UMILOU
Environmental Protection
Agency
Office of Pesticides and
Toxic Substances
Washington DC 20460
                            EPA-560/12-80-005
Toxic Substances
Support  Document
Economic Impact Analysis
of Proposed Section 5
Notice Requirements

Part I - Analysis of the Impacts
on the Chemical Industry
of Proposed Section 5 Notice Requirements

Part II - Issue Papers

Proposed Rule   Section 5
Toxic Substances Control Act
                              W    A

-------
                                               EPA-560/ 12~8Q~005
                                               September 1980
          ECONOMIC  IMPACT ANALYSIS OF PROPOSED
              SECTION 5 NOTICE REQUIREMENTS
PART I:  ANALYSIS OF THE IMPACTS ON THE CHEMICAL INDUSTRY
        OF PROPOSED SECTION 5 NOTICE REQUIREMENTS
                 PART II:  ISSUE PAPERS
                 Contract  No.  68-01-5878


                    Project Officer:

                       Sammy K.  Ng
            ECONOMICS  AND  TECHNOLOGY  DIVISION
               OFFICE OF TOXIC SUBSTANCES
                 WASHINGTON, D.C.   20460

          U.S. ENVIRONMENTAL PROTECTION AGENCY
        OFFICE OF PESTICIDES AND  TOXIC  SUBSTANCES
                 WASHINGTON, D.C.   20460

-------
                               TABLE OF  CONTENTS

                                                                           Page

EXECUTIVE SUMMARY

PART I.   ANALYSIS OF IMPACTS ON THE CHEMICAL INDUSTRY OF PROPOSED
          SECTION 5 NOTICE REQUIREMENTS


CHAPTER 1:  Background and purpose  	   1


CHAPTER 2:  A Methodology for Analyzing the Economic impact
            of Regulations implementing Section 5 of TSCA 	   3

     Analytical Method  	   3

     Assumptions About the Methodology   	   8


CHAPTER 3:  Data Sources and Collection	15

     Publicly-Available Data  	  15

     Studies by Arthur D. Little	15

     Industry Representatives Willing to Provide information  	  16

     Meeting of independent Chemical industry and innovation Experts  ...  16

     Public comment on Earlier Notice Proposals 	  17


CHAPTER 4:  impact of Section 5 Notice Regulation	19

     Theoretical Expectations About impacts 	  19

     Theoretical Basis of Section 5 Notice impacts   	  21

     Impact of t1	21

     impact of c' and p'  	22

     Empirical Manifestation of increased
     Costs and Delays in introductions	23

     Problems with Cost Estimation	24

     Summary	24

-------
CHAPTER 5:  Analysis of Comprehensive Program Options  	   27




     Reporting Requirements 	   29




     issue 1.  Confidentiality	31




     issue 2.  Customer Contact	35



                                                                              37
     issues 3 and 4.   importer Provisions  	




     Issue 5.  Exporter Provisions	3^




     issue 6.  Supplemental Reporting 	   ^




     issue 7.  Insufficient Submissions  	   44




     Issue 8.  Processor Reporting   	   ^




     issue 9.  Possession or control	48




     Interrelationships Among  issues 	   50




     Scope of Reporting Requirements 	   51




     Confidentiality and the Other  issues  	   52




     Comparison of comprehensive Program Options   	   55






 CHAPTER 6:   identifying innovative  Chemical  Industry Segments 	   63




     inorganic Chemicals  	   63




     Synthetic High polymers   	   65




     Amphipathic Compounds  	   67




     Elementary Organic Chemicals  	   68




     Organic Chemicals, Not Elsewhere Classified   	   70




     Catalysts	72




     Other Chemical Products   	   73




     innovative  Segments  identified 	   75
                                        11

-------
                                                                           Page

CHAPTER 7:  Measuring the impact  	  79


CHAPTER 8:  Models of Innovation in the Chemical Industry 	  83

     Summary of Literature  	  83

     Literature Review  	  85

     Models of innovation 	  92


CHAPTER 9:  Qualitative Analysis of impact
            on New Chemical introductions 	  97

     Impact on commercialization Activities 	  97

     implications for the Chemical Industry 	 100

     Segment Analysis 	 102

     Summary   .................«••••••••••••• 106

     Macroeconomic Effects  	 107

     Conclusion	HO


PART II.   ISSUE  PAPERS

INTRODUCTION   	 113

CHAPTER 1:  CONFIDENTIALITY	115

     Specification of EPA Options	115

     Analysis of issues	123

     When to Substantiate	128

     Generic information  	 131

     Chemical  identity as Part of Health and Safety Study 	 136


CHAPTER 2:  CUSTOMER CONTACT	141

     Customer Contact Options 	 141


                                       iii

-------
                                                                           pag_e





     Cost of Compiling Customer Lists 	 143




     Summary	1







CHAPTER 3:  DEFINITION OF IMPORTERS	15




     importer Definition Options  	




     Coverage 	




     Cost to industry	155




     Summary	l56






CHAPTER 4:  IMPORTER CONTACT OF FOREIGN MANUFACTURERS/SUPPLIERS 	 157




     Upstream Contact Options 	  ^




     Coverage	15 9







CHAPTER 5:  EXPORTERS	l65




     Exporter Options 	 165




     Coverage	166




     Summary	168






CHAPTER 6:  SUPPLEMENTAL REPORTING  	 169




     Supplemental Reporting Options 	 169




     Evaluation Framework 	 170




     Summary	174






CHAPTER 7:  INSUFFICIENT SUBMISSIONS  	 175




     Insufficient Submission Options  .....  	 175




     Evaluation Framework 	 176




     Summary	I83






                                        iv

-------
CHAPTER 8:  PROCESSOR REPORTING	185

     Introduction 	  185

     Overview	187

     Extent of impacts	190

     Options	192

     Major Differences Among Alternatives 	  195

     Analysis of Economic Costs 	  195

     Appendix:  Differences Between Options 	  205


CHAPTER 9:  POSSESSION OR CONTROL 	  215

     Possession or Control Options  	  216

     industry Costs 	  218

     Cost of Options	225

     Summary	228

     Appendix:  State of the Art—chemical information Science  	  231


APPENDICES

APPENDIX;  VOLUME I

     APPENDIX A:  Cost of Forms	A-l

     APPENDIX B:  Assessment of the Feasibility of Developing
                  a Methodological Framework for Formal
                  Economic impact Analysis  	  B-l

     APPENDIX C:  Discounted Cash Flow Analysis	C-l

     APPENDIX D:  Discussion of the Economic Burden of Section 5
                  Notice Requirements on EPA and Society	D-l

APPENDIX;  VOLUME II

     APPENDIX E:  chemical Segment Profiles 	  E-l

-------
                                    PREFACE
     The attached document is a contractor's study done with the supervision
and review of the Office of pesticides and Toxic Substances of the U.S.
Environmental Protection Agency.  The purpose of the study is to analyze the
potential economic impact of proposed section 5 notice requirements.  These
requirements were prepared by the EPA Office of Pesticides and Toxic
Substances to implement section 5 of the Toxic Substances Control Act.

     This report was submitted in fulfillment of Task Order Number 3 of
Contract Number 68-01-5878 by ICF Incorporated.  Work was completed as of
August 1980.

     This report (consisting of a report and two volumes of appendices) is
being released and circulated to coincide with publication in the Federal
Register of a notice announcing the availability of the report and of a draft
regulatory analysis.

     The study is not an official EPA publication.  All comments received by
EPA will be considered in establishing the final analysis to be published
along with the final regulations.  Prior to final promulgation of the section
5 notice requirements, the accompanying document shall have standing in any
court proceeding only to the extent that it represents the views of the
contractor who prepared it.  The document cannot be cited, referenced, or
represented in any respect in any such proceedings as a statement of EPA's
view regarding the subject, the industry or the economic impact of the
regulation.
                                       VI1

-------
                                EXECUTIVE SUMMARY
      ICF  Incorporated  was  issued  a  task  order  by  the  Environmental Protection
 Agency  to analyze  the  economic  impact on the chemical industry  of  the  Toxic
 Substances Control Act (TSCA) section 5  rules.  These rules  would  require
 chemical  firms  to  notify EPA  about  new chemicals  before  introduction into
 commerce.   This report presents the results of  the  analysis.

      When the rules are promulgated, there will be  six distinguishable
 consequences for the chemical industry's new chemical introduction process.
 These are:

      1.    Direct costs associated with completing reporting  require-
           ments;
      2.    Indirect costs associated with the delay  in the  introduc-
           tion  of  new  chemicals;
      3.    Uncertainty  regarding direct costs;
      4.    Uncertainty  regarding length of delay;
      5.    Possible trade secret disclosure; and
      6.    Possible restrictions on  the ability  to market the
           chemical.

      The  segments  of the chemical industry most likely to  be affected  by the
 notice  requirements are catalysts,  surfactants, cyclic intermediates,  rubber
 processing chemicals,  plasticizers,  synthetic organic chemicals  (NEC),
 adhesives and sealants, industrial  inorganic chemicals (NEC). and  plastics and
 resins.

      The  section 5 notice  requirements have a unique  impact.  Unlike most reg-
 ulations,  these do not require  companies to  (1) cease production of specific
 chemicals,  (2)  install equipment, or (3)  incorporate  items directly into the
 product that will  raise its cost.   Rather, the  notice regulation requires that
 the administrative apparatus  within companies producing new chemical entities
 notify EPA of their  intention to  introduce them into  commmerce.  Although
 there are  readily  measurable  costs  associated with  notifying EPA,  these costs
 are not the major  source of the economic  impact.  Rather,  the uncertainty
 about EPA's action on  a notice  (which is  a function of the very existence of
 TSCA  not  just Section  5) and  the  other "difficult to  quantify" consequences
 only  indirectly related to the  notification rule  (e.g., the decision to
 undertake  a health and  safety study, protection of  trade secrets revealed in
 the notice, potential  restrictive action) will  be the major source  of  impact.

     The  impact of  these direct and  indirect effects  will occur when the
companies make  capital  allocation decisions.  The direct cost of complying
with proposed reporting requirements is  estimated,  based on the October 16,
 1979 form,  to range  from $1,000 to  $9,000 per new chemical.  However,  this is
 the least  important  cost factor influencing chemical  company decision  making.
When firms  make capital allocation  decisions — whether or not to  invest in a
new product, for example — the uncertainties associated with possible EPA
actions will generally  outweigh the direct costs of complying with  notice
 requirements.
                                       IX

-------
      The nonquantifiable uncertainty consequences represent a major impediment
 to new product development if firms are unable to obtain adequate information
 about EPA's rationale in making notice decisions.  Stated differently, the
 economic impact of section 5 notice requirements is primarily a function of
 EPA's regulatory practices, not the administrative costs of notification.

      Smaller companies may suffer disproportionately from the uncertainty
 caused by notice requirements because they generally have less access  to
 expertise about government decision making and normally direct fewer resources
 toward overhead activities which could provide this expertise.   In addition,
 these companies may also face relatively larger direct costs per  new product
 as a result of their reliance on low volume products which may  generate lower
 absolute revenues and require the submission of more notices.

      One likely result of these regulations will be a reduction in the number
 of new chemicals introduced into the market in the  short  run.   In the  long
 run, chemical companies can be expected to shift their  innovative activities
 to "safe" chemicals.   What constitutes a "safe"  chemical  depends  on the state
 of chemical knowledge and EPA administrative practices  with regard to  various
 chemicals.   The point is that efforts to reduce  the  uncertainties created by
 notice requirements are likely to influence the  mix,  if not the volume, of
 chemicals available in the marketplace.   While  it is clear  that any initial
 drop in the introduction of new chemicals cannot help the  current economy,  it
 is difficult to estimate the  magnitude  of any  near-term macroeconomic
 effects.   This inability is attributable  in part to  limitations in techniques
 of macroeconomic analysis,  but  more  importantly  to  the  uncertainty of  EPA's
 use of notice requirements data  and  the  lack of data on chemical  innovation.

      On the basis of  current  data, it is  difficult  to estimate  either  the in-
 dustry's  rate of new  chemical introductions or the  extent  of the  reduction
 caused by the section 5  notice  requirements.  Even with all of  the necessary
 data to measure the current rate and the  likely  reduction  (data that industry
 has been  reticent to  provide) ,  it  is doubtful that  the  level of the reduction
 could  be  predicted ex ante.

     Without  the  ability  to quantify any  but the direct costs of  these
 regulations or  to estimate the change in  the rate and composition of new
 chemical  introductions, economic theory must remain  the primary source of
 insight into  the effects of notice requirements.  Theory would  predict that a
 drop in the rate of introductions will probably occur,  that the industry and
 the economy will probably suffer at minimum  the costs  involved  in the
 transition from  "unsafe" to "safe" chemicals and that,  in  the long  run,  the
 industry may be composed of fewer, larger competitors better able to absorb
 the direct costs and regulatory uncertainty associated with the section 5
notice requirements.

-------
                   PART I
ANALYSIS OF IMPACTS ON THE CHEMICAL INDUSTRY




        OF PROPOSED SECTION 5 NOTICE




                REQUIREMENTS

-------
                                    CHAPTER 1

                             BACKGROUND AND PURPOSE
     The Environmental Protection Agency  (EPA) published proposed premanu-
facture notification rules in the Federal Register on January 10, 1979  (44 FR
2242), hereafter referred to as the January 10tn proposal under section 5 of
the Toxic Substances Control Act (TSCA).  After reviewing public comments, EPA
reproposed a portion of these rules in the Federal Register of October 16,
1979 (44 FR 59764), hereafter referred to as the October 16th proposal.

     A preliminary economic impact analysis, prepared by Arthur D. Little,
Inc. (ADL) for EPA, was published concurrently with the January 10tn
proposal.  This analysis estimated the costs of preparing section 5 notices
and briefly examined their potential impact on new chemical development.  In
addition, a unit cost analysis for the October 16 proposal form was also
completed by ADL.JL/

     EPA engaged ICF Incorporated to conduct an economic analysis of several
regulatory alternatives (developed by EPA) for implementing section 5 and a
study of the impacts of this program on the chemical industry.  Specifically,
ICF was asked to extend the ADL preliminary analysis to address impacts on
industry performance and structure and impacts on the economy.

     ICF Incorporated performed six tasks:

     1.   developed project workplan;

     2.   collected data;

     3.   assessed feasibility of formal approaches to economic
          impact analysis;

     4.   analyzed regulatory provisions;

     5.   performed cost analysis of processor reporting and minimum
          guidance reporting; and
    .I/These documents, impact of TSCA Proposed Premanufacture Notification
Requirements, December 1978, and Estimated Costs of Preparation and Submission
of Reproposed Premanufacture Notice Form, October 1979, are available from the
EPA Industry Assistance Office.

-------
                                      -  2  -


     6.   performed economic impact analysis (based on previous
          tasks).

     Tasks 1 through 5 were necessary to provide the framework for the
economic analysis under Task 6.  To a considerable extent,  the output of these
tasks determined the appropriate approach to Task 6.

     The details of the methodology are presented in the next chapter.

-------
                                      - 3 -
                                    CHAPTER 2

                 A METHODOLOGY FOR ANALYZING THE  ECONOMIC  IMPACT
                  OF REGULATIONS  IMPLEMENTING  SECTION 5 OF TSCA
     This chapter presents an ideal methodology for measuring the economic
impact of the new chemical notice information provisions in section 5 of the
TSCA and is organized around two topics:   (1) the proposed analytical method,
and  (2) assumptions made in creating the methodology.  This chapter is similar
to the methodology published in the Federal Register on June 10, 1980.

     The analytical method section of the chapter is divided into two
subsections.  The first subsection describes an overall approach to the
economic analysis of regulations that was amended to account for the unique
aspects of this set of regulations.  The second subsection lists the
particular questions that were addressed.

     The assumptions section discusses one procedural and five substantive
assumptions used in developing this methodology.

     On the basis of comments received concerning the methodology and data
limitations faced during the study, we amended the methodology so that greater
emphasis was placed on the qualitative factors than on the quantitative
factors presented in this chapter.
ANALYTICAL METHOD

Approach to Economic impact Analysis

     The first task of an economic impact analysis is to create a baseline
against which the effects of the notice regulations can be measured.  This
baseline should constitute an economic picture of the industry independent of
the TSCA regulations but dependent on other changes in the regulatory and
legal environment (e.g., liability laws, OSHA, CPSC).  In order to develop a
baseline, substantial data about the companies within the industry and the
industry's role in the U.S. economy are needed.  Because the chemical industry
produces a diverse set of products which will be impacted differently by
premanufacture requirements, it is necessary to segment the industry.
Considerable data is needed for each segment, and for this analysis it is
important to obtain:

     •    production data over time that distinguishes between
          intermediate and final products, notes capacity
          utilization rates and expansion plans and,  when possible,
          denotes new vs. old product relationships;

-------
                                      -  4 -


     •    sales volumes or uses over time;

     •    price histories;

     •    existence and prices of substitutes;

     •    financial profiles of producers;

     •    import/export trends;

     •    engineering processes;

     •    research and development expenditures;

     •    decision rules and processes for  new chemical
          introductions; and

     •    amount, type, and cost of health  and safety testing.

     This data permits tentative assessments of:

     •    market concentration;

     •    degree of vertical and horizontal integration;

     •    competitiveness and forms of competition;

     •    capital budgetary decision processes; and

     •    demand and supply determinants.

     Using this picture of the industry, the analysis requires a demand
forecast for the industry's products and a  concomitant supply forecast.

     Demand Forecast.  It is rare that new  chemicals subject to section 5
notice requirements will be consumed directly by individual consumers.  In
almost every instance the chemical industry segment  produces intermediate
products that are substantially altered before reaching the final consumer.
Tracing the demand for these products is exceedingly difficult because of this
level of detail; moreover, little or no existing econometric forecasts are
available.  Thus, we are forced to rely on  a projection technique unique to
the segment or on projections provided by companies  within the impacted
segments.

     We use historical trend lines to project future developments when such
trends appear reliable, and in other cases  we use industry-provided
projections.   The choice was necessarily judgmental.

     Supply Forecast.  Once the overall baseline demand projection is
prepared, it may be necessary to differentiate how this demand will affect
individual chemical manufacturers.  By examining the following areas,
potential recipients of the changed demand  might be  identified:

-------
                                      - 5 -
     •    whether the markets are geographically isolated;

     •    relative strength of competitors and their varying
          abilities to attract capital;

     •    individual company product mix and how it might change
          over time; and

     •    whether foreign competition will have a significant impact
          on domestic production.

