995-1998



          SIC definition                         SBA definition
% likely sma
3356
3354
3325
3324
3299
3297
3724
3281
3463
3263
3255
3253
?111
3086
3083
3061
3053
3011
2895
2892
2836
3291
3561
2833
3669
3661
3651
3646
3543
3641
3624
3585
3577
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
1
1
1
. 1
1
1
1
1
1
1
1
1
1
1
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
Rolling, Drawing, and Extruding of Nonferrous Metals, Except
Copper and Aluminum
Aluminum Extruded Products
Steel Foundries, N.E.C.
Steel Investment Foundries
Nonmetallic Mineral Products, N.E.C.
Monday Refractories
Aircraft Engines and Engine Parts
Cut Stone and Stone Products
Nonferrous Forgings
Fine Earthenware (Whiteware) Table and Kitchen Articles
Clay Refractories
Ceramic Wall and Floor Tile
Leather Tanning and Finishing
Plastics Foam Products
Laminated Plastics Plate, Sheet, and Profile Shapes
Molded, Extruded, and Lathe-Cut Mechanical Rubber Goods
Gaskets, Packing, and Sealing Devices
Tires and Inner Tubes
Carbon Black
Explosives
Biological Products, Except Diagnostic Substances
Abrasive Products
Pumps and Pumping Equipment
Medicinal Chemicals and Botanical Products
Communications Equipment, N.E.C.
Telephone and Telegraph Apparatus
Household Audio and Video Equipment
Commercial, Industrial, and Institutional Electric Lighting Fixtures
Current-Carrying Wiring Devices
Electric Lamp Bulbs arid Tubes
Carbon and Graphite Products
Air-Conditioning and Warm Air Heating Equipment and
Commercial and Industrial Refrigeration Equipment
Computer Peripheral Equipment. N.E.C.
750 ppl
750 ppl
500 ppl
500 ppl
500 ppl
750 ppl
1,000 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
1,000 ppl
500 ppl
750 ppl
500 ppl
500 ppl
500 ppl
750 ppl
750 ppl
1,000 ppl
750 ppl
500 ppl
500 ppl
1,000 ppl
750 ppl
750 ppl
1,000 ppl
Approximately 93%
Approximately 84%
91%
82%
95%
Approximately 88%
Approximately 94%
99%
85%
97%
85%
93%
95%
92%
94%
88%
92%
Approximately 92%
56%
Approximately 90%
89%
93%
89%
Approximately 96%
Approximately 96%
Approximately 93%
Approximately 98%
94%
87%
Approximately 98%
Approximately 90%
Approximately 94%
Approximately 97%
                                                                                                    Attachment 3, page 12 of 13

-------
SICs of 4,345 new Facilities in PCS Database, 1995-1998
£
3569
3568
3449
3564
3451
3559
3556
3554
3552
3546
3545
3544
3524
3523
^189
3484
3469
3711
3566
Total
#of
facilities
1
1
1
1
1
1
1
1
1
1
1
^
i
1
1
1
1
•1
1
1
4345
% of total
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
100.00%
SIC definition
General Industrial Machinery and Equipment N.E.C.
Mechanical Power Transmission Equipment, N.E.C.
Miscellaneous Structural Metal Work
Industrial and Commercial Fans and Blowers and Air Purification
Equipment
Screw Machine Products
Special Industry Machinery, N.E.C.
Food Products Machinery
Paper Industries Machinery
Textile Machinery
Power-Driven Handtools
Cutting Tools, Machine Tool Accessories, and Machinists'
Precision Measuring Devices
Special Dies and Tools, Die Sets, Jigs and Fixtures, and Industrial
Molds
Lawn and Garden Tractors and Home Lawn and Garden
Equipment
Farm Machinery and Equipment
Ordnance and Accessories, N.E.C.
Small Arms
Metal Stampings, N.E.C.
Motor Vehicles and Passenger Car Bodies
Speed Changers, Industrial High-Speed Drives, and Gears

SBA definition
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
500 ppl
1,000 ppl
500 ppl
1,000 ppl
500 ppl