     These factors are important because they can lead to an alteration in the
structure of the segment.  If they significantly affect the structure, they
will affect the economics of the industry and should be reflected in the
baseline forecast.

     Changes as a Result of Section 5.  The next step is to identify the
economic impacts on the industry that may result from section 5 notice
requirements .  If the baseline is properly constructed, sufficient
information will have been generated to provide not only a benchmark against
which to measure the effects of section 5 notice requirements, but also a
useful analytic framework for projecting the economic consequences resulting
from implementation of the requirements.

Special Questions

     Regulatory impacts can be measured at three different levels:  at the
level of individual companies affected by regulation, at the more aggregate
level of an industry affected by regulation, and at the broad level of the
U.S. economy.  The impacts on individual companies and the impacts on industry
are derived primarily from an analysis of (1)  innovation,  (2)  market
structure, and (3) profits.  Impacts on the U.S. economy are derived from the
analysis of industry impacts.

     The five most critical subjects that we examined in our analysis were
innovation, market structure, profits, GNP and employment, and foreign trade.
Each is discussed below.   (This should not be construed to mean that we did
not address all of the elements of a typical economic analysis, as shown
above, only that these are the five most critical.)

     Innovation.   Innovation is a broad term which is used throughout this
report to refer only to the new chemical entity development process.  Process
innovation and creation of new mixtures are excluded from  the definition in
this report.

     The first step in examining the impact of premanufacture notification on
the stream of new chemicals introduced each year was to characterize the
historical pattern for each segment.  Using this baseline  we then tried to
estimate the way in which section 5 notice requirements would alter that
stream.  From a methodical survey of previous studies, trade journals, buyer's
guides, and U.S.  International Trade Commission data on chemical production,
we developed an understanding of the role that innovation  plays in each

-------
                                      - 6 -


segment.  Lacking detailed industry input, we were unable to develop detailed,
historical, new product introductions in terms of research and development,
sales, physical quantities, and expected and actual returns.

     Having characterized the baseline stream of new chemicals on the basis of
public information, we attempted to explore the chemical firms' decision-
making process regarding new chemical introduction.  Some information about
this process was available in the comments made by firms about the section 5
notice requirement proposals, and more information was obtained from a survey
of trade journals.  However, the bulk of data on firm decision-making
processes was, of course, in the possession of chemical firms and, in some
cases their customers.  Lacking direct access to this data,  we had to rely on
industry experts to provide insight into these processes.

     With this information we attempted to answer the following questions for
typical chemical companies in different segments of the industry:

     1.   How much is budgeted for R&D and how is that amount
          determined?

     2.   What are the key factors in the decision to produce a new
          chemical?

     3.   How important is time in the R&D and the commercialization
          of new chemicals?

     4.   How important is confidentiality to the success of a new
          chemical venture?

     Combining the information on the baseline stream of new chemicals, firm
decision-making processes,  and the estimates of the costs of completing the
forms developed earlier,  we intended to estimate the impacts of the section 5
notice program on innovation in each segment.   As subsequent pages will
reveal, segment detail proved more elusive in practice than in theory.

     Market Structure.  There are four questions about market structure that
we addressed.   The information came from United States International Trade
Commission (USITC)  data,  buyers'  guides, trade journals, other literature
sources,  and industry experts.

     1.    Who  are the competitors in each segment?

     2.    What form does  their competition take—pricing, product
          differentiation,  new product development, other?

     3.    What role do small firms play within the segment?

     4.    Does company size influence the kinds of new chemicals
          produced?

     Profits.   Four questions concern the role of profits in segment growth:

-------
                                      - 7 -
     1.   What profits are associated with new chemicals and how do
          they compare to profits on established chemicals?

     2.   To what extent are companies dependent on profits in the
          development of new chemicals?

     3.   What do companies do with their profits and other cash
          flows?  What are their sources and uses of funds?

     4.   With what degree of certainty are new chemical profit
          streams estimated?

     The literature, comments from companies, and industry experts provided
the information to answer these questions.  In particular, question four has
been discussed in many of the public comments on prior proposals.

     GNP and Employment.  In this part of the analysis (and the foreign trade
area as well) we intended to aggregate segment projections to determine the
overall impact of notice requirements on the U.S. economy.  Several economic
forecasters  (DRI, Chase Econometrics, etc.)  of the U.S. economy have
previously addressed these questions, and we intended to replicate their basic
methodological approach using segment projections.  Given the lack of these
quantitative segment projections, we instead addressed these related questions:

     1.   What is the relationship between chemical industry indicators and
          general economic indicators?

     We reviewed existing model data on the GNP increment provided by chemical
segments to estimate the degree to which a given percentage change in a
chemical segment growth would be reflected in a GNP percentage change.  We
also reviewed input-output data to determine the effect the chemical industry
has on the economic performance of other individual industries.

     2.   To what extent is innovation in other industries dependent on
          chemical industry innovation?

     We relied for the most part on the literature and on independent experts
highly experienced in the chemical industry for this information.  The
information provided a multiplier to assess the adverse impact on other
industry innovation.

     3.   What has been the relationship between employment and sales in the
          chemical industry?

     To adequately assess this relationship, we have obtained chemical
industry sales information from the U.S. Department of Commerce and employment
information including productivity from the U.S. Bureau of Labor Statistics
(BLS).   We have also gathered information on other potential impacts on
chemical industry employment from the literature and from chemical industry
interviews.   As we explain later, the inability to quantify the program's
direct uncertainty effects precluded the quantification of the GNP and
Employment impacts.

-------
                                      - 8 -
      Foreign Trade.   Most  of  the foreign trade questions have been answered
 through the literature  search, discussions with trade associations, and
 interviews with government officials.  A few would have been aided by greater
 industry input.

      1.    What types  of chemicals are handled in foreign trade?

      2.    Do U.S.  chemical firms have many foreign subsidiaries?
           What types  of relationships exist among them?  How do
           multinationals fit  into the particular segments?

      3.    Does international  competition vary across the industry
           segments?   How do foreign chemical firms' R&D programs
           compare  to  U.S.  firms' programs?

      4.    To what  degree are  chemicals developed by U.S. firms
           licensed to foreign firms for production?

      5.    What are the  factors influencing the locating of R&D, test
           marketing,  and production activities?
 ASSUMPTIONS ABOUT THE METHODOLOGY

     To perform this particular analysis we made certain procedural and
 substantive assumptions.

 Procedural Assumption

     Without question, the most important procedural assumption was that
 industry could be used as a source of specific data about particular questions
 raised during our review of publicly available data.  This was an important
 assumption because, although the industry may not be completely objective, it
 was a potentially excellent source of data.  It was our intention to use the
 industry—through interviews and meetings with trade associations—in a very
 directed way.  We wanted to ask them to supply data in response to specific
 questions which were developed after our analysis of the publicly-available
 data.  The combination of public and industry sources certainly provided a
more substantive and sound analysis.  Tthe trade associations we contacted
 explained that they were not in possession of the extremely confidential
 detailed data we sought, and generally they encouraged us to interview
 individual companies.

     If, however,  we had relied solely on the individual companies to tell us
what they would do, we believe we might have gathered potentially
 unsubstantiable data for two reasons.  First, regulations implementing TSCA
are not final at this time.  Therefore, each potential industry interviewee
would have a different picture of the TSCA regulatory program in his mind and
would tell how he will react to the program as he perceives it.  Second, we
 think it unlikely that industry interviewees would be able to distinguish

-------
                                      - 9  -
section 5 impacts from those of TSCA in general.  Given these problems, we
chose instead to rely on a selected group of independent industry experts as a
major source of data.

Substantive Assumptions

     Five substantive assumptions essential to the form of analysis and
methodology were:

     1.   The baseline projection assumed that no TSCA exists and
          covered the entire chemical industry.

     2.   To the extent possible, results were quantified.  In most
          cases, the direct impact was provided.

     3.   The non-baseline projections of chemical industry supply
          and demand were based on comprehensive program options.

     4.   The appropriate segmentation scheme was chemical-entity
          based.

     5.   Previous estimates of the number of chemicals that will be
          introduced without TSCA are reasonable and are unlikely to
          be improved through further analysis.

The Baseline Projection Will Assume No TSCA and Cover the Entire Industry

     This assumption has two dimensions:  the scope of the chemical industry
to be studied and the choice of a baseline regulatory program.  Section 5 has
no direct impact on currently manufactured chemicals.  Therefore, it can be
argued that the regulations cannot directly impact these chemicals and that
the scope of the economic analysis should only be directed at new chemicals.
We would note, however, that in the long run the potential lack of
introduction of new chemicals would affect existing chemical markets, the
industry, and the U.S. economy.  Accordingly we concluded that the appropriate
baseline should include both new and existing chemicals consistent with the
latter approach.

     The second dimension is the decision to assume no TSCA as the baseline.
If the baseline were any set of section 5 regulations that could potentially
cause significant changes in the industry, it would be practically impossible
to develop a defensible data base to project impacts, because no usable
historical data would exist.  Consequently, we believe the baseline should be
the "no-statute" case.  Although we will assess the difference among various
program formulations, we plan for these to be projections from a baseline that
assumes no TSCA.  Performing the analysis this way allows us to consider
historical data about the industry in developing working hypotheses about
demand factors, supply factors, market structure, degree and forms of
competition, and concomitant conduct and performance.

-------
                                      - 10  -
 To the Extent Possible,  Results  Will  Be  Quantified.   In Most  Cases Range
 Estimates Will Be Provided.

      Although attempts have  been made to measure precisely  the effects of
 changes in new chemical introductions, ICF has, in a  qualitative analysis,
 projected that the weight of program  changes will fall on the companies, the
 industry, and the U.S. economy.  Chemicals are not economic units; they are,
 rather, the prime factor in  an analysis  of this type.

      We believe that the results of proposed changes  in chemical introductions
 cannot be assessed qualitatively,  because the data we have will not yield
 defensible,  measurable estimates.  We  have, however, established a baseline
 with historical quantitative data.

      An example of the problems  involved in quantifying the impact of section
 5 is the difficulty in estimating  the  effect on employment.   Because section 5
 is a premanufacture rule,  it does  not  directly cause existing plants to close
 (though it may indirectly cause  plants to close due to a restructuring of the
 industry), but instead impedes development of potential new plants.  For
 example, if  a new chemical simply  replaced an old chemical, then it would be
 possible to  associate only slight  employment impacts with changed income and
 profit flows to the industry.  Clearly, measurement of this kind of
 consequence  is subject to considerable error, both because  it is removed from
 the activity directly affected and because it is based on other estimates of
 direct impacts.

 The Non-baseline  Projection  Was  Derived from Comprehensive Program Options

      Comprehensive program options (CPOs) are combinations of alternatives
 available to resolve  numerous regulatory issues.  If two items in a new
 regulatory program were  unresolved and there were 3 options for each, there
 would be nine possible CPOs.  There are two reasons why CPOs  should delineate
 the boundaries of  the economic projections.  First, the alternative to
 comprehensive analysis is  issue-specific analysis.  This kind of analysis
 requires specificity  of  data well  beyond that currently available.

      Second,  it is probable  that the bulk of the economic impact may be the
 existence  of  the program  itself  without regard to the particular implementation
 plan  (choice  among  issue alternatives).  Thus, we believe that broad options
 which combine  a set of issue alternatives must be examined; but they are only
 examined  to provide indications of the relative impact.  For example, the
 economic  impact of any confidentiality provision, no matter how structured,
may be substantial.   Likewise, even without a form, the requirement to notify
may represent  the  bulk of the impact.   In both cases, the incremental
 difference among the  individual  issue options may be quite small, but,  if
 combined with  several other issues' alternatives, that difference might be
 considerable.

     Our approach, therefore, is to analyze the comprehensive program options
which have been developed by EPA.  They are based on combinations of program

-------
                                     - 11 -
element  (issue-specific) options.  This approach provides useful information
about the relative costs of various CPOs.

     To conclude, we favored the CPO approach because (1) the major economic
impact was the existence of the program without regard to the issue-specific
option chosen, (2) the data do not exist to perform issue-specific analysis
with any rigor, and  (3) the reasonableness of the regulations depended on the
marginal cost-effectiveness of the regulatory program, which should be readily
identifiable from the CPO analysis.

The Appropriate Segmentation Scheme was Chemical-Entity Based

     In order to estimate the magnitude of the effect of premanufacturing noti-
fication on the introduction of new chemicals, it is necessary to characterize
past chemical innovation.  For a sector as large and diverse as the chemical
industry, this analysis is greatly facilitated by disaggregating the industry
into smaller units more suitable for study.

     Because of the availability of economic data from the U.S. Bureau of the
Census, it is tempting to divide the industry along Standard Industrial
Classification (SIC) lines.  Unfortunately, these categories bear no relation
to the need for filing section 5 notices.  Instead, SICs fragment individual
chemicals according to their source and use—both distinctions to which the
requirement to provide notice under section 5 is blind.   A single compound
like methanol, for example, is classified in two segments depending upon
whether it is derived from wood (SIC 2861)  or petroleum  (2869).  Similarly,
ammonium hydroxide is listed separately as an industrial inorganic chemical
(2819), a household cleanser (2842), and a nitrogenous fertilizer (2873),
despite the fact that only one notice would have to be submitted if the
chemical were new.

     In addition to assigning chemicals to many different categories, SICs
also include mixtures such as paints (285).  Such combinations are not subject
to section 5 regulations and would be affected only indirectly through their
components.

     SICs are awkward vehicles for studying chemical innovation because they
are not chemical groupings.  An alternative segmentation scheme, based on
common divisions of chemicals and chemical science, is shown in Exhibit 2-3.
The portion of the analysis that focuses directly on chemical innovation
requires categories based on chemical identities.  For other aspects of the
analysis we have regrouped the segments.

     The Inorganic Chemical group consists primarily of acids, bases, and
salts of the more common elements.  These compounds are not expected to be the
subject of substantial innovation.  The majority of these chemicals are
produced in large quantities.

     High Polymers are chain-like macromolecules comprised of an indefinite
number  of smaller molecules (monomers)  which are chemically bonded.  Most
synthetic structural materials, such as plastics and rubbers, are made of such
polymers and usually produced in large quantities.  Although the monomeric
units change little over time,  new polymers are constantly being developed to
adapt to new uses.

-------
                                     - 12 -
                                  EXHIBIT  2-3

                               SEGMENTATION SCHEME
Segment
Inorganic Chemicals
 Component Products
Alkalies and Chlorine
Industrial Gases
Inorganic Pigments
Inorganic Acids
Potassium -and Sodium Compounds
Fertilizers
Industrial Inorganic
  Chemicals, NEC	
   Roughly
Cor re spond i ng
  SIC Codes
  2812
  2813
  2816
  28193,4
  28197
  2873,4

  28195,6,9
High Polymers
Thermoplastic Resins
Thermosetting Resins
Synthetic Rubber
Organic Fibers, Non-Cellulosic
  28213
  28214
  2822
  2824
Amphipathic
Compounds
Soaps and Detergents
Surfactants
Fatty Acids	
  2841
  2843
  28992
Elementary Organic Chemicals
Primary Petrochemicals
Gum and Wood Chemicals
 Cyclic Crudes	
  2911
  2861
  28655
Organic Chemicals, NEC
Cyclic Intermediates                  28651
Organic Dyes and Pigments             28652,3
Miscellaneous Cyclic
  Chemical Products                   28691
Miscellaneous Acyclic
  Chemical Products                   28692
Rubber-Processing Chemicals           286933
Plasticizers                          286935
Synthetic Organic Chemicals, NEC      28695
Organic Explosives                    2892
Catalysts
Catalytic Preparations
  28198
Other Chemical Products
Cellulosic Fibers
Polishes and Sanitary Goods
Paints and Allied Products
Adhesives and Sealants
Explosives
Printing Ink
Carbon Black
Salts, Essential Oils, and
   Chemical Preparations, NEC
  2823
  2842
  2851
  2891
  2892
  2893
  2895

  28991,5

-------
                                     - 13 -
     Fatty Acids are weak organic acids isolated from animal or plant fats or
oils.  Reaction with inorganic bases yields soaps.  Other detergents, such as
the linear alkyl sulfonates, are made synthetically from simpler organic
chemicals.  These substances are amphipathic in that they are soluble in both
water and oil.  Innovation is expected to be low to moderate.

     The Elementary Organic Chemicals are the simple alkanes, olefins, and
aromatics which can be obtained directly from petroleum refining.  Paraffins
and asphalt are also included in the category.  Little or no innovation is
expected in this group.  These substances are the basic building blocks for
all organic chemicals and are produced in very large quantities.

     Organic Chemicals are used as intermediates for synthesizing chemicals
in most of the other categories and also have a variety of uses as end
products.  Innovation is rapid in this area, and the amounts produced vary.

     Catalysts contain both organic and metallic constituents.  They are
commonly used in small quantities as synthetic reagents and fuels.  Because
the field is still in its infancy, innovation is expected to be quite high.

     Other Chemical Products contains a variety of products which do not fit
into the other categories and may have low rates of innovation.

     In Exhibit 2-4 we show some of our expected characteristics for each
segment.  Within each segment economic analysis is performed along
product-market lines.  For example, within the Other Chemical Products
category we developed nine profiles roughly along SIC segments.

                                   EXHIBIT  2-4

               GENERAL  CHARACTERISTICS EXPECTED FOR EACH SEGMENT

                                                               Non-Occupational
Segment                       Innovation      Production^/         Exposure	

Inorganic Chemicals              L                M-H                V
High Polymers                    H                H                  H
Amphipathic Compounds            L-M              M-H                M-H
Elementary Organic
  Chemicals                      L                H                  V
Organic Chemicals, NEC           H                V                  V
Catalysts                        H                L                  L
Other Chemical Products          V                V                  V

Index Key:
  L = Low
  M = Moderate
  H = High
  V = Variable
    a/Production quantities and exposure levels for new chemicals in each
      group will generally be lower than average.

-------
                                     - 14 -


     In concept, our segmentation scheme did not differ greatly from one
proposed earlier by ADL.  We grouped ADL's 41 segments according to chemical
characteristics.  Because business enterprises of all sizes produced similar
types of chemicals, we split the chemical segment, as best we could, in order
to enhance economic discussion.  In the case of primary petrochemicals, we
combined ADL's four segments into one.

Previous Estimates of the Number of New Chemicals That Will Be Introduced
Without TSCA Appear Reasonable and Are Unlikely to be Improved Through Further
Analysis

     Both Foster D. Snell, Incorporated in 1975,  and ADL in 1978,  have
estimated the number of new chemicals introduced annually.   ADL projected the
number to be 3,000, of which 2,000 are almost exclusively for R&D purposes.
For the remaining 1,000, only 300 are produced in quantities greater than
1,000 pounds.  Foster D. Snell estimated 3,300 new substances, a figure close
to the ADL estimate.

     Three ways to obtain information about the number of new chemicals
introduced annually are:  (1)  to use buyers' guides, (2)  to extrapolate from
patent data, and (3) to survey the industry.  All three approaches have been
used in previous analyses, and the ADL estimates considered the results of all
three.  For this reason, we have used the ADL estimates.

     We did attempt, however,  to refine the ADL estimate of the likely sales
growth over time for new chemicals and the likely cash flow associated with
them,  using the cash flow streams extrapolated from the sales data, we
created net present values of new chemical investments under differing
assumptions about how the uncertainty consequences affected expected real
returns.  Then, we compared the direct cost consequences of filling out the
notice to the total investment.  From this comparison, we made a judgement
about the likelihood of the chemical not being introduced.

     This financial analysis approach initially appeared the most useful way
to explore the problem of new chemical introduction.  However, the lack of
data about past new chemical introduction made these results too speculative
to be included in this report.  Appendix C in "Appendix:   Volume 1" provides
the details of the analysis performed.

-------
                                     -  15 -
                                    CHAPTER 3

                           DATA SOURCES AND COLLECTION
      ICF considered  the  following sources of data in preparing this report:
 industry interviews,  formal contact with trade associations, chemical industry
 literature, government statistics, trade publications, chemical marketing
 literature, and previous studies bearing on chemical innovation.  Although ICF
 used  all these sources to varying degrees, most information came from:
 publicly-available data  in trade journals, in literature, and in previous
 studies about the chemical industry and innovation; the previous work
 performed by ADL, including their industry interviews; a meeting of several
 independent chemical industry experts  (whose views are not necessarily
 reflected in this analysis, but who were a source of data); those industry
 representatives willing  to provide information; and the public record on the
 January 10tn and October 16^ proposals.
 PUBLICLY-AVAILABLE DATA

      ICF developed initial profiles of the industry segments and identified
 data  gaps to be closed by industry experts and industry representatives by
 first consulting the trade literature.  The documents reviewed included all of
 the issues as far back as 1965 of Chemical Marketing Reporter, Chemical Week,
 Chemical and Engineering News, and Chemical Purchasing.  ICF also examined
 Stanford Research Institute's confidential reports on chemical segments and
 the various volumes of the Kline Guide to the Chemical Industry, the Census of
 Manufactures, U.S. Industrial Outlook, Current Industrial Reports, and annual
 reports and 10-Ks of major competitors in selected industry segments.
ARTHUR D. LITTLE STUDIES

     The second important source was ADL's previous work.  ICF used three
components of ADL's work as sources of data for the analysis.  The first was
the data from their economic analysis of new chemical introductions.  Although
ICF discovered that the sample was skewed to the low end of the new chemical
sales volumes, ICF used the data because it was representative of the mix of
new chemical sales of companies in the $10-200 million annual sales
range—companies with the potential to be affected more severely, according to
the rules, than larger companies.

     Secondly, the estimates of the cost of providing specific notice
information to EPA was the building block for cost estimates ICF made for the
minimum reporting guidance, processor reporting requirements, and importer
reporting requirements.

-------
                                     - 16 -
     The third aspect of ADL's work that ICF used initially was their
methodology for estimating the number of new chemicals that would not have
been introduced if section 5 requirements had been in effect at the time of
their introduction.  Their analytical approach with which ICF concurred was to
perform discounted cash flow analysis (DCF) of potential new chemicals based
on new chemicals introduced in the past.  As developed in Appendix C, some
aspects of their analysis were not refined to the extent ICF believes they
should have been.  ICF performed this refinement to the extent possible and in
the Appendix shows how dramatically different the results could be.
INDUSTRY REPRESENTATIVES WILLING TO PROVIDE INFORMATION

     ICF met with representatives from the Chemical Manufacturers Association
(CMA), the Synthetic Organic Chemicals Manufacturers Association (SOCMA), the
Chemical Specialties Manufacturers Association (CSMA), the Soaps and Detergent
Association, and the National Paint and Coatings Association to discuss both
our approach and our data needs.  Additionally we contacted the Society of
Plastics Industry, the Man-Made Fibers Association, the Dry Color
Manufacturers Association, the Adhesives Council, the Adhesives Manufacturers
Association, and the American Importers Association.  A critical piece of data
that ICF sought was information on research and development costs,  sales
volumes, and profit margins of new chemicals introduced in the past on an
industry segment-specific basis.  All industry representatives expressed doubt
that ICF could obtain this type of data on new chemicals introduced in the
past because of the highly confidential nature of such information.  Several
industry representatives did share with us specific information about their
companies and their segments of the industry.  ICF incorporated this data into
the analysis.  ICF and EPA are still attempting to meet with industry
representatives in a continuing effort to obtain additional data regarding the
various industry segments.


MEETING OF INDEPENDENT CHEMICAL INDUSTRY AND INNOVATION EXPERTS

     After reviewing the ADL interview notes and after learning from the trade
associations that they could not finish the needed raw data, we realized that
obtaining representative data would require either an extensive and time-
consuming interviewing program or a very detailed formal questionnaire.
Alternatively, a group of industry and innovation experts could be convened to
discuss the chemical industry and its segments.  ICF chose to convene a panel
of such experts.  The experts supplied data and analysis during a two-day
discussion of the impact of the notice requirements on the chemical industry.

     The backgrounds of the experts were varied.  Two participants were
academics who have studied innovation.  One was the author of numerous
articles on innovation in the chemical industry.  The three other
participants, independent consultants to the chemical industry, represented a
total of 100 years of industry experience.  Each had his own perspective on
industry problems.  One came from a marketing background.  Another  had held a
variety of R&D positions and owned patents on sane chemicals.  The  third had
been a general manager and senior executive for a major chemical company
before becoming an independent consultant and academic.  Their resumes are
available upon request.

-------
                                     - 17 -
     The experts provided valuable information relative to the focus of the
analysis and the innovative sector of the industry.  First, they cautioned us
against taking too myopic a view of the notice requirement.  They recognized
the need to identify those consequences associated solely with the notice
requirement.  But they also cited examples of past analyses that had been too
narrowly focused and that had, as a result, completely overlooked the broader
implications of regulatory programs.  In addition, the experts provided more
specific data on the innovative sectors of the industry, that is, those
segments of the industry that had routinely introduced new chemical entities
or that gave signs of doing so.  For example, they furnished us with estimates
of the expected gross margins on new products and the number of competitors
broken down by company size — in annual sales volume.  They considered this
last factor crucial to understanding of the impact of the notice requirement.
PUBLIC COMMENT ON EARLIER NOTICE PROPOSALS

     The final important source of data was the public record on the January
10tn proposal and October 16*^ proposal concerning the notice require-
ment.  This source was particularly helpful in performing the issues analysis
presented in Chapter 5.

-------
                                     - 19 -
                                    CHAPTER 4

                      IMPACT OF SECTION 5 NOTICE REGULATION
     This chapter presents an overview of the costs the section 5 notice
 regulation  imposes on the chemical industry.  The chapter is organized to
 first provide  the theoretical basis for assessing impacts, followed by a
 discussion  of  the kinds of costs the regulation imposes.


 THEORETICAL EXPECTATIONS ABOUT  IMPACTS^/

     Three  classes of impacts that can be expected from the section 5 notice
 program are described below.

     (1)  Impacts on individual chemical firms.  Income and profits may flow
          from one set of firms to another within the industry.  It is
          generally thought that small chemical firms which rely on innovation
          are particularly threatened.

     This is an area in which the impacts of section 5 notices might be great.
 It is important to mention that the impacts discussed here, up to and
 including the bankruptcy of an individual firm and the assumption of its
 production  by other firms, do not cause any loss to society in the long run
 unless the  industry becomes non-competitive.  Economic theory suggests that
 there may be short-term disruptions and there may be severe costs to sane
 people.  But short-term disruptions should eventually disappear, and the
 losses to one group of people will be gains for another group.  [We do not
mean to imply that society should be unconcerned about changes in the
 distribution of benefits.  We mean simply that we should recognize the
 difference  between such changes and welfare losses.]

     (2)  Impacts on the chemical industry.  These impacts include income and
          profit flows from the domestic chemical industry to foreign chemical
          industries and other domestic industries as well as changes in
          market structure in the chemical industry.  To the extent that
          changes in market structure affect the productive efficiency and the
          flow of profits to the chemical industry, the economic analysis
          should offer evidence about the impact.
    2/This analysis is concerned with the impacts on submitters of section 5
notices.  A discussion of impacts on EPA and society can be found in Appendix D.

-------
                                     - 20 -
     The reduction in chemical innovation and the increased costs of
innovations that do survive could result in the replacement of chemicals by
other materials in some applications, e.g. use of wood instead of plastic in a
construction application.  Thus, the hypothesis to test is that this
replacement is not significant.

      (3)  Impacts on the U.S. as a whole.  This includes any welfare losses
          due to inefficient production, changes in the balance of trade in
          chemicals, and changes in the balance of trade of products that
          consume chemicals.

     We expect welfare losses!/ caused by inefficient production to be the
most  important economic impact of section 5 notices.  Examples of welfare
losses would be the delay in or lack of introduction of new chemicals into the
marketplace.

      The net effect of the regulations on the U.S. balance of trade in
chemicals will depend on the form that new chemical regulation takes (or has
taken in the case of japan) in other major chemical-producing countries.  if
every other country had regulations identical to those of the U.S., we would
expect no TSCA-prompted changes in the balance of trade.  [Laws governing
toxic substances already exist or will soon exist in other major industrial
countries; Japan has a law, and the members of the European community are
expected to pass laws by 1981 which conform to a model law.]  ideally, we
would assess the relative barriers to the introduction of new chemicals posed
by  each of these laws, but the undefined aspects of each law or prospective
law make it difficult to assess their consequences.

      Economic theory indicates that the potential for disturbances in the
balance of trade may have been greatly exaggerated because of the typically
close relationship between the manufacturer and the customer.  The specialty
chemical manufacturers have described a situation in which frequent
interaction between chemical firms and their customers occurs when a new
chemical is first introduced.i/  it is hard to accept the idea that foreign
firms introduce their chemicals here in a similar fashion, because the
    .2/A welfare loss can be contrasted to the income and profit flows from
one set of firms to another discussed earlier.  The latter represents a
transfer from one set of economic actors to the other, but no loss to the
economy as a whole.  However, welfare loss occurs when there is a loss of
productive efficiency in one sector of the economy without a counterbalancing
increase in productive efficiency in another sector.

    i/Source:  American Chemical Society. Chemistry in the Economy
(Washington: ACS, 1973), p. 202; Comments of Reilly Tar and Chemical Company,
March 21, 1979, pp. 9-17.

-------
                                     - 21 -
interactions with customers would be much more cumbersome and expensive,
making their marginal cost curve higher.  Consequently, increases in chemical
imports and decreases in exports will occur only if some chemicals are never
introduced here because of TSCA, but are introduced elsewhere.

THEORETICAL BASIS OF SECTION 5 NOTICE IMPACTS

     The impacts of section 5 suggested above flow from a key fact—the direct
costs of regulation fall solely on new chemicals.  Consequently, changes in
the timing of introduction of new chemicals, production costs, and customer
prices will be the focus of this analysis.^/

     One way to identify these impacts is to focus on a single chemical which
would have been introduced at time t, price p, and cost c before passing TSCA.
After the passage of TSCA, a single chemical will be introduced at time t+f,
price p+p1, and cost c+c', where t1, p1, and c1 are presumed to be positive.
The magnitude of t1, p1, and c1 will then determine the impacts of section 5
notices on individual firms in the chemical industry and on the U.S. economy.
IMPACT OF t'
     First, consider the impact when the only change is that it takes longer
to bring new products into the marketplace.  Assuming that the only impact of
section 5 notices is to delay the introduction of a new chemical by time t',
an analysis can be performed to estimate the the result of this delay.  Most
interesting is the impact when a chemical is never introduced (t1 =
infinity).  Because the chemical foregone is presumably more cost effective
than the item it might replace, its lack of introduction causes society to
suffer a welfare loss from producing inefficiently (which may or may not be
offset by the welfare gain to society of not introducing a potential health
hazard).

     There are also distributional consequences when t' equals infinity: income
flows from both the producers of the foregone new chemicals and the users of
the chemicals to the producers of the items they would have replaced.  To the
extent these items are other chemicals, the chemical industry as a whole
suffers no loss at any given level of national income.  There is simply a
redistribution among chemical companies.  To the extent that the foregone
chemicals would have replaced nonchemicals, e.g. glass, at any given level of
national income the chemical industry loses income to other industries.
    .5/This presumes the main focus is on product innovation.  Process
innovations often result in the production of new intermediates.  Because
notices must be prepared for isolated intermediates, process innovations may
also be inhibited.  But because the success of the final product (if not on
the new process) is already assured, we would expect the impacts of stifled
process innovations to be much less than those of stifled product innovation.

-------
                                     - 22 -
     The impact on industry profits of chemicals never being introduced is
less clear.  A new chemical could be patented and replace a competitively
produced commodity chemical.  In this case,  all of the welfare gains would be
absorbed by the producing firm's profits.!/   Therefore, the foregoing of a
new chemical would mean a reduction in profits for the firm and for the
industry.  On the other hand, a new chemical could break up a monopoly,
thereby reducing profits for the firm and the industry.  In this case, the
foregoing of a new chemical would prevent a  decrease in profits for the
industry.Z/

     The situation would change somewhat if  innovation were blocked in the
United States but not in another country. Section 5 notices might also block
import of the new chemical.  If so, foreign  manufacturers who use the new
chemical would produce more inexpensively than American firms.  If the final
product were not subject to section 5 notice requirements and could be
exported to the U.S., this could undercut American manufacturers.  If the
chemical were imported and replaced other imported chemicals, the net balance
of trade would not change.  If the new chemical replaced a domestically
produced chemical, the balance of trade would change, and income would flow
from the domestic chemical industry to the foreign chemical industry.
IMPACT OF C' AND p'

     When t1 is less than infinity, the new chemical would be introduced, but
its cost would be raised by c1  and its price by p1.   Because of the increased
price, the new chemical would not be used as widely  as it otherwise would have
been, and society would suffer  a welfare loss like the one described above.
Because of the decreased use of the new chemical,  income would flow from new
chemical producers to the producers of the goods or  services the new product
would have displaced.

     In addition, there could be differential section 5 notice requirement
barriers for importers and domestic producers.   If so, these differential
barriers would impose different costs on foreign and domestic producers,
shifting income between foreign and domestic producers and changing the
balance of trade.
    I/In addition, welfare gains would be smaller than if the product were
competitively priced.

    2/Even if the section 5 notice requirements were to inhibit the
production of a large  number of new chemicals in the short run, we would
expect this to happen  infrequently in the long run.

-------
                                     -  23 -
EMPIRICAL MANIFESTATION OF  INCREASED COSTS AND DELAYS IN INTRODUCTIONS

     Costs will  increase  (C+c1) and the timing of introduction of some
chemicals will be  lengthened  (t+t1) when final section 5 regulations are
promulgated because,  for  each new chemical entity, there will be associated
regulatory costs.  C1 and f can be translated into six "real world" factors:
direct out-of-pocket  costs, delays in the introduction of new chemicals,
uncertainty regarding direct out-of-pocket costs, length of the delay,
possible disclosure of trade secrets, and possible restrictive actions.

     The direct  out-of pocket cost can be considered to be  (1) the
pre-submission costs  of planning the submission, collecting and organizing
information, completing and submitting the notice; and (2)  the post-submission
costs of responding to requests for clarifications or additional information,
appealing EPA decisions according to established Agency procedures, and
pursuing legal challenges to Agency decisions.  These costs are incurred and
can be quantified  on  an individual company basis.  For some companies, efforts
to minimize these  costs will include the added one time cost of restructuring
their information  data base.5/

     Delays in the introduction of new chemicals are the result of the
statutory requirement for a 90-day review period and a possible extension of
90 days.  In addition to  these two requirements, the possibility for
indefinite delay exists if a section 5(e) action is taken.

     Uncertainty regarding the out-of-pocket costs focuses primarily on the
possibility that the  submitter will be required to submit additional
information.  Authorities under which additional information could be required
include supplemental  reporting, section 5(e) actions, or incomplete notice
determinations.

     Uncertainty regarding the length of delay derives from the gap in the
statute which requires a  complete notice for the 90-day clock to run.  Without
well-defined procedures for determining if submissions meet the criteria for a
notice, companies  will always be concerned about lengthy delays.  Uncertainty
will peak if delay occurs precisely at the moment when a competitor introduces
an alternative product that captures the first firm's anticipated market.

     Uncertainty about possible trade secret disclosure.  The act requires
that within five days of  receiving a submission EPA place a notice of receipt
in the Federal Register.  Although efforts to disguise the information in the
    1L/ADL estimated the pre-submission direct out-of-pocket costs for the
January 10, 1979 proposal and the October 16, 1979 reproposal.  Using the ADL
methodology. ICF estimated these costs for the minimium reporting guidance
option, processor reporting, and importer reporting.  For details about the
cost estimates the reader should refer to the ADL reports and Appendix A to
this report.

-------
                                     - 24 -
notice are substantial (as we discuss later),  the possibility exists that the
notice will not sufficiently protect trade secrets.   There is also some
concern that a Freedom of Information Act request will be improperly handled
causing disclosure.

     The final cost is uncertainty about regulatory  actions.  The entire
process is affected by the concern that a chemical will be regulated by EPA.
Consistent action by the agency will reduce this uncertainty.  This cost often
overwhelms the others because it dictates strategies concerning pre-submission
decisions about what data to provide.  In the  worst  case, the regulatory
action is a 5(f)  decision causing infinite delay.

PROBLEMS WITH COST ESTIMATION

     At the present time, not all of the costs to submitters readily translate
into dollars.  Even for the costs which would  generally be specified in
dollars (i.e., direct out-of-pocket costs), information does not exist which
adequately captures the pre-submission and post-submission expected
out-of-pocket costs for a variety of industry  segments.   As ICF explains
later, it is clear that uncertainty costs far  outweigh direct out-of-pocket
costs within all industry segments.  For some  firms  in each segment, direct
costs may outweigh uncertainty; but this is the exception.  In particular,
industry experts steadfastly maintain that direct out-of-pocket costs are not
the most important type of cost.  Because these out-of-pocket costs readily
translate into dollars, previous attempts to estimate the costs of complying
with section 5 requirements have focused on pre-submission out-of-pocket
costs.  It should not be inferred, however, that because pre-submission,
out-of-pocket costs are the only costs that currently lend themselves to
monetization, that they are necessarily the most important costs.