93%
91%
90%
90%
98%
94%
96%
93%
97%
89%
97%
98%
85%
97%
85%
% likely small















Approximately 97%
96%

Approximately 96%
91%



                                                           Attachment 3. page 13 of 13

-------
ATTACHMENT 4: UNIT COST / POUND OF OFFSETS FOR PROCESS WATER
DISCHARGERS AND CONCENTRATION OF POLLUTANTS AT END-OF-PDPE

-------
     BASIS FOR COSTS PER POUND OF OFFSETS AND WATER QUALITY CRITERIA VALUES

Derivation of Unit Costs Used in Calculating Annual Costs of Offsets

       Metals costs: The costs are based on cost functions as given in U.S. EPA, Office of Water.
       Development Document for Proposed Effluent Limitations Guidelines and Standards for the
       Centralized Waste Treatment Industry. December 1998. Multimedia filtration is expected to
       remove 80% of TSS.

              $162 is derived from the expected cost to a medium sized (0.05 MOD), existing
              discharger to install multimedia filtration for metals removal. (The assumption is that
              new dischargers will seek metals offsets from dischargers already at advanced
              treatment.) Assuming that metals comprise 20% of TSS, then going from  15 mg/L TSS
              (common to several industries with metals bearing waste steams) to 3 mg/L TSS will
              reduce 2.4 mg/L of metals.  For a 0.05 MGD discharger, 2.4 mg/L is equivalent to 259
              pounds/ year. The annualized cost of the treatment in 1st quarter 1999 dollars (capital
              costs at 7% over 20 years + O&M) is $41,823.  $41,823 / 259 Ibs = $162/lb.

       BOD costs: $1.21 is the 1999 1st quarter dollar equivalent of $0.484 in 1976 dollars. $0.484 is the
       cost of BOD control based on $0.25 /ib of BOD and TSS control, in which BOD accounted for
       51.7% of the pounds reduced. $0.25 is a benchmark cost used by EPA. The following is standard
       language in EAD preambles: "EPA first calculates the cost per pound of conventional pollutant
       removed by industrial dischargers in upgrading from BPT to BCT candidate technology and then
       compares this cost to the cost per pound of conventional pollutants removed in upgrading
       POTWs from secondary treatment.  The upgrade cost to industry must be less than the POTW
       benchmark of $0.25 per pound (in 1976 dollars). EPA  explained its methodology for the
       development of BCT limitations in July 1986 (51 FR 24974)."

       TSS costs: $1.29 is the 1999 1s1 quarter dollar equivalent of $0.518 in 1976 dollars. $0.518 is the
       cost of TSS control based  on$0.25/ Ib of TSS and BOD control, in which TSS accounted for
       48.3% of the pounds reduced. See above note for further detail.

       Toxic Organics Costs:  The costs are based on cost functions as given in U.S. EPA, Office of
       Water. Detailed Costing Document for the Centralized Waste Treatment Industry. January 1995.
       Carbon adsorption is expected to remove approximately 80% of organics, based on "organics
       option 2" for the organic waste treatment and recovery  subcategory.

              $7,532 is derived from the expected cost to a medium sized (0.05 MGD), existing
              discharger to install granular activated carbon adsorption. (The assumption is that new
              dischargers will seek metals offsets from dischargers already at advanced treatment.)
              Reducing a 2.13 mg/L discharge by 80% L is equivalent to reducing 184 pounds/ year.
              The annualized cost of the treatment in 1SI quarter 1999 dollars (capital costs at 7% over
              20 years + O&M) is 51,385,848. $1,385,848 / 184 Ibs = $7,532/lb.

       Nitrogen and Phosphorous Costs: The costs are based on the costs to buy nutrient reductions
       from non-point sources via installation of BMPs.

              Nitroaen at $11/ Ib - State of North Carolina Department of Environment, Health and

                                                                             Attachment-4. 1 of 3

-------
               Natural Resources. Report of Proceedings on the Proposed Neuse River Basin Nutrient
               Sensitive Waters Management Strategy. July 1997.