     Chapter 5 quantifies in dollars the pre-sutmission,  out-of-pocket costs
when existing data permit.  Otherwise, costs are presented ordinally.


SUMMARY

     Section 5 imposes six kinds of costs on submitters which can be grouped
into pre-submission and post-submission categories as follows:

     (1)   Before  the notice is submitted, the  submitter incurs the
          following pre-submission costs;

          •    out-of-pocket costs of planning the submission,
               collecting and organizing information, and completing
               and submitting the notice; and

          •    time delay costs,  as defined by the delays in the
               commencement of manufacturing attributable to
               completing the notice.

-------
                                - 25 -
(2)   After  the notice is submitted,  the submitter  could  incur
     the following post-submission costs;

     •   additional out-of-pocket costs in responding to
          requests for clarifications or additional
          information, appealing EPA decisions  within
         established Agency procedures, or pursuing legal
         challenges to Agency decisions;

     •    additional delays in the commencement of manufacture
          for the same reasons cited above;

     •    the disclosure of trade secrets;  and

     •    restrictions placed on how the chemical  can enter
          commerce.

-------
                                     - 27 -
                                    CHAPTER 5

                    ANALYSIS OF COMPREHENSIVE PROGRAM OPTIONS


     The magnitude of the costs imposed by section 5 varies with the
regulatory option chosen.  This chapter analyzes the costs of each regulatory
option across the main issues:

     •    confidentiality;

     •    customer contact;

     •    importer definition;

     •    importer contact of foreign manufacturer/supplier;

     •    exporter reporting;

     •    supplemental reporting;

     •    insufficient submissions;

     •    processor reporting; and

     •    possession or control.

Three comprehensive program options  (CPOs) were constructed by EPA by
selecting one regulatory option for each of the nine issues.

     The three CPOs are defined in Exhibit 5-1.  The nine issues were analyzed
according to the cost dimensions discussed in Chapter 4:

     •    pre-submission
          —   out-of-pocket costs, and
               delay costs; and
     •    post-submission
          —   out-of-pocket costs,
          —   delay costs,
          —   potential trade secret disclosure costs, and
          —   restrictive action costs.

Costs for each CPO were constructed from the costs of the option appropriate
to that CPO for each of the nine issues.  These costs were compared on a
monetized basis when sufficient data were available.  Otherwise, they were
compared on an ordinal basis.  ICF developed the ordinal comparisons because
the only costs that lent themselves to monetization were the pre-submission
out-of-pocket costs.  To compare the magnitude of the non-monetized costs, ICF
employed an ordinal ranking system.

-------
                                     - 28 -


                                  EXHIBIT  5-1

                   SUMMARY OF COMPREHENSIVE PROGRAM OPTIONS*
                         CPO 1
                                              CPO 2
                                                                  CPO 3
Reporting
Requirement
Confidentiality
Assert
Substantiate
Timing
Generic
Chemical ID
Timing
Chemical ID
Inventory
Customer Contact
Importer
Definition
Upstream
Contact
Minimum Guidance

Item-by-Item
Series of questions
5(d) (2) at
submission
None
Manufacturing
Generic
None
Principal
Importer
Health and envi-
ronmental effects
EPA Reproposal

Category
Reproposal questions
All at submission
4 categories
Manufacturing
Specific
Unknown Uses
Combined
responsibility
(proposal)
Form based on
reproposal
EPA Proposal

Category
Series of questions
All at submission
4 categories
Submission
Specific
Proposal
Combined
responsibility
(proposal)
Proposed form
Exporter
Supplemental
  Reporting

Insufficient
  Submissions


Processor
  Reporting


Possession or
  Control
data and risk
assessments only

Section 8 notice
(coverage equiv-
alent to proposal;
authority differs)

Case-by-Case


Right to
invalidate; no
procedures

Processor reporting
under section 8
Subsidiaries
only, same
venture
Reproposal
Reproposal


January 1979
proposal


Processor reporting
for exempt sub-
stances (section 5)
Entire corporate
hierarchy, same
venture
Proposal
Proposal


Time limit;
appeal procedures


Manufacturer
and importer
reporting
Entire corporate
hierarchy, any
venture
 These options are explained in subsequent pages.
issue is found in Part II of this document.
                                 In-depth analysis of each

-------
                                      -  29 -

          /
     The system assigned values according to the following rules:

     1.   Regulatory options  imposing no costs to submitters were
          assigned a zero rating.

     2.   Regulatory options  which according to data and logic
          imposed the lowest  costs were assigned a 1.

     3.   Regulatory options  imposing higher costs were given higher
          rankings.

     4.   Regulatory options  which according to data and logic
          imposed equal costs were assigned equal ratings.

 In  the  remainder of this chapter we first discuss the costs of reporting
 requirements  for the three comprehensive program options.  Next we summarize
 the regulatory options and costs for  each of the nine regulatory issues.  This
 is  followed by a discussion of the inter-relationships between the nine issues
 and their costs.  The chapter closes  with a discussion of the costs imposed by
 the comprehensive program options on  domestic manufacturers producing for
 domestic consumption, importers, foreign manufacturers, exporters, and
 processors.

     The issues analyzed here are multi-faceted and complex.  In the interests
 of  clarity of presentation, the summaries of the CPOs and the nine issues are
 highly  simplified.  Complete  discussion of each CPO and the nine issues can be
 found in Part II of this report.  A presumption of familiarity with the
 January 10, 1979 proposal and October 16, 1979 reproposal has been made.

 REPORTING REQUIREMENTS

     In addition to the nine  issues,  the costs of the reporting requirements
 as  explained below, were considered.  CPO 1 imposes the least reporting
 requirements and CPO 3 the most.  This does not mean that CPO 1 is least
 costly.

 Pre-submission Costs

     Exhibit 5-2 contains the cost summary for reporting requirements.  The
 dollar estimates for pre-submission out-of-pocket costs were derived using a
methodology similar to that used by ADL.jV  presubmission delay costs
 increase from CPO 1 through CPO 3 because the more information needed to
 comply with the reporting requirements, the longer it will take to gather the
 necessary information and submit the section 5 notice.
    9/Arthur D. Little, Inc.  (ADL), Impact of TSCA Proposed Premanufacturing
Notification Requirements  (Cambridge, Mass:  December 1978).

-------
                                               EXHIBIT 5-2

                      COMPARISON OF COSTS ATTRIBUTABLE TO NOTIFICATION  REQUIREMENTS
      Cost
Pre-submission costs

    Out-of-pocket costs
    Delay

Post-submission costs

    Potential out-of-pocket
    Potential delay
    Potential trade secret
      disclosure
    Potential restrictions

Coverage
     CPO 1
Minimum Guidance
 $1,000-7,500l/
       1
       3
       3

       1
       1
       CPO 2
10/16/79 Reproposal


   $1,200-8,900k/
         2
         2
         2

         2
         2
     CPO 3
1/10/79 Proposal


  $3,700-42,000£/
         3
         1
         1

         3
         3
                                                                         CO
                                                                         o
£/Appendix A to this report in "Appendix:  Volume 1," p. A-25.

^/Arthur D. Little, Inc., Estimated Costs for Preparation and Submission of Reproposed Premanufacture
Notice Form, (Cambridge, Mass.:  September 1979), p. 38.

£/Arthur D. Little, Inc., "Impact of TSCA Proposed Premanufacturing Notification Requirements,"
(Cambridge, Mass.:  December 1978), p. V-17.
Note:  for ordinal estimates, 0 = no cost, 1 = lowest cost, 2 = next lowest costs, etc.

-------
                                     - 31 -
Post-Submission Costs

     Post-submission out-of-pocket costs decrease from CPO 1 through CPO 3
because the probability that EPA will require additional existing information
to evaluate the submission decreases with the breadth of the initial reporting
requirements.  That is, the more existing information required at the time of
submission, the less additional existing information will be needed after
submission.  Post-submission delay also decreases from CPO 1 through CPO 3
because the less defined and less comprehensive the notice requirements, the
more likely it is that the notice period will be extended for up to an
additional 90 days in order to reach additional existing data.

     The risk of potential trade secret disclosure increases from CPO 1
through CPO 3 because the broader the initial reporting requirements, the more
confidential business information is likely to be submitted.  The risk of
trade secret disclosure is the product of the quantity of trade secrets
submitted multiplied by the probability of disclosure (as defined by the
confidentiality policy).  Therefore, risk of trade secret disclosure increases
with the breadth of reporting requirements.

     As the initial reporting burden increases, some potential new chemicals
will not be able to meet the more burdensome notice requirements.  When
submitters choose not to subject these chemicals to section 5 review because
the review process is considered too burdensome, the resultant decision not to
enter the market becomes a de facto restriction on that chemical entering
commerce.

     In addition, the changes in explicit EPA regulatory decisions that are
brought about by increasing the initial reporting burden is an issue that
needs consideration.  The rationale for increased information must be improved
decision making.  In the case of EPA regulatory decisions (i.e., the decision
to initiate a 5(e) or 5(f) action), the additional information must sometimes
lead to changed decisions.  In some cases, chemicals which would not have been
subjected to some restrictions without the additional information now would
be.  In other cases, chemicals which would have been subjected to some
restrictions without the additional information now would not be.  The
combination of voluntary submitter decisions not to enter the market and
possible changed EPA regulatory decisions contribute to fewer new chemicals
coming to market as initial reporting requirements increase from CPO 1 through
CPO 3.

Coverage

     The reporting requirements for the three CPO's do not differ on coverage
of manufacturers.  Any potential coverage differences in customer contact,
importers, exporters, or processors are addressed later in this chapter.

ISSUE 1.  CONFIDENTIALITY

     provisions relating to the confidentiality of submitted information
attempt to reconcile the concerns of business that trade secrets will be

-------
                                     - 32 -
disclosed with public concerns that adequate information on the hazardous
nature of the chemicals should be made available.  The key elements of this
issue are the method for asserting the claim of confidentiality (assert), the
method for substantiating the claim (substantiate), the timing of the claim
(timing), the manner in which the information is masked (generic), the timing
of identification of the chemical (chem ID timing), and the type of name by
which the chemical will be placed on EPA's inventory (chem ID inventory).  The
CPOs are defined across these elements as shown below:
   Element

Assert

Substantiate

Timing


Generic

Chem ID Timing
   CPO 1

Item by Item

Minimum Guidance

5(d)(2) at Sub-
  mission

None

Manufacture
   CPO 2
                                                                CPO 3
Category              Category

Reproposed Questions  Minimum Guidance
Chem ID Inventory  Generic
All at Submission


4 Categories

Manufacture

Specific
All at Sub-
  mission

4 Categories

Submission

Specific
Costs to the submitter are summarized in Exhibit 5-3 and are discussed below.
Method and Timing of Assertion and Substantiation

     Pre-submission Costs.  In assessing the pre-submission out-of-pocket
costs associated with confidentiality, the benchmark is an ADL estimate
(assuming a claim is made) derived while costing the reproposal.i2/  The ADL
cost ranges were specified for "a generalized method of asserting and
substantiating claims of confidentiality."  The ADL estimate for
pre-submission out-of-pocket costs is assumed to apply to both CPO 2 and CPO
3, because they differ only on the substantiation questions.  The
pre-submisson out-of-pocket costs are less for CPO 1, because not all
information asserted as confidential needs to be substantiated at the time of
submission.  These same relationships hold for pre-submission delays:  delays
for CPO 1 are less than delays for CPO 2 and CPO 3, because not all items
asserted as confidential need to be substantiated at the time of submission.

     Post-Submission Costs.  CPO 1 has larger post-submission out-of-pocket
costs than CPO 2 or CPO 3 because of the potential need to substantiate some
items after submission.  There are no post-submission delays associated with
    ill/Arthur D. Little, Inc., Estimated Costs for Preparation and
Submission of Reproposed Premanufacture Notice Form (Cambridge, Mass.:
September 1979), p.  52.

-------
                                                         33 -
                                                    EXHIBIT 5-3

                         COMPARISON  OF COSTS ATTRIBUTABLE TO CONFIDENTIALITY PROVISIONS!/
        Cost
      CPO 1

Assert:  Item by Item
Substantiate:  Interim
   questions
Timing: 5(d) (2)  at
   submission
Generic:  none
Chera ID Timing:   Mfg
Chem ID Inventory:
   Generic
        CPO 2

Assert:  Category
Substantiate:  Reproposal
    questions
Timing:  All at submission

Generic:  4 Categories
Chem ID Timing:  Mfg
Chem ID Inventory:
   Specific
      CPO 3

Assert:  Category
Substantiate:  Interim
    questions
Timing:  All at submissior

Generic:  4 Categories
Chem ID Timing: Submissior
Chem ID Inventory:
   Specific
Method  and  Timing  of  Assertion
    and  Substantiation

Pre-submission  costs

    Out-of-Pocket Costs
    Delay

Post-submission costs

    Potential  out-of-pocket
    Potential  delay
    Potential  trade secret  disclosure
    Potential  restrictions

Coverage
   less than
   $900-6400
        1
     $900-6400Jy
          2
   $900-6400
        2
Generic  Requirements

Pre-submission costs

   Out-of-pocket costs
   Delay

Post-submission costs

   Potential out-of-pocket
   Potential delay
   Potential trade  secret disclosure
   Potential restrictions
When to Disclose Chem ID

Potential trade secret disclosure


How to Place on Inventory

Potential trade secret disclosure
^/Assume that a confidentiality claim is made.

^/"Estimated. Costs for Preparation and Submission of Reproposed Premanufacture Notice Form" ADL, Inc.,
  September 1979, p. 52.
Note:  For ordinal estimates, 0 = no cost, 1 = lowest cost, 2 = next lowest.

-------
                                     - 34 -
asserting and substantiating confidential items because policies related to
the withholding and disclosure of confidential business information do not
have any effect on whether the 90-day notice period will be extended.
Potential trade secret disclosure is higher under CPOs 2 and 3 than under CPO
1 because of the timing of substantiation,   under CPOs 2 and 3,  it is possible
that there will be some confidentiality claims that are not sufficiently sub-
stantiated which can conceivably enter the  public domain.  Because there is no
initial substantiation and no possibility of not sufficiently substantiating
at this point under CPO 1, the opportunity  for disclosure decreases.

     We assume that a submitter's motive in voluntarily initiating
restrictions will not stem solely from considerations of confidentiality.  Of
course, it is possible that submitters might not submit new chemicals for
section 5 review, but the underlying reason for refraining to do so is more
likely to be the appreciable costs of trade secret disclosure.

Coverage

     There are no coverage differences attributable to any of the confiden-
tiality provisions.

Generic Requirements

     Because there are no generic requirements under CPO 1, there are no costs
under the generic heading.

     Pre-subtnission Costs.  Pre-submission  out-of-pocket costs under CPOs 2
and 3 involve developing generic masking for items asserted as confidential.
Development of such generic masks would also carry some pre-submission delay.

     Post-Submission Costs.  CPOs 2 and 3 carry potential post-submission
out-of-pocket costs because additional details related to generic masking may
have to be negotiated between the submitter and EPA.  CPOs 2 and 3 also carry
some risk of potential trade secret disclosure, because generically masked
information might be used by others to uncover trade secrets.  There are no
post-submission delays or potential restrictions resulting from generic
requirements.

Other Confidentiality Provisions

     Other provisions for confidentiality concern how and when chemical
identity, as part of a health and safety study, would be disclosed.  Cost
differences between the comprehensive program options relate only to potential
trade secret disclosure.  The risk of trade secret disclosure depends on the
CPO.  CPO 3  (disclosure at submission)  carries a greater risk of disclosure
than do CPOs 1 or 2.  CPO 1 (add to inventory generically) carries a lesser
risk of disclosure than CPO 2 or 3 (add to  inventory by specific chemical
identity).

-------
                                     - 35 -


ISSUE 2.  CUSTOMER CONTACT

     In some circumstances, new chemical customers could provide data in such
areas as worker exposure, waste disposal, and environmental release.   The CPOs
vary in terms of the number of customers contacted and the amount of
information sought:

     •    CPO 1:  no customer contact provisions;

     •    CPO 2:  the reproposal asks for reports on approximate
          production quantities and number of customers for which
          use is unknown  [note:  a recent variation on this general
          approach is to report the expected use distribution broken
          down by quantities and number of customers]; and

     •    CPO 3:  the initial proposal (mandatory contact for a very
          broad definition of customers and mandatory submission of
          customer lists).

Costs are addressed below.  Costs to submitters and costs to customers are both
considered and are summarized in Exhibit 5-4.

Cost to Submitters

     Pre-submission Costs.  Pre-submission out-of-pocket costs increase from
CPO 1 through CPO 3.  Mandatory contact for all of the categories of  customers
specified in CPO 3 is clearly more costly than providing use information.
There are pre-submission delay costs for CPO 3 because mandatory contact will
consume additional time before the notice can be submitted.

     Post-Submission Costs.  There are post-submission costs for CPO  2,
because this is the only option where EPA would go back to the submitter and
ask for information about some customers as a direct result of reporting
requirements about customer uses.  EPA could also ask for customer information
under CPO 1, but the reporting requirements under CPO 1 are not as likely to
generate use information which would trigger these requests.  Furthermore,
because there are no customer contact provisions under CPO 1, any such
requests to submitters would have to be considered costs of supplemental
reporting.  Under CPO 3, all out-of-pocket costs have been incurred at the
time of submittal.   Thus, only CPO 2 has potential post-submission
out-of-pocket costs directly attributable to customer contact.

     Post-submission delays are more likely under CPO 3 than under CPO 2.  The
requirement under CPO 3 that submitters must contact customers and request that
they submit customer forms may result in EPA's invoking section 5(c)  more
often in order to receive all of the relevant information on which to base
a decision.  EPA may also, have to use supplemental reporting to obtain
receiving information from customers under this option.  Under CPO 2,
information needs from customers are more focused, resulting in infrequent
need for 5(c).

-------
          Cost
Pre-submission costs
                                             EXHIBIT 5-4

                         COMPARISON OF COSTS ATTRIBUTABLE TO CUSTOMER CONTACT
CPO 1
(none)
    CPO 2
(reproposal)
   CPO 3
(proposal)
   Out-of-pocket costs
   Delay
  0
  0
      1
      0
     2
     1
Post-submission costs

   Potential out-of-pocket                 0
   Potential delay                         0
   Potential trade secret disclosure       0
   Potential restrictions                  0
                             1
                             1
                             0
                             1
                                0
                                2
                                1
                                2
                                                                          OJ
                                                                          CTl
Coverage
Cost  to customers
Note:  0 = no cost,  1 =  lowest  cost,  etc.

-------
                                     - 37 -


     Costs associated with potential trade secret disclosure are only incurred
under CPO 3.  Under CPO 3, mandatory contact, particularly to potential rather
than actual customers, coupled with the mandatory submission of a customer
list, increases the problem of trade secret disclosure.

     Potential restrictions on how chemicals, can enter commerce are likely to
increase from CPO 1 to CPO 3.  Voluntary submitter actions to refrain from
submitting chemicals to section  5 review could increase as customer contact
provisions grow.  The impact of  the additional information on explicit EPA
regulatory decisions is not clear and could result in changes in either
direction.  The combination of voluntary submitter actions and changed EPA
explicit regulatory decisions increases the potential cost of restrictive
action as customer contact provisions become more burdensome.

     Coverage.  The options move toward broader coverage.  CPO 1 has no
customer contact policy, so the definition of customer is not a factor.  CPO 2
is limited to those customers with firm commitments and CPO 3 reaches beyond
this to potential customers with less than firm commitments.

     Cost to Customers.  Under CPO 1 there are no costs to customers.  Under
CPO 2, there are costs only when EPA makes inquiries to selected customers
based on information provided by the submitter on customer use.  Under CPO 3,
there are optional information requests to customers, and responses to them
could be quite lengthy.  Thus, the costs to customers increase from CPO 1
through CPO 3.
ISSUES 3 and 4.  IMPORTER PROVISIONS

     The issue of importer definition deals with designating the entity which
must comply with section 5 requirements for imported chemicals.  The CPOs vary
in terms of the specific identity of the importer and the amount of
information requested of the foreign manufacturer:


   Issue        	CPO 1	  	CPO 2	  	CPO 3	

Definition of   principal importer  Combined Responsi-  Combined Responsi-
   Bnporter                          bility (Proposal)   bility (Proposal)

   Upstream         Mandatory           Mandatory             Mandatory
   Contact        Health and        Revised Information      1/10/79 Form
                  Safety Data Only  Request

The cost of these alternatives is addressed below and summarized in Exhibit
5-5.

Definition of Importer

     Pre-submission Costs.  Pre-submission out-of-pocket costs to importers
are provided in Exhibit 5-5.  These costs were derived by applying the ADL
cost-estimating methodology to the reporting requirements for importers under

-------
                                                          38 -
                                                     EXHIBIT 5-5

                               COMPARISON OF COSTS ATTRIBUTABLE TO IMPORTER PROVISIONS
            Cost                           CPO 1
                                                                    CPO 2*
                                                                                                   CPO
                                 def :  principal  importer   def:  combined responsibility  def:  combined responsibility
                                   upstream:  mandatory         upstream: mandatory           upstream: mandatory
                                    health  and safety      revised information request           1/10/79 form
                                         data
Definition of Importer

Pre-submission costs  (Importer)
   Out-of-pocket                      $800-7500^7
   Delay                                   1

Post-submission costs

   Potential out-of-pocket                 3
   Potential delay                         3
   Potential trade secret disclosure       1
   Potential restrictions                  1

Coverage                                   1
                                                                  $900-8900£/
                                                                       2
$2300-26,700^7
      3
Upstream Contact

Pre-Submission Costs

   Out-of-pocket  (Foreign Mfr.)       $300-1500£/
    Delay                                  1

Post-submission costs

   Potential out-of-pocket                 3
   Potential delay                         3
   Potential trade secret disclosure       1
   Potential restrictions                  1

Coverage                                   1
                                                                  $1800-7200l/
                                                                       2
$3000-11,600S/
      3
^./Assumes parties will make a good faith effort to comply with  the definition.
           s minimum guidance form less $200 for cost of completing  a minimum  guidance  section II-A.   Maximum is
  maximum cost of complete minimum guidance form.

£/Minimum is minimum cost of repropbsed form less $300  for cost of completing  a  reproposed section II-A.   Maximum
  is maximum cost of reproposed form.

^'Minimum is minimum cost of completing sections I,  II-A, II-D of proposed  form.  Maximum is  maximum  cost of
  completing section I, II-A, II-C,  II-D, III-A, III-B, III-C of proposed form.

^/Minimum is minimum of completing section III-B of  the reproposed form.  Maximum is  maximum  cost of  completing
  section III-B of the reproposal.

I/See Part II, Chapter 4,  Exhibit 4-3 for explanation.

2/Minimum is minimum cost of completing Section I-A, I-B, I-C, I-D,  II-A, II-B,  II-C  of  January 1979  Proposed
  Foreign Manufacturer/Suppliers Form as approximated by comparable  sections of  proposed domestic form.   Maximum
  is maximum cost of same sections (See Part II, Chapter 4, Exhibit  4-3  for more details.
Note:  For ordinal estimates, 1 = lowest cost, 2 and 3   next lowest costs,  respectively.

-------
                                     - 39 -


each of the comprehensive program options.  Pre-submission delays are assumed
to be in proportion to the breadth of reporting requirements and, therefore,
increase from CPO 1 through CPO 3.

     Post-Submission Costs.  Potential post-submission out-of-pocket costs
decline from CPO 1 through CPO 3 because potential requests for additional
information are inversely related to the scope of the initial reporting
requirements.  The more information that is provided with the initial notice,
the less additional information will be needed to make regulatory decisions.
Potential post-submission delay also decreases from CPO 1 through CPO 3,
because the probability of EPA initiating 5(c) action decreases with the
increased quantity of information provided by the initial submission.
Potential trade secret disclosure increases from CPO 1 through CPO 3, because,
the more information that is provided with the initial submission, the more
likely it is that trade secrets are included and, therefore, subject to
disclosure independent of the confidentiality policy.

     The potential for submitters to decide not to introduce a new chemical
into commerce is likely to increase from CPO 1 to CPO 3.  As reporting
requirements for importers become more stringent, submitters are more likely
to voluntarily  not submit some chemicals to section 5 review.  These
voluntary decisions  are responsible for the increasing cost of possible
restrictive actions.

     Coverage.  CPO 1 results in notice submission from the single most
knowlegeable importer. The combined responsibility policy under CPOs 2 and 3
applies to a broader set of firms than just the single defined importer.

Upstream Contact

     Presubmission Costs.  The pre-submission costs to the foreign
manufacturer/supplier are estimated assuming that the foreign manufacturer
will respond to requests for information.  Pre-submission out-of-pocket costs
are estimated by applying the ADL methodology to the appropriate reporting
requirements for foreign manufacturers/suppliers.  Pre-submission delay is
assumed to be proportional to the scope of reporting requirements.

     Post-Submission Costs.  The logic for the ordinal rankings of
post-submission costs is identical to the logic presented for post-submission
costs for the definition of importer.  Potential post-submission out-of-pocket
costs and delays decrease as the breadth of initial reporting requirements
increase,  potential trade secret disclosure and restrictive actions increase
as more information is required in the initial notice.

     Coverage.  There are no coverage differences; all three comprehensive
program options reach the identical set of foreign manufacturers/suppliers.


ISSUE 5.   EXPORTER PROVISIONS

     The issue of exporter responsibility deals with the manufacture and
transport of chemicals for export.  The CPOs vary from no reporting

-------
                                     - 40 -
responsibility to the same responsibility as domestic manufacturers who
distribute domestically.  The CPO exporter options, applicable only to
domestic manufacturers/suppliers producing solely for export,  are as follows:

     •    CPO 1:  submit section 8 notice equivalent to October 16»
          1979 reproposal;
     •    CPO 2:  submit section 5 notice equivalent to October 16,
          1979 reproposal; and
     •    CPO 3:  submit section 5 notice equivalent to January 10,
          1979 notice.

Exhibit 5-6 contains the cost summaries,  described in more detail below.

Pre-submission Costs

     Pre-submission out-of-pocket costs are shown in Exhibit 5-6.  These
estimates were derived by applying the ADL cost-estimating methodology to the
appropriate reporting requirements, as specified in the exhibit.
Presubmission delay is assumed to be proportional to the breadth of notice
requirements.  Therefore, pre-submission delay under CPO 3 is  greater than
under the other two options.

Post-Submission Costs

     iJo post-submission out-of-pocket costs related to section 5 are
attributable to CPO 1, because no action would be taken under  section 5
rule.ii/  post-submission out-of-pocket costs are higher under CPO 2 than
under CPO 3 because EPA is more likely to request additional information  under
CPO 2.  Also, the potential for post-submission delay is greater under CPO 2
because the less extensive the initial reporting requirements, the more likely
are 5(c) actions.

     The potential for trade secret disclosure and the potential for
restrictive action increase as initial reporting requirements  grow more
extensive.  More trade secrets are assumed to be submitted as  reporting
requirements increase.  Also, submitters will voluntarily restrict the flow of
new chemicals into commerce as the burden of section 5 grows.

Coverage

     CPO 1 applies to a smaller set of exporters because there is a small
business exemption under section 8:  manufacturers and processors with total
annual sales under $1 million are generally exempted from section 8 rules.
There are no other coverage differences between the definitions.  All apply to
domestic production solely for export.
    ii/We recognize that section 8 rules have costs associated with them.
However, if section 8 authority is used, there is no section 5-related cost.
Since we are analyzing section 5 impacts (see Chapter 2),  we cannot consider
section 8-related costs.

-------
       Cost
                      EXHIBIT 5-6

COMPARISON OF COSTS ATTRIBUTABLE TO EXPORTER PROVISIONS

                CPO 1

         (Section 8 Notice)
Pre-Submission Costs

   Out-of-pocket
   Delay

Post-submission costs

   Potential out-of-pocket
   Potential delay
   Potential trade secret disclosure
   Potential restrictions

Coverage
            $1200-810(£/
                0
                0
                0
      CPO 2
  (Must Submit
10/16/79 Notice)
$1200-810Qb/
      1
      2
      2
      1
      1
      CPO 3
  (Must Submit
  1/10/79 Notice)
$2500-40,300£/
         2
         1
         1
         2
         2
a/Assumes section 8 reporting requirements are  identical to the October 16 requirements.

^/Minimum is minimum cost of October 16 form less minimum cost of completing parts II-B and Il-C.
Maximum is maximum cost of October  16 form less maximum cost of completing part II-C.

£/Minimum is minimum cost of January 10 form less minimum cost of completing parts II-C and II-D.
Maximum is maximum cost of January  10 form less maximum cost of completing part II-D.

^/Although there may be costs here, they would  not be attributable to section 5.  As mentioned in
Chapter 2, we are analyzing the costs of section 5.  Therefore, a rule based on section 8 authority
has no section 5-related cost.
Note:  For ordinal estimates, O = no cost, 1 = lowest cost, 2 = next lowest cost.

-------
                                     - 42 -
ISSUE 6.  SUPPLEMENTAL REPORTING

     Supplemental reporting would require information in addition to the data
included in the original submission.  The CPOs vary according to the type of
additional information and the circumstances under which it could be required:

     •    CPO 1:  section 8 rules on a case-by-case basis;

     •    CPO 2:  reproposal of October 16, 1979 (section 8 letter
          writing authority with reasonably well defined criteria
          and appeal mechanism); and

     •    CPO 3:  proposal of January 10, 1979 (section 8 letter
          writing authority with less well-defined criteria).

Costs are summarized in Exhibit 5-7 and are discussed below.

Pre-submission Costs

     There are no pre-submission costs.

Post-Submission Costs

     Submitters are required to provide the most post-submission information
under CPO 3 and the least under CPO 1.

     Potential post-submission delays can occur through 5(c)  or 5(e)  actions.
The probability of a 5{c) action to extend the notice period is assumed to be
independent of the procedures for supplemental reporting.  The probability of
a 5(c) action is likely to be a function of the quality and quantity of
information initially submitted and the complexity of the evaluation decision
based on available chemical-specific information.  The probability of a 5(e)
action is greater under CPO 1 than under the other two options because EPA
might use 5(e) actions under CPO 1 because there is no other way to obtain
additional information.  Therefore, the potential for post-submission delay is
greater under CPO 1 than under the other two options.

     The relative costs for potential trade secret disclosure and potential
restrictions are based on the principle that the more information which can be
submitted under supplemental reporting, the higher likelihood it is that:

          •    more trade secrets will be submitted and, therefore,
               subject to possible disclosure; and

          •    additional information requirements create a higher
               hurdle for submitters and may result in a voluntary
               reduction in the number of chemicals submitted to
               section 5 review.  Also, the additional information
               may result in changes to specific regulatory
               decisions.

-------
                                              EXHIBIT 5-7

                      COMPARISON OF COSTS ATTRIBUTABLE TO SUPPLEMENTAL REPORTING
       Cost

Pre-Submission Costs

   Out-of-pocket
   Delay

Post-submission costs

   Potential out-of-pocket
   Potential delay
   Potential trade secret disclosure
   Potential restrictions

Coverage
CPO 1
(section 8
case-by -case)
0
0
1
2
1
1
CPO 2

(October 16)
0
0
2
1
2
2
CPO 3

(January 10)
0
0
3
1
3
3
Note:  For ordinal estimates, 0 = no cost, 1 = lowest cost, 2 and 3 = next lowest costs,  respectively.

-------
                                     - 44 -
Coverage

     There are no coverage differences among the three supplemental reporting
options.
ISSUE 7.  INSUFFICIENT SUBMISSIONS

     The issue of insufficient submissions deals with the failure of
submissions to meet the statutory requirements and errors which impede notice
review.  The CPOs vary in their characterization of the deficiencies,
prescribed time delays, and appeal processes:

          •    CPO 1:  case-by-case determination, no general
               procedures;

          •    CPO 2:  the proposed policy of January 10, 1979; and

          •    CPO 3:  correction of errors, specification of
               inadequacies in submissions, and appeal procedures.

     Exhibit 5-8 contains a summary of costs for insufficient submissions
policy.  Costs for CPO 1 were estimated assuming that submitters and EPA would
behave identically under CPOs 1 and 3.  It is recognized, however, that under
CPO 1, EPA would have discretion to deviate from the policy defined by CPO 3,
which would create additional uncertainty for submitters.

Presubmission Costs

     Assuming the pre-submission period is defined by when the submitter
submits an initial piece of paper which it believes is a section 5 notice,
rather than with the submission of a notice which meets the statutory
criteria, there are no pre-submission costs associated with any of the
alternatives.

Post-Submission Costs

     Out-of-pocket costs are the same under all three options because roughly
the same submissions would be found incomplete or invalid under all three
options, and the same corrections would be made.  Delay associated with CPO 2
is greater than CPO 3; under CPO 3, the finding must be made within the first
30 days, but there is no such restriction under CPO 2.  Also, the notice clock
is suspended under CPO 2 for minor deficiencies provided they are corrected
within a specified time.  The clock is not stopped for minor deficiencies
under CPO 3.  For purposes of this analysis, delays attributable to CPO 1 are
assumed to be identical to delays under CPO 3.  Delays under CPO 1, however,
could be greater or lesser than under CPO 3, because of uncertainty about the
way in which CPO 1 would be implemented.

     The three options are assumed to require the same additional informa-
tion.  Thus, they do not differ in their potential for trade secret disclosure
or their potential to result in restrictive actions.

-------
                                               EXHIBIT 5-8

                       COMPARISON OF COSTS ATTRIBUTABLE TO INSUFFICIENT SUBMISSIONS
         Cost

Pre-Submission Costs^/

   Out-of-pocket
   Delay

Post-submission costs
            CPO
(Case-by-case determinations
    no general procedures)
             0
             0
   Potential out-of-pocket               1
   Potential delay                       1
   Potential trade secret disclosure     1
   Potential restrictions                1
                                                                 CPO 2
(Proposed 1/10/79)
      0
      0
                                        1
                                        2
                                        1
                                        1
    CPO 3
  (Time Limit;
Appeal Procedures)
        0
        0
                                1
                                1
                                1
                                1
Coverage
.5/In estimating costs, the behavior of submitters and EPA are assumed to be identical under
  CPOs 1 and 2.  It is recognized, however,  that under CPO 1, EPA would have discretion to deviate
  from the policy defined by CPO 2, which would create additional uncertainty for submitters.

^./Defined by when submitter submits notice,  not when notice period begins.
Note:  For ordinal estimates, 0 = no cost, 1 = lowest cost, 2 = next lowest cost.

-------
                                     - 46 -


Coverage

     There are no coverage differences attributable to the options.

ISSUE 8.  PROCESSOR REPORTING

     The issue of processor reporting deals with the reporting of chemicals
previously exempt from reporting under TSCA.  The CPO's vary from a section 8
requirement to the January 10 proposal.

          •    CPO 1:  processors submit section 8 notices;

          •    CPO 2:  processors submit section 5 notice equivalent
               to October 16, 1979 reproposal; and

          •    CPO 3:  manufacturers and importers submit section 5
               notice equivalent to January 10, 1979 (no separate
               processor).

Exhibit 5-9 contains the cost summaries which are explained below.

Pre-submission Costs

     Pre-submission out-of-pocket costs are shown in Exhibit 5-9.  The
estimates were derived by applying the ADL cost-estimating methodology to the
appropriate reporting requirements.  Pre-submission delays increase as more
information is initially requested for review.  Because CPO 3 requests more
information than CPO 2, delays under CPO 3 are greater than those under the
CPO 2.

Post-Submission Costs

     Because CPO 1 requires reporting under section 8, no post-submission
costs arise.  Post-submission out-of-pocket costs and delay costs are greater
under CPO 2 than under CPO 3 because those potential costs decrease with the
stringency of initial reporting requirements.

     The potential for trade secret disclosure and restrictive actions
increases as initial reporting requirements grow more stringent.  More trade
secrets are assumed to be submitted as reporting requirements increase, and
increased reporting requirements are expected to lead to more voluntary
restrictive actions by submitters  (i.e., a decision not to market).  Also, the
increased reporting requirements presumably will lead to some changed
regulatory decisions for some specific chemicals, although these changes could
be in either direction.

Coverage

     CPO's 1 and 2 cover all processors whether they are the manufacturer (or
importer)  or separate entities.  CPO 3 covers only processors who are
manufacturers or importers; coverage  does not extend to separate processors.