               Nitrogen at $20 /lb -based on upper end of $12 - $19 /lb, which are derived after
               adjusting for "safety factors" and cost-share. North Carolina Department of
               Environment, Health and Natural Resources. Guide to Establishing a Point /Nonpoint
               Source Pollution Reduction Trading System for Basinwide Water Quality Management:
               The TarPamlico River Basin Experience.  July 1995. Citing Research Triangle Institute,
               1/95 and Danielson, 1994)

               Phosphorous at $ 17 /lb. Based on $ 16.29/ lb from Faeth, Paul. The Potential for Nutrient
               Trading in Minnesota: The case of the Minnesota River Valley (DRAFT). World
               Resource Institute. February 13, 1998.

        •       Phosphorous at $62 /lb - based  on low end of $62 - $93/ lb, which are derived after
               adjusting for "safety factors" and cost-share. North Carolina Department of
               Environment, Health and Natural Resources. Guide to Establishing a Point /Nonpoint
               Source Pollution Reduction Trading System for Basin-wide Water Quality Management:
               The TarPamlico River Basin Experience.  July 1995. Citing Research Triangle Institute,
               1/95 and Danielson, 1994)

Derivation of Water Quality Criteria Values  Used in Calculating Annual Loads

        Metals criteria of 0.426 mg/ L: Based on summation of freshwater Criteria Continuous
        Concentration for arsenic, cadmium, chromium III, chromium VI, copper, lead, mercury, nickel,
        selenium, silver, thallium and zinc in U.S. EPA Office of Water, National Recommended Water
        Quality Criteria - Correction. April 1999.  0.426 mg/L is a very high estimate of metals
        concentration, since few facilities are likely to discharge all of these metals.  As such, the annual
        costs represent an upper bound.

        BOD Criteria of 16 mg/L: There is no criterion for BOD. We estimate that the most that any
        facility would be allowed to discharge is the BCT value for POTWs. Note that this value
        represents an upper bound estimate, as facilities located on  impaired waters are likely to be
        required to achieve more stringent levels of control.  As such, costs for offsets of BOD are likely
        to be lower, since the amount discharged will be smaller. EPA determined that 16.14 mg /L is
        the long term average performance of POTWs meeting secondary treatment requirements.  July
        1986,51 FR 24974."

        TSS Criteria of 16  mg/L:  There is no criterion for TSS. We estimate that the most that any
        facility would be allowed to discharge is the BCT value for POTWs. Note that this value
        represents an upper bound estimate, as facilities located on  impaired waters are likely to be
        required to achieve more  stringent levels of control.  As such, costs for offsets of  TSS are likely
        to be lower, since the amount discharged will be smaller. EPA determined that 15.84 mg /L is the
        long term average performance of POTWs meeting secondary treatment requirements.   July
        1986, 51 FR 24974.

        Toxic Organics criteria of 2.13 mg/L:  Based on the NSPS  Total Toxic  Organics (TTO) standard
        for the Metal Finishing Point Source Category (40 CFR §433.16).

Attachment-4. 2 of 3

-------
Nitrogen Criteria of 6 mg/L: Tar-Pamlico NSW Implementation Strategy: Phase II, December 8 '
1994. There are no national criteria yet developed for nitrogen, and many states have only
criteria for drinking water standards. 6 mg/L is the total nitrogen monthly average for new
dischargers in the Tar-Pamlico watershed. The Tar-Pamlico was designated as a nutrient
sensitive water by the State of North Carolina,  and is one of the most developed "effluent
trading" programs in the country. As such, the  6 mg/L limit for new dischargers would seem to
be an appropriate criterion to use to estimate costs for these situations on a national level.The
Neuse River Watershed, also a nutrient sensitive water in North Carolina, has proposed limits of
3.5 mg/L for new dischargers. Costs for offsetting loads remaining after meeting a 6 mg/L limit,
may thus be somewhat of an overestimate for some waters.

Phosphorus criteria of 0.1 mg/L: U.S. EPA, Quality Criteria for Water, July 1976 ("Red Book").
Discussion of phosphate phosphorus in streams or other flowing waters, p. 188.
                                                                        Attachment-4. 3 of ?

-------