-------
                                                EXHIBIT 5-9

                         COMPARISON OF  COSTS ATTRIBUTABLE TO  PROCESSOR REPORTING
             Cost
   CPO 1
(Section 8
Reporting)
         CPO 2
(August 1980 Proposal)
   October  16 form
       CPO 3
(January 10 Proposal)
  January 10 form
Pre-Submission Costs

   Out-of-pocket
   Delay
724-4,450
    1
    $1,200-8,900
           1
  $3,700-42,000
         2
Post-submission costs

   Potential out-of-pocket               0
   Potential delay                       0
   Potential trade secret disclosure     1
   Potential restrictions                0
                                2
                                2
                                1
                                1
                                      1
                                      1
                                      2
                                      2
Coverage
Note:  0 = none, 1 = lowest cost, 2 = next lowest cost.

-------
                                     - 48 -
ISSUE 9;  POSSESSION OR CONTROL

     The issue of possession or control arises from the requirement that all
test data in the possession or control of the submitter be included with the
section 5 notice.  The CPOs vary by the extent of the possession or control
definition, horizontally, vertically, and by involvement in the current
venture.

     •    CPO 1:  possession or control of test data limited to the
          submitter and the companies controlled by the submitter
          that are involved in the current venture;

     •    CPO 2:  possession or control of test data limited to the
          submitter and the parents, subsidiaries, and other
          companies controlled by the parent company which are
          involved in the current venture, (i.e., any corporate
          entity in the corporate hierarchy involved in the current
          venture); and

     •    CPO 3:  possession or control limited to submitter and
          parents, subsidiaries, and other companies controlled by
          the parent company which are not necessarily involved in
          the current venture (i.e., all corporate entities in the
          corporate hierarchy).

Costs are summarized in Exhibit 5-10 and are described below.

Pre-submission Costs

     The further the submitter must search within its own corporate structure,
the greater the pre-submission costs in both time and money.  Thus, the costs
of pre-submission possession or control increase from CPO 1 through CPO 3.

Post-Submission Costs

     Potential post-submission out-of-pocket costs decrease with an increasing
scope of the definition of possession or control, because the more information
required at time of submission, the less post-submission information will be
required.

     Post-submission delay decreases as the scope of the definition of
possession or control increases, because delaying actions are less likely when
there is additional information provided initially.

     The potential for trade secret disclosure increases with the scope of the
definition, because more information must be submitted.

     The relationship between potential restrictive actions and possession or
control is similar.   The further submitters are required to search, the more
likely it is that some submitters will choose not to enter the market because
the perceived costs of notice requirements are too great.

-------
                                              EXHIBIT  5-10

                       COMPARISON OF COSTS ATTRIBUTABLE TO POSSESSION OR CONTROL


                                         CPO 1                     CPO 2                    CPO 3
                               (entities controlled by       (whole hierarchy,        (whole hierarchy,
          Cost                  submitter, same venture)        same venture)             any venture)

Pre-Submission Costs

   Out-of-pocket                        123
   Delay                                123

Post-submission costs

   Potential out-of-pocket              3                          2                         1
   Potential delay                      3                          2                         1
   Potential trade secret disclosure    123
   Potential restrictions               123

Coverage                                1                          11
Note:  For ordinal estimates, 0 = no cost, 1 = lowest cost, 2 and 3 = next lowest costs,  respectively.

-------
                                     - 50 -
Coverage

     There are no coverage differences attributable to the possession or
control options.

INTER-RELATIONSHIPS AMONG ISSUES

     The differences between comprehensive program options were addressed
above by issue.  There are, however, potential impacts which cannot be
attributed to a single issue or to an alternative within that  issue.  For
example, mandatory customer contact by itself may not pose much of a threat of
trade secret disclosure.  Yet, mandatory customer contact in combination with
a disclosure policy in which trade secrets are not adequately  protected could
pose a significant threat to submitters and their customers.   In essence,  this
relationship is an example of a possible synergistic impact.   Other types  of
relationships between issues may not be synergistic, but may still involve
dependencies.  For example, the cost of supplemental reporting depends on  the
amount of the information requested and the probable extent of investigation
by submitters to obtain that information.

     The purpose of this chapter is to point out the nature of these
inter-ralationships, so they can be taken into account when comparing
comprehensive program options.  For purposes of analysis, the  issues which
make up a comprehensive program option are grouped as follows:

            SCOPE OF
     REPORTING REQUIREMENTS          PROCEDURES        	COVERAGE	

    • Notice Requirements        • Confidentiality     • Importer Provisions
    • Possession or Control      • Insufficient        • Exporter Provisions
    • Supplemental Reporting         Submissions       • Customer Contact!?/
    • Customer Contact                                 • Processor Reporting

     The four issues included under scope of reporting requirements provide
EPA with access to all the information theoretically available that is
relevant to a section 5 decision for any particular chemical.  Confidentiality
and insufficient submissions are procedural issues because they do not define
the reporting burden placed on submitters.  Rather, they define procedures
which help ensure that submitters comply with the reporting requirements.  The
specification of insufficient submission provisions are independent of the
other issues and, therefore, are not discussed in this section.  Coverage
issues apply only to specific subgroups of affected parties; their
differential impacts over comprehensive program options are defined primarily
by the choice of reporting requirements and procedural policies.
     -12/customer contact is a hybrid issue in the sense  that  the burden  it
imposes on submitters is related to depth and breadth of reporting
requirements, but the burden it imposes on customers is  primarily a coverage
issue.

-------
                                      -  51 -


 SCOPE OF  REPORTING  REQUIREMENTS

     In theory,  the combination of  notice requirements, possession or control,
 supplemental  reporting,  and customer  contact provide EPA with the authority to
 obtain all  the available information  that is relevant to a section 5
 decision.   It is possible to devise a rational vertical ranking in terms of
 the concern of submitters about these issues, regardless of the comprehensive
 program option being addressed.   In order of decreasing concern to submitters,
 this ranking  is  as  follows:

     •    notice requirements,

     •    possession or  control,

     •    supplemental reporting, and

     •    customer  contact.

 The logic for this  ranking is discussed below.  First, the low rank of
 customer  contact is explained, and  then the other issues are addressed in
 pairs.

     Customer contact, as defined in  CPO 1 and CPO 2, will have little impact
 on submitters and is placed below supplemental reporting as an issue of
 concern,  assuming only these two CPOs are being considered.  The potential
 impact of customer  contact on submitters is significantly greater under CPO 3,
 given the mandatory contact requirements for a very broad definition of
 customers.  Therefore, under CPO  3, customer contact could be assumed to be of
 relatively  greater  importance to submitters.

     The  rationale  for the relative importance of notice requirements,
 possession  or control, and supplemental reporting is developed through
 comparison  by pairs as follows:

     •    Notice Requirements/Supplemental Reporting;

     •    Possession or  Control/Supplemental Reporting; and

     •    Notice Requirements/Possession or Control.

 This section  will be followed by a  summary of the relative importance of each
 to submitters.

 Notice Requirements/Supplemental Reporting

     The more narrowly notice requirements are defined, the more likely supple-
mental reporting will be used.  In  the case where notice requirements are
 relatively  comprehensive (i.e., the January 10, 1979 proposal), submitters may
 provide EPA with more information than is needed to make decisions, thereby
 incurring unnecessary costs; but supplemental reporting costs would then be
minimal.  In  general, it would be less costly for submitters to meet less
detailed notification requirements  and risk additional supplemental reporting

-------
                                     - 52 -
costs.   In this way, submitters provide EPA with exactly the information
needed to make section 5 decisions, and no more.  Limiting the amount of
information reduces the risk of trade secret disclosure and also insures that
attention is paid to the correct information in the section 5 decision-making
process.  Thus, in terms of relative costs, notice requirements dominate
supplemental reporting requirements as an issue of concern to submitters.

Possession or Control/Supplemental Reporting

     A similar argument can be developed regarding the relationship of
possession or control to supplemental reporting.  As with notiqe requirements,
submitters would prefer to minimize the initial burden so that any unnecessary
costs can be avoided.  Therefore, submitters would choose a narrow definition
of  "possession or control" and run the risk of additional supplemental report-
ing costs, no matter how the supplemental reporting rules are ultimately
defined.

Notice Requirements/Possession or Control

     The relationship between notice requirements and possession or control is
a synergistic one.  Thus, slight changes in both directons can have signifi-
cant impacts on the joint effects.  The relative importance of these two
issues depends on individual data items.  For example, given the choice, a
submitter might choose to eliminate particularly sensitive requirements from
the notice in exchange for a broader definition of possession or control for
the remaining elements.  Because notice requirements are the primary deter-
minant of the reporting burden faced by submitters, notice requirements domi-
nate possession or control.
CONFIDENTIALITY AND THE OTHER ISSUES

     Confidentiality concerns are paramount.  The relationship between
confidentiality and reporting requirements is addressed in the next section.
The following two sections will address the synergisms of confidentiality/
customer contact, and confidentialty/ importer provisions, respectively.

Confidentiality and Reporting Requirements

     To submitters, the essential confidentiality issue is the degree to which
trade secrets will be protected.  In essence, the primary confidentiality
concern can be cast in terms of the risk of disclosure.  Notice requirements
provide submitters with a risk of trade secret disclosure which could be
measured as the product of two factors:

          •    quantity (and sensitivity)  of trade secret data
               submitted, and

          •    probability of trade secret disclosure.

-------
                                     -  53 -
Presumably, the  first  factor  is directly proportional to the scope of
reporting  requirements, and the second factor represents the result of any
given confidentiality  policy.  Given  the multiplicative nature of the
relationship  between these two factors, simultaneous small increases in both
components could imply a  significant  increase in the burden to submitters.
The multiplicative  relationship between notice requirements, possession or
control, and  confidentiality  policy creates the possibility that a
comprehensive program  option  which is only slightly broader on all three of
these issues  may be significantly more burdensome to submitters.

     Consider how the  three comprehensive program options differ on reporting
requirements  and confidentiality policy:
                            CPO 1
                   CPO 2
Reporting Requirements
     Notice

   Possession or
     control
Confidentiality
     Generic
       Policy
     When to
       substantiate
minimum guidance   reproposal
downward, same
   venture
    none
5(d)(2) with
   notice
whole hierarchy,
same venture
chem. ID, manuf.
ID., use, phys.
and chemical
properties

all with notice
                  CPO 3
proposal

whole hierarchy,
no venture
limits
chem. ID. manuf.
ID., use, phys.
and chemical
properties

all with notice
The reporting requirements increase from CPO 1 to CPO 3 according to both
relevant dimensions.  The major provisions of the confidentiality policy carry
an increased burden in progressing from CPO 1 to CPO 2, but not from CPO 2 to
CPO 3 (the differences between CPO 2 and CPO 3 relate to the substantiation
questions and the issue of when and how to disclose chemical identity as part
of a health and safety study).  Although the generic requirements carry some
additional risk of trade secret disclosure, it is reasonable to assume that
this risk is independent of the scope of reporting requirements.  (The
responses to the four items for which generic information is required probably
do not change significantly as reporting requirements grow.)  However, the
risk associated with substantiating all information at the time of submission
increases with the scope of notice requirements.  Thus, even though the
differences between CPO 1 and CPO 2 do not appear great on an issue-by-issue
basis, the combined effects of the differences in individual issues may be
significant.

-------
                                     - 54 -
Confidentiality/Customer Contact
           /
     Although requirements for mandatory customer contact may not by themselves
pose risks of trade secret disclosure, mandatory customer contact (especially
with a broad definition of customers), in conjunction with a confidentiality
policy where trade secrets may not be protected adequately, could pose a
substantial threat to submitters and  their customers.  Submitters could be
concerned that their customers might  be able to piece together enough
information about the new chemical to manufacturing it themselves.il/  Such
subsequent manufacture by customers would not be subject to any section 5
requirements because the original manufacturer would have already submitted
the notice.  Similarly, submitters might be able to gather enough information
from their customers' submission to consider downstream integration.

     Only CPO 3 appears to present this type of joint risk from
confidentiality/customer contact.  However, this additional risk should be
considered as an incremental cost of  moving from CPO 2 to CPO 3.

Confidentiality/Importer Provisions

     The problem here is that the importer defined for the purposes of section
5 may not be able to protect the trade secrets of foreign suppliers or
manufacturers.  Substantiation requirements under all three comprehensive
program options require submitters to demonstrate that their competitive
position will be harmed by disclosing the information asserted as
confidential.  Depending upon how an  importer under section 5 is ultimately
defined, it is possible that the importer will have access to sensitive
information.  The disclosure of such  information may not harm the importer but
could harm the foreign supplier or manufacturer.  In this situation, the
importer may not be able to assert that this information is confidential.  In
some cases, it may even be to the importers' advantage to have foreign
manufacturers' confidential business  information released  (e.g., chemical
identity and use) in the hope that a  potentially cheaper source of the
chemical becomes available.il/  The impact of this joint effect for CPOs 1
and 2 will depend on the specification of Alternative 2 for definition of
importer (most knowledgeable person as defined by EPA).  For CPO 3, the
mechanics of implementing the combined responsibility approach to the
definition of an importer are not sufficiently clear to allow for an
assessment of the joint effects of confidentiality/importer provisions.
    ^./Comments of the Manufacturing Chemists Association on EPA  Proposed
Regulations for Premanufactures Notification Under  Section  5 of TSCA,  March
26, 1979, pp. 212-220.

    M/Comments of the American Importers Association, OTS-05002E, November
30, 1979, p. 13.

-------
                                     -  55 -


 COMPARISON OF  COMPREHENSIVE  PROGRAM OPTIONS

      In  this section,  the  comprehensive program options are compared.
 Although it  is not  possible  yet  to describe the costs of any comprehensive
 program  option completely  in dollar terms, it is useful to organize the
 comprehensive  program  options  by the groupings suggested in Chapter 2.  This
 facilitates  the comparison of  comprehensive program options both horizontally
 and  vertically and  provides  a  rough framework for understanding the relative
 differences  among them.


 Domestic Manufacturers Producing For A Domestic Market

      Exhibit 5-11 summarizes the issue-by-issue costs to domestic manufacturers
 producing for  a domestic market, ranked by the hierarchy discussed earlier.
 The  costs presented in Exhibit 5-11 are identical to those developed
 previously,  and they have  not  been adjusted for interdependences.  Costs
 which are in dollars (pre-submission out-of-pocket notice and confidentiality
 costs) must  be thought of  in conjunction with costs attributable to other
 issues,  as the footnotes to  Exhibit 5-11 indicate.  Although the direction in
 which these  issues  affect  costs  is known, the magnitude of the cost effects is
 not.  Presubmission out-of-pocket notice costs need to consider possession or
 control.  Presubmission out-of-pocket confidentiality costs need to consider
 notice requirements and possession or control.  The rationale behind these two
 additional considerations  is provided below.

      The  pre-submission out-of-pocket notice costs were derived from estimates
 prepared  by  ADL.l^/ These estimates were developed under ADL's assumed
 "view of  the likely response of  a prudent firm."  This probably would
 correspond to  the definition of  possession or control assumed in CPO 1 or CPO
 2.   Because  CPO 1 and  CPO  2  are  both limited to companies participating in the
 same venture with the  submitter,  there is probably not much difference between
 the  two,  provided that CPO 2 does not extend the reach of CPO 1 to foreign
 firms.   It is  assumed  that international considerations will significantly
 complicate possession  or control because some countries have laws limiting the
 international  flow  of  scientific and technical information.  This problem
might occur  in  a case  where  the  submitter is a domestic subsidiary partici-
 pating in the  same  venture with  a foreign subsidiary of the same parent.
    JJ>/The costs for CPO 3 were derived from Impact of TSCA Proposed
Premanufacturing Notification Requirements, December 1978; the costs for CPO 2
were estimated from Estimated Costs for Preparation and Submission of
Reproposal premanufacture Form, September 1979; the costs for CPO 1 were
derived by ICF using the ADL methodology.

-------
                                                        - 56

                                                      EXHIBIT 5-11

                       SUMMARY OF COSTS  FOR  DOMESTIC  MANUFACTURERS  PRODUCING  FOR A DOMESTIC MARKET
          Cost
                             Coverage
                                                Pre-Submission
                                                                                       Post-Submission


Reporting Requirement
CPO 1
CPO 2
CPO 3
Possession or Control
CPO 1
CPO 2
CPO 3
Supplemental Reporting
CPO 1
CPO 2
CPO 3
Customer Contact
CPO 1

CPO 2

CPO 3


Out-of-Pocket Delay

Same $1,000 to 7,500 1
Same $1,200 to 8,900 2
Same $3,700 to 42,000£/ 3

1 1
2 2
3 3

0 0
0 0
0 0

No Customer g g
Contact
No Mandatory i Q
Contact
Mandatory 2 1
Contact

Out-of-Pocket

3
2
1

3
2
1

1
2
3

0

1

0


Delay

3
2
1

3
2
1

2
1
1

0

1

2

Trade Secret
Disclosure

1
2
3

1
2
3

1
2
3

0

0

1


Restrictions

1
2
3

1
2
3

1
2
3

0

1

2

Confidentiality

  How and When to Assert
  and Substantiate
    CPO 1
    CPO 2
    CPO 3

  Generic
    CPO 1
    CPO 2
    CPO 3

  When to Disclose
  Chemical ID
    CPO 1
    CPO 2
    CPO 3

  How to Disclose
  Chemical ID
    CPO 1
    CPO 2
    CPO 3

-Insufficient Submissions
    CPO 1
    CPO 2
    CPO 3
less than
to 6,400£/
to 6,400
to 6,400£/
0
1
1
0
0
0
0
0
0
0
0
0
1
2
2
0
1
1
0
0
0
0
0
0
0
0
0
2
1
1
0
1
1
0
0
0
0
0
0
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
1
2
1
1
2
2
0
1
1
1
1
2
1
2
2
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
£/Range specified is likely an underestimate due to the reach of possession or control under CPO  3.

£/Range specified is likely an underestimate due to notice requirements and possession or control
  under CPO 1.

£/Range specified is likely an underestimate due to notice requirements and possession or control
  under CPO 3.

Note:  For ordinal estimates, 0   no cost, 1 = lowest cost, 2 = next lowest cost, etc.  Ordinal estimates  are
       meaningful only within issues.

-------
                                     -  57 -
 Under  CPO  1,  the  submitter would not be in possession or control of the
 foreign  firm's  data, whereas under CPO 2 the submitter would be.  Thus, CPO 2
 could  extend  the  reach of possession or control to foreign firms in limited
 situations.   However, the definition of possession or control under CPO 3 has
 a much broader  reach, because it is not limited to the current venture.
 Therefore,  in any case where the submitter is part of an overall corporate
 hierarchy  which includes foreign subsidiaries, problems of international
 information transfer will occur.  Thus, out-of-pocket pre-submission notice
 costs  for  CPO 3 should be adjusted to account for this difference.

     Presubmission out-of-pocket costs for confidentiality need to be
 similarly  adjusted.  The estimates which appear in Exhibit 5-11 are derived
 from ADL's out-of-pocket estimates for the reproposed confidentiality policy.
 Although the  cost ranges are specified for "a generalized method of asserting
 and substantiating claims of confidentiality," it seems reasonable to assume
 that these  ranges were based on a notice form and a definition of possession
 or control approximated by the reproposal.  Therefore, the pre-submission
 out-of-pocket confidentiality costs for CPO 3 would be significantly higher,
 and the pre-submission out-of-pocket confidentiality costs for CPO 1 would be
 slightly lower, as the footnotes indicate.  In making these assessments, one
 should remember that the relationship between notice requirements, possession
 or control, and confidentiality is multiplicative rather than additive.
Importers

     Exhibit 5-12 contains a summary of costs for importers.  Coverage
differences and pre-submission out-of-pocket cost differences are the major
differences between Exhibit 5-11 and 5-12.  Adjustments to pre-submission
out-of-pocket costs for both reporting requirements and confidentiality need
to be considered, as discussed previously and as the footnotes in Exhibit 5-12
indicate.  Confidentiality costs are complicated further by the potential
problem of an importer who is unable to substantiate the trade secrets of a
foreign supplier/manufacturer.  The inability to substantiate the claims could
either increase or decrease out-of-pocket confidentiality costs to importers,
depending upon whether or not EPA develops a mechanism for foreign
manufacturers to protect their trade secrets.


Foreign Manufacturers/Suppliers

     Exhibit 5-13 summarizes the costs to foreign manufacturers/suppliers under
each of the comprehensive program options.  Because the foreign manufacturer/
supplier submission is essentially a supplement to an importer section 5
submission, the issues of possession or control, supplemental reporting,
customer contact, and insufficient submissions are assumed not to affect the
foreign manufacturer/supplier submission.  Foreign manufacturers/suppliers are
assumed to make an effort in good faith to respond.

-------
                                                           58
                                                      EXHIBIT 5-12

                                             SUMMARY OF COSTS FOR IMPORTERS
         Cost
Reporting Requirement

  CPO 1
  CPO 2
  CPO 3

Possession or Control

  CPO 1
  CPO 2
  CPO 3

Supplemental Reporting

  CPO 1
  CPO 2
  CPO 3

Customer Contact

  CPO 1

  CPO 2

  CPO 3


Confidentiality

  How and When to Assert
  and Substantiate^/

    CPO  1
    CPO  2
    CPO  3

  Generic
    CPO  1
    CPO  2
    CPO  3
                             Coverage
Principal
Combined Resp.
Combined Resp.
                                                Pre-Submission
                                                                                       Post-Submission
                                           Out-of-Pocket
$  800 to  7,500     1
$  900 to  8,900     2
$2,300 to 26,700£/   3
                                                                  Trade Secret
                                            Out-of-Pocket  Delay   Disclosure   Restrictions
No Customer
  Contact
No Mandatory
  Contact
Mandatory
  Contact
                 less than
                  $900 to  6,400.k/   1
                  $900 to  6,400     2
                  $900 to  6,400£/   2
0
0
1
0
1
0
0
1
2
0
0
1
  When to Disclose
  Chemical ID
    CPO 1
    CPO 2
    CPO 3

  How to Disclose
  Chemical ID
    CPO 1
    CPO 2
    CPO 3

Insufficient Submission
    CPO 1
    CPO 2
    CPO 3
^/Range specified is likely an underestimate due to reach of possession or control under CPO 3.

.b/Range specified is likely an underestimate due to notice requirements and possession or control under
  CPO 1.  Also, the phenomenon of importers not necessarily being able to substantiate the trade secrets of
  foreign suppliers complicates this cost assessment.

S/Range specified is likely a significant underestimate due to notice requirements and possession or control  under
  CPO 3.  There are additional complications due to considerations as mentioned above.

I/The cost ranges specified here are based on confidentiality costs for the domestic manufacturer's form,  even  though
  the importer's form is less broad than the domestic manufacturer's form.

Note:  For ordinal estimates, 0   no cost, 1   lowest cost, 2   next lowest cost, etc.  Ordinal estimates  are meaning-
       ful only within issues.

-------
                                                            59 -

                                                       EXHIBIT  5-13

                                   SUMMARY  OF COSTS  FOR FOREIGN MANUFACTURERS/SUPPLIERS
       Cost
 Reporting  Requirement

   CPO 1
   CPO 2
   CPO 3

 Possession or  Control

   CPO 1
   CPO 2
   CPO 3

 Supplemental Reporting

   CPO 1
   CPO 2
   CPO 3

 Customer Contact

   CPO 1

   CPO 2

   CPO 3
 Confidentiality

   How and  When to Assert
   and Substantiate.^/
     CPO 1
     CPO 2
     CPO 3

   Generic
     CPO 1
     CPO 2
     CPO 3

   When to  Disclose
   Chemical ID
     CPO 1
     CPO 2
     CPO 3

   How to Disclose
   Chemical ID
     CPO 1
     CPO 2
     CPO 3

 Insufficient Submission
     CPO 1
     CPO 2
     CPO 3
                              Coverage
  Same
  Same
  Same
                                                 Pre-Submission
                                                                                       Post-Submission
No Customer
  Contact
No Mandatory
  Contact
Mandatory
  Contact
                                            Out-of-Pocket
$  300 to  1,500     1
$1,800 to  7,200     2
$3,000 to 11,600£/   3
                      0              0
                      0              0
                      0              0
                      0              0
                      0              0
                      0              0
      0              0

      0              0

      0              0
                                                                       Out-of-Pocket  Delay
                                                                                             Trade Secret
                                                                                              Disclosure   Restrictions
less than
$900 to 6,400
$900 to 6,400
$900 to 6,400
0
1
1
0
0
0
0
0
0
0
0
0

1
2
2
0
1
1
0
0
0
0
0
0
0
0
0
5/Range specified is likely an underestimate due to reach of possession or control under CPO  3.

.b/Range specified is likely an underestimate due to reduced scope of the  foreign manufacturers/suppliers  form.

Note:  For ordinal estimates, 0   no cost, 1 = lowest cost, 2   next lowest cost, etc.  Ordinal  estimates are
       meaningful only within issues.

-------
                                      -  60  -


      Foreign manufacturers/suppliers  have  special confidentiality concerns.
 It may be that importers will possess information that  is sensitive  to  foreign
 manufacturers/suppliers which they will not be able to  protect.  Thus,  unless
 EPA  institutes a mechanism  to ensure  that  foreign manufacturers have an
 opportunity to assert  and substantiate  information in the importer's
 submission,  foreign manufacturers/suppliers may bear a  special risk  of  trade
 secret disclosure.
 Exporters

      Exhibit  5-14  summarizes the costs of the three comprehensive program
 options  to domestic manufacturers who produce new chemicals solely for
 export.  Under CPO 1, these manufacturers must submit a section 8(a) notice
 before manufacturing solely for export.  Because these manufacturers are
 exempt from section 5 reporting, no costs of this option are technically
 attributable  to section 5.  In Exhibit 5-14, no costs are shown for CPO 1
 (other than pre-submission costs for reporting requirements, which are
 provided for  comparative purposes) because CPO 1 is not a section 5 rule.
 However, it is still possible for submitters to incur other costs, but such
 costs would be attributable to other parts of the statute.

      Differences between CPO 2 and CPO 3 parallel the explanations developed
 previously.   Presubmission out-of-pocket costs for reporting requirements need
 to be adjusted upward for the deeper reach of possession or control under CPO
 3.   Presubmission  out-of-pocket costs for confidentiality under CPO 3 need to
 be adjusted upward for the broader reporting requirements and deeper possession
 or control definition applicable to CPO 3.  However, this increased cost would
 be offset somewhat by the fact that reporting requirements for domestic
manufacturers that produce for domestic consumption (on which the
 confidentiality out-of-pocket costs were based)  are greater than reporting
 requirements  for exporters.
Processors
     Exhibit 5-15 summarizes the costs for processors.  CPO 1 requires
processors to file a section 8(a) notice equivalent to the reproposed form of
October 1979.  CPO 2 requires processors to file a section 5 notice as
proposed on August 15, 1980.  CPO 3 requires only processors who are also the
manufacturer or importer to file a section 5 notice equivalent to the proposed
form of January 1979 (i.e. the separate processor does not file a notice).
The logic for the cost estimates parallels the logic for domestic
manufacturers producing for a domestic market.

-------
                                                           - 61

                                                       EXHIBIT 5-14


                                              SUMMARY OF COSTS FOR EXPORTERS
          Cost
                            Coverage
                                                  Pre-Submission
                                                                                         Post-Submission


Reporting Requirement
CPO 1 Small Bus. Exemp.
CPO 2 No Exemptions
CPO 3 No Exemptions
Possession or Control
CPO 1
CPO 2
CPO 3
Supplemental Reporting
CPO 1
CPO 2
CPO 3
Customer Contact
CPO 1 No Customer
Contact
CPO 2 No Mandatory
Contact
CPO 3 Mandatory
Contact
Confidentiality
How and When to Assert
and Substantiates/
CPO 1
CPO 2
CPO 3
Generic
CPO 1
CPO 2
CPO 3
When to Disclose
Chemical ID
CPO 1
CPO 2
CPO 3
How to Disclose
Chemical ID
CPO 1
CPO 2
CPO 3
Insufficient Submission
CPO 1
CPO 2
CPO 3

Out-of-Pocket

$1,200 to 8,100
$1,200 to 8,100
$2,500 to 40,30Qb/

0
1
2

0
0
0

0

1

2




none
. $900 to 6,400
$900 to 6,400l/

0
1
1


0
0
0


0
0
0

0
0
0

Delay

Oa/
1
2

0
1
2

0
0
0

0

0

1




0
1
1

0
1
1


0
0
0


0
0
0

0
0
0

Out-of-Pocket

0
2
1

0
2
1

0
1
2

0

1

0




0
1
1

0
1
1


0
0
0


0
0
0

0
1
1

Delay

0
2
1

0
2
1

0
1
1

0

1

2




0
0
0

0
0
0


0
0
0


0
0
0

0
2
1
Trade Secret
Disclosure

0
1
2

0
1
2

0
1
2

0

0

1




0
1
1

0
1
1


0
1
2


0
1
1

0
1
1

Restrictions

0
1
2

0
1
2

0
2
1

0

1

2




0
0
0

0
0
0


0
0
0


0
0
0

0
1
1
^/Although this cost is a cost, it is not a Section 5 cost.  Therefore, a rule based on section 8 authority has no
sector 5-related cost.

.b/Range specified is likely an underestimate due to reach of possession or control under CPO 3.

£/Range specified is likely an underestimate due to the fact that the exporter's form is narrower than the domestic
  manufacturer's form.


-------
                                                             -62-



                                                       EXHIBIT  5-15

                                              SUMMARY OF COSTS FOR PROCESSORS
                                                  Pre-Submission
VUS l-

Repocting Requirement
CPO 1
CPO 2
CPO 3

Possession or Control
CPO 1
CPO 2
CPO 3
Supplemental Reporting
CPO 1
CPO 2
CPO 3
Customer Contact
CPO 1

CPO 2

CPO 3

	 a 	 	
Out-of-Pocket Delay

all $724 to 4,450 1
all $1,200 to 8,900 1
only manufacturer $3,700 to 42,00027 2
or importer

0 0
1 1
2 2

0 0
0 0
0 0

No Customer g 0
Contact
No Mandatory ]_ Q
Contact
Mandatory 2 1
Contact

Out-of-Pocket

0
2
1


0
2
1

0
1
2

0

1

0


Delay

0
2
1


0
2
1

0
1
1

0

1

2

Trade Secret
Disclosure

1
1
2


0
1
2

0
1
2

0

0

1


Restrictions

0
1
2


0
1
2

0
1
2

0

1

2

 Confidentiality

   How and  When to  Assert
   and Substantiate
     CPO 1
     CPO 2
     CPO 3

   Generic
     CPO 1
     CPO 2
     CPO 3

   When  to  Disclose
   Chemical  ID
     CPO 1
     CPO 2
     CPO 3

   How to Disclose
   Chemical  ID
     CPO 1
     CPO 2
     CPO 3

 Insufficient Submissions
     CPO 1
     CPO 2
     CPO 3
None
$900
$900













to 6,400
to 6,40flb/
0
1
1
0
0
0
0
0
0
0
0
0
0
1
1
0
1
1
0
0
0
0
0
0
0
0
0
0
1
1
0
1
1
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
2
1
0
1
1
0
1
1
0
1
2
0
1
1
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
I/Range specified is likely an underestimate due to reach of possession or control under CPO  3.

-/Range specified is likely an underestimate due to notice requirements and possession or control under  CPO  3.

Note:  For ordinal estimates, 0 = no cost, 1 = lowest cost, 2 = next lowest cost, etc.  Ordinal estimates  are
       meaningful only within issues.

-------
                                      -  63 -
                                    CHAPTER 6

                IDENTIFYING INNOVATIVE CHEMICAL INDUSTRY SEGMENTS
      In  previous  chapters  three factors critical to the economic impact
 analysis were  presented.   In Chapter  2 a segmentation scheme for the industry
 based on chemical properties was presented.  The broad segments were
 inorganics,  high  polymers, amphipathic compounds, elementary organic
 chemicals, organic chemicals, NEC, catalysts, and other chemical products.  In
 Chapter  4 the  kinds of  impacts that section 5 notice provisions may impose
 were  identified.  These impacts were direct out-of-pocket costs, delays, and
 various  costs  tied to uncertainty.  In Chapter 5 the regulatory issues were
 analyzed to  determine how  the costs differed among options and monetary costs
 were  assigned  whenever  possible.  The next step in analyzing the impact of
 these costs  on the chemical industry is to define the affected groups.

      Specifying the groups most affected is difficult because all chemical
 segments are engaged in broadly defined innovative activity.  Although some
 innovations may be new  products, equally rewarding process innovations can be
 placed into  the production process.  By limiting the analysis to those
 segments that  introduce many new chemical entities, we can better focus the
 impact analysis.

      This chapter presents a summary description of the industry segments and
 an identification of the sub-segments which are potential sources of
 substantial  new chemical entity innovation.  The chapter concludes with a
 review of the most innovative segments.

      The segments are discussed in this order:  inorganic chemicals, synthetic
 high  polymers, amphipathic compounds,  elementary organic chemicals, organic
 chemicals, NEC, catalysts, and other chemical products.  Appendix E contains
 the comprehensive profiles of segments and sub-segments briefly discussed
 here.  In addition, ADL's report published in December 1978 includes
 information about various chemical industry segments.
INORGANIC CHEMICALS

     Inorganic chemicals are chemicals that are not carbon-based.  Their
relatively limited ability to bond restricts their capacity to form larger
molecules and, as a result, reduces their structural and functional diversity.

     Lowered bonding ability also accounts for the comparatively minor product
innovation in the area of inorganic chemicals.  The fact that the study of
inorganic chemicals is one of the oldest fields in chemistry also explains, in
part, why innovation is slow and infrequent in this segment of the industry.

-------
                                     -  64 -


          •    Many important products  have  the status of  commodity
               chemicals—standardized  products produced on a very
               large scale.

          •    Commodity chemicals tend to be produced by large
               companies, and the markets are often highly
               concentrated.  Because of the importance of commodity
               chemicals in the inorganic chemicals segment, market
               concentration is higher  in inorganics than in other
               segments of the chemical industry.

          •    There is a great premium on process innovation in the
               commodity chemicals and  in the inorganic chemicals in
               general.  Even though producers find it difficult to
               capture new markets by introducing new products,
               manufacturers may be able to  lower costs, lower
               prices, and capture new  markets or additional market
               shares by improving the  efficiency of their
               production processes.

          •    There is a concentration of research and development
               expenditures in engineering process innovation rather
               than in basic scientific research.

     Despite the limited innovation in  the inorganic chemicals segment, this
section of the industry has developed some new boron, phosphate, and silicon
compounds in recent years.  Some firms  have  been working with rare earth
salts, but they have yet to produce these salts in sufficient quantities for
commercial development,  in addition, process innovation often leads to the
development of new chemical intermediates.

     The direction of future innovation in inorganic chemicals may not
resemble historical trends,  in recent  years an increased theoretical
understanding of inorganic chemistry has emerged from research laboratories.
To date, these theoretical breakthroughs have not lead to the development of
new products.  But there may be some commercialization in the future which
would contribute substantially to further innovation.

     Some of the most important producers of inorganic chemicals are Diamond
Shamrock, FMC, Freeport Minerals,  Kerr-McGee, NL industries, Occidental
Petroleum (through its Hooker Chemical  subsidiary), Olin,  PPG, and Stauffer
Chemical.  Union Carbide, Airco, and Air Products & Chemicals are the three
leading producers of industrial gases,  and DuPont,  NL Industries, and Glidden
are the three leading producers of inorganic pigments.  Major fertilizer
producers include The Williams Companies, Farmland industries, CF Industries,
American Cyanamid, W.R. Grace, Occidental petroleum (Hooker Chemical),  and
U.S. Steel.

-------
                                     - 65 -
SYNTHETIC HIGH POLYMERS

     Synthetic high polymers are chain-like macro-molecules formed by the
chemical linkage of smaller molecular units known as monomers.   Depending upon
the nature of the monomers, these compounds can assume the form of plastics,
resins, fibers, or rubbers and are used in the fabrication of a wide variety
of industrial and consumer goods.

     The basic polymer types currently in use date largely from the period
1930 through 1960.  Some important new polymer types have been developed since
then—polysulfone plastic by Union Carbide (1965), olefin fibers by Hercules
(1962), and Anidex fibers by Rohm & Haas (1969)—but the synthetic tigh
polymers industry is built upon polymer types that were first developed a
generation ago.  A relatively small number of these basic polymer types have
large market shares.  For example, in 1978 polyester accounted for almost half
of domestic non-cellulosic organic fiber production and nylon accounted for
almost one-third.  Forty to 50 basic plastics are available commercially, but
in 1976 polyethylene accounted for 29 percent of domestic production;
polyvinyl chloride, for 15 percent; and polypropylene, for almost nine
percent,  in 1976 styrene-butadiene rubber constituted more than half of all
U.S. synthetic rubber output.16/

     These characteristics of synthetic high polymers—a small number of basic
products that are made in very large volumes which have dominated the industry
for years—-lead to a market structure which is dominated by large plants owned
by large companies.  Concentration in synthetic high polymer markets is
relatively high and especially apparent in the synthetic fibers market,  in
1976 DuPont alone controlled 35 percent of domestic synthetic fiber
production,  and the top four firms (DuPont, Celanese, Monsanto, Eastman Kodak)
controlled two-thirds of domestic production.  The top four synthetic rubber
producers (Goodyear, Firestone, DuPont, Goodrich) accounted for half the
domestic production in 1976, and the top eight for 72 percent.   The plastics
market displays the lowest degree of concentration:  in 1976 the top four
producers (DuPont, Dow, Monsanto, Union Carbide) accounted for 27 percent of
domestic production, and the top eight firms for 38 percent.17/

     The leading plastics producers are large, diversified chemical firms
whose production of plastics represents only a small part of total
production.   The plastics sales of only two, of the top twenty firms in 1976,
exceeded 16  percent of the total chemical sales of the company.18/  sales of
plastics represented less than 10 percent of total chemical sales of most
plastics-producing companies.  Man-made fiber production represents a much
greater percentage of total corporate chemical sales among man-made fiber
producers.   Man-made fiber sales as a percentage of total chemical sales
    i§/Mary K. Meegan, ed., Kline Guide to the Chemical industry (3rd ed.,
Charles Kline & Co., 1977).

-------
                                     - 66 -
exceeded 10 percent for all of the top ten man-made fiber producers,  and for
five of them these sales were more than one-third of total chemical
sales.19/  Most of the leading synthetic rubber producers are tire and oil
companies for whom synthetic rubber represents only a small proportion of
total sales.

     Because of the characteristics of synthetic high polymer markets
discussed above, the chances are slight that a new, radically different
generic fiber able to compete successfully with the current major fibers could
be developed and produced.   First of all,  finding such a fiber in this well-
researched field would be difficult.  Then, building large-capacity plants and
establishing the product in the man-made fiber market would require major
expenditures.

     The last generic fiber introduced was Anidex by Rohm & Haas in 1969.
Rohm & Haas reputedly spent $20 million and 10 years in research,  k 1977
estimate by an industry executive set expenditures of time and money
respectively at $50 million and five to ten years of development to
commercialize a new generic product for broad-based market application.JO/

     Although the roster of basic polymers is quite steady, the synthetic high
polymer industry sends out a steady stream of new polymers that eventually
bring new products into being.  Recently,  there has been intense innovation in
copolymers.  Copolymers and polymers differ only by virtue of the number of
types of monomers in their composition:  copolymers use two or more;  polymers
use a single type.  Both use the same components but in different constructs.

     Producing copolymers is only one way  to introduce new products to the
market.  For example, manufacturers are able to produce literally thousands of
different formulations, tailored to specific uses, by using additives to
change color, flexibility,  stability, tensile strength, electrical conductiv-
ity and other properties.  Manufacturers can also create new products by
varying the methods used to process the basic materials.  For example, in
1977. Hercules, the largest producer of polypropylene, had more than 150
polypropylene products available.2l/  Most of the additives which are used
to produce these products are covered in another segment—Organic Chemicals,
NEC.
    12/Ibid.
         erican Chemical Society,  chemistry in the Economy,  1973,  a  study
supported in part by the National  Science Foundation.

    Ji/"Polypropylene:   R&D is the Key,"  Chemical Marketing  Reporter, March
14, 1977.

-------
                                     -  67 -
AMPHIPATHIC COMPOUNDS

     All amphipathic compounds consist of two parts.  The hydrophilic part/ or
hydrophile, is soluble in water.  The hydrophobic  (lipophilic) part is soluble
in oil.  Usually these molecules are large enough for each part to display its
own solubility behavior.  The parts that are not soluble in the particular
chemical environment surrounding them are chemically attracted to each other.
As a result,  the molecules huddle together with the insoluble parts forming
their own environment inside the huddle.  These huddles, called micelles, are
of microscopic size and are spherically shaped.  Particles not soluble in the
solution may  be dissolved in these micelles.  This is what accounts for the
relatively high dissolving capabilities of many amphipathic compounds such as
detergents.

     Amphipathic compounds comprise an interconnected group of products with
diverse industrial structures and levels of innovation.  They consist of
soaps, synthetic detergents, surfactants, and fatty acids.  Soaps, glycerin,
and fatty acids are primarily natural products and are not innovative in the
sense used in this report.  The synthetic detergent industry is highly
innovative, but much of this is process innovation or the development of new
formulations  of the same chemical substances.  The bulk of the innovation in
synthetic detergents comes from the development of new surfactants.

     Surfactants (short for surface active agents)  are organic compounds that
reduce surface tension.  They wet surfaces easily,  remove and suspend dirt,
disperse particles, emulsify oil and grease, and produce foam.  The first
synthetic surfactants were developed during World War I by the Germans because
of a shortage of natural fats from which natural surfactants (soaps)  are
derived.  New types of synthetic surfactants were developed during the period
between the two world wars,  but rapid growth in the use of synthetic
surfactants did not come until the 1950s with the development of alkylbenzene
sulfonate (ABS).  In the 1960s concern over the accumulation of
nonbiodegradable chemicals (such as ABS)  in the environment led industry to
spend several hundred million dollars to develop linear alkylbenzyl sulfonate
(LAS).22/  LAS is not completely acceptable because it breaks down slowly
and incompletely.  As a result, LAS was banned in many areas, and research on
new surfactants which are both environmentally acceptable and effective is
underway.

     The pressure to develop environmentally acceptable surfactants has
produced a significant degree of innovation in the  surfactant industry.  The
surfactant industry has also developed new surfactants for uses other than in
synthetic detergents,  such as food processing, oil  processing, and emulsion
polymerization.   Overall, ICF believes, based on our analysis (see Appendix
E),  that the level of innovation in the surfactant  industry is high.
         erican Chemical Society, Chemicals in the Economy, p. 200.

-------
                                     - 68 -
     Approximately half the total surfactant production is captively consumed
by detergent companies such as Proctor & Gamble, Lever Brothers, and
Colgate-Palmolive.  Diamond Shamrock, Stepan, ICI United States and a number
of smaller firms sell surfactants to other firms.
     Synthetic detergents are mixtures of surfactants, builders, bleaching
agents, corrosion inhibitors, and other agents that are used to clean surfaces
such as textiles, cooking and eating utensils, walls and floors, and metals.
They are also used for human cleaning.  The growth of synthetic detergents has
been tied to the increased use and development of surfactants.

     The synthetic detergent industry has two separate markets:  household and
industrial.  The domestic household detergent market is dominated by Proctor &
Gamble  (50 percent of the market)  and by Colgate-Palmolive and Lever Brothers
 (each with 17 percent of the market) .  In contrast, the industrial market is
highly diversified.  Most firms are regional:  fewer than 30 companies operate
on a national basis, and only 14 have domestic sales of $25 million or more.
These fourteen companies share 50 percent of the market, and the other 3000
firms in the industry share the remaining 50 percent. 24/

     The synthetic detergent industry is highly innovative, but much of that
innovation involves new formulations and better processes.  The chemical
innovation primarily involves surfactants which were discussed  above.

     Fatty acids, glycerins, and soaps are primarily natural products, and the
industries which produce them are not very innovative.  Fatty acids and
glycerins are produced by a large number of firms, but soap production is
highly concentrated.
ELEMENTARY ORGANIC CHEMICALS

     Elementary organic chemicals are the simple alkanes,  olefins and aromatics
which are obtained directly from petroleum,  coal,  and wood.   Although the
products can be varied to some extent by making minor adjustments in the
production processes, emphasis is always placed on the production of a
relatively small number of basic compounds which serve as  the building blocks
for the entire organic chemical industry.

     Before World War II,  the basic chemical building blocks were derived
primarily from coal tar, a by-product of coke manufacture.   But,  with the
development of the Middle East oil fields after World War  II, petroleum
    il/Meegan,  Kline Guide,  p.  166.

    jl/Chemical Week,  August 22,  1973.

-------
                                     - 69 -
displaced coal as the primary chemical feedstock.  Although the basic chemical
building blocks could also be obtained from gum and wood chemicals,  in practice
they are not.  The major types of gum and wood chemicals are turpentine,
charcoal, rosin, and tall oil.

     Product innovation in this segment is virtually non-existent, but
increases in petroleum prices during the last ten years have stirred renewed
interest in coal-based chemical feedstocks.  Such a switch would result in
significant process innovation in the chemical industry and the development of
new catalytic preparations.  However, the intricate oil-based network,
developed over the last 30 years, would undoubtedly act to slow any conversion
from petroleum to coal.  The increase in petroleum prices has also renewed
interest in biomass-based chemicals, but again, any expansion will be slow.

     The production of petroleum refinery products, by far the predominant
activity in this segment, is relatively unconcentrated and has changed little
over the years.  In both 1958 and 1972, the top four firms accounted for 13
percent of the total value of shipments.25/  petroleum products are produced
by almost all major chemical and petroleum companies.

     Cyclic crudes are organic chemicals which are isolated from coal tar.
Almost all cyclic crudes have identical counterparts produced in petroleum
refineries, and the petroleum-derived products dominate the market.
Consequently, production of cyclic crudes is quite small:  in 1977 production
was $250 million compared to almost $3 billion for the petroleum-derived
products.  The market is also highly concentrated:  in 1972 the top four firms
produced more than 90 percent of the cyclic crudes.26/

     Gum and wood chemicals such as rosin, turpentine, tall oil, and charcoal
are obtained from living and dead trees,  concentration in the entire industry
is relatively high:  in 1972 the four-firm concentration ratio was 49 percent,
and the eight-firm concentration ratio was 64 percent.  The production of soft
wood distillation products is highly concentrated  (a four-firm concentration
ratio of 90 percent in 1972), but the production of other gum and wood
chemicals, including gums, charcoal, and tall oil is much less concentrated (a
four-firm concentration ratio of 36 percent in 1972).27/

     The most important producers of petroleum refinery products are the
leading petroleum refiners—Exxon, Standard Oil of Indiana, Standard Oil of
California, and Texaco.  Major producers of gum and wood chemicals include
Hercules, Monsanto, Reichhold, union Camp, Westvaco,  and Shelton Naval Stores.
    25/u.s. Department of Commerce, Bureau of Census, "Concentration Ratios
in Manufacturing Industry," 1972 Census of Manufacturers, (Washington, D.C.:
Government Printing Office).

    .26/ibid.

    12/ibid.

-------
                                     - 70 -
 ORGANIC CHEMICALS, NOT ELSEWHERE CLASSIFIED

      The  organic chemicals, NEC segment consists of organic chemicals that do
 not  fit into  the other organic chemical categories because of their unique
 structures  and  functions.  As a whole, this segment is highly innovative,
 although  some of its components, such as synthetic organic dyes and pigments,
 and  miscellaneous acyclic chemicals, are not.

      Cyclic intermediates are distinguished from other chemicals by both
 structure and function.  Cyclic intermediates all consist of various
 functional  groups which are attached to, or built into, a carbon framework
 that contains at least one ring.  Cyclic intermediates are typically used in
 the  production of end-use products, although they themselves may also have end
 uses as biocides, photographic chemicals, fuel additives, and antioxidants.

      There  are a few cyclic intermediates which have always dominated the
 industry.   A group of 11 (cumene, cyclohexane, ethylbenzene, styrene, o-xylene,
 p-xylene, alkyl benzene, aniline, chlorobenzene, phenol, and phthalic
 anhydride)  has traditionally accounted for 65 to 70 percent of production
 volume and  40 to 50 percent of sales volume.^8/  gut despite the dominance
 of these  large-volume commodity chemicals, innovation in this component is
 quite high.  In general, there are many paths by which end-use chemicals can
 be produced from the initial feedstock.  Each process innovation, in which a
 new  path  from feedstock to end use product is discovered, may result in the
 production  of several new chemical intermediates for use in the new process.
 In addition, new intermediates may result from the search for new end-use
 products.

      It is  difficult to discuss the structure of the cyclic intermediate
 industry  in its entirety.  The large-volume commodity chemicals typically have
 more than 10 major producers, yet many of the small-volume intermediates are
 produced  by only one firm.   Because there are many ways to produce most cyclic
 compounds,  there is some competition among the different intermediates.  It is
 also important to note that oil companies command a significant share of the
 market, particularly in the production of large volume commodity chemicals.
 They have been attracted to the industry by the close relationship of many
 cyclics to  refinery products.

     Miscellaneous cyclic chemicals include end-use cyclic compounds such as
 gas  and oil additives, many oil field chemicals, photographic chemicals,
 tanning materials,  paint driers, enzymes, and some flavorings.  In general,
 there is a  high rate of innovation among end-use cyclics, just as there is
 among intermediate cyclics.  It is difficult to say much more about the
 innovative  nature and structure of the industries producing miscellaneous
 end-use cyclics, because data on these chemicals are often grouped with
miscellaneous acyclic chemicals, a much larger though non-innovative group.
          . International Trade Commission, Synthetic Organic Chemicals
(Washington, D.C.:  Government Printing Office).

-------
                                     - 71 -
     Miscellaneous acyclic chemicals are frequently used as intermediates, but
also find end uses as gasoline additives, refrigerants, preservatives, oilwell
chemicals, foods, antifreeze, and biocides.  Little else is known about this
component, because its data are often combined with data on miscellaneous
cyclics.  However, the industry experts that ICF contacted stated that this
segment is non-innovative.

     Plasticizers are organic chemicals that are physically incorporated into
plastics to make their manufacture easier or to make the final product more
flexible.  About 85 percent of all plasticizers are used in the plastics
industry  (about two thirds of the total in polyvinyl chloride alone).   The
rest are used in the production of rubber and cellulose products.29/

     Our analysis (see Appendix E) has led us to conclude that there  is
considerable innovation in the plasticizer industry, particularly in  the
relatively new polymeric plasticizers.  Even though the roster of basic
plastics has been relatively unchanged for 20 years, thousands of new end-use
products have been developed during that time from these basic plastics.
Often, a new product will be associated with a new type of plasticizer.

     In the mid-1970s, the leading four firms (Monsanto, W.R. Grace,  U.S.
Steel, and Exxon) produced just over 50 percent of all plasticizers,  and the
leading eight firms (the firms previously named and FMC, Union Carbide,
Eastman Kodak, and Stauffer)  produced just over 70 percent.12/  jn more
narrowly defined markets  (e.g., phthalates, phosphate esters, polymeric
plasticizers), the concentration ratios are somewhat higher.  There has been
some turnover among the leading plasticizer producers in the last 20  years.
Allied Chemicals and Celanese, both among the leading plasticizer producers in
the 1960s, no longer produce plasticizers, and Union Carbide no longer
produces phthalate plasticizers.  U.S. Steel, Exxon, Stauffer, Tenneco, and
BASF Wyandotte have become major plasticizer producers either by buying small
producers and expanding or by beginning production from scratch.31/

     Rubber-processing chemicals include a wide variety of substances that are
used to modify rubber so it can be used in commercial applications.  Not all
of the chemicals used in rubber production are included in this category.
Products such as sulfuric acid, salt, alum, sulfur, zinc oxide, fatty acids,
silicas, clays, carbon black, nylon, rayon and pigments have many uses outside
the rubber industry and are not included in this segment.  The major
categories of rubber-processing chemicals are:
    29/"piasticizers." Modern Plastics, September 1975, p. 47.

    30/Meegan. Kline Guide.

    Ji/Meegan, Kline Guide and chemical and Engineering News, November 13,
1961, pp. 134-135.

-------
                                     -  72  -


     (1)   Accelerators:   cause  rubber to vulcanize  faster and  often
          retard aging;

     (2)   Activators:   increase the  efficiency of vulcanization;

     (3)   Antioxidants:   protect rubber from deterioration  due to
          the action of oxygen  and oxidizing chemicals;  and

     (4)   Antiozonants:   protect rubber from deterioration  due to
          the action of ozone.

     As with plastics,  there  has not been  much innovation in the  roster  of
basic synthetic rubbers during  the last 20 years, but  hundreds of new end-use
products have been developed  from those basic rubbers.   One or more  new
rubber-processing chemicals are associated with  many of  these  products.   New
rubber-processing chemicals have been introduced — established types, as well
as entirely new types.

     The major tire companies are important producers  of rubber-processing
chemicals.  Goodrich, Goodyear, and  Uniroyal produce about  half of the total.
Much of the volume produced by  major rubber producers  is used  in  their own
rubber factories.  It should  be noted that the advantage that  a high degree of
concentration would normally  give the rubber-processing  chemicals industry is
somewhat negated by the high  degree  of  concentration among  rubber producers,
the buyers of rubber-processing chemicals.

     Synthetic organic dyes and pigments'  production is  highly concentrated.
In 1976,  the four largest producers  of  organic dyes (DuPont, CibaGeigy,  Mobay,
and Sandoz) accounted for 55  percent of U.S. sales  in  organic  dyes,  and  the
four largest producers of pigments  (Dutont, Chemetron, American Cyanamid, and
Sun Chemical)  accounted for 48  percent  of  U.S. pigment sales.12/   This
industry is quite mature and  innovation is not high.   EPA has  received a large
number of section 5 notice submissions  for chemicals in  this segment.
However,  ICF does not believe that generalizations  regarding long-term trends
can be made now based on this experience.
CATALYSTS

     A catalyst is a substance which promotes a chemical reaction without
being consumed by the reaction.  After the reaction is over, the catalyst can
usually be used once again to promote the same reaction.  Many commodity
chemicals with a wide range of applications are also used as catalysts.
Sulfuric acid, phosphoric acid, hydrofluoric acid, and caustic soda are
important examples of this type of catalyst.  But catalysts are more commonly
chemicals whose sole function is to promote reactions.  Many contain metals
such as cobalt, nickel, molybdenum, vanadium, platinum, aluminum, palladium,
and copper.  In 1978 about 35 percent of all catalysts  (by value) were used
    Ig/Kline Industrial Marketing Guide, (Charles Kline & Co.). p.  158.

-------
                                     - 73 -
in petroleum refining, 25 percent in catalytic converters, and the rest in a
variety of chemical processes.33/

     According to the Census of Manufactures, in 1972 the four firm
concentration ratio in the catalyst industry was 38 percent, and the eight
firm concentration ratio was 63 percent.   However, the Census of Manufactures
excludes captive consumption and commodity chemicals used as catalysts, so the
real level of concentration of the industry is much lower.  The components of
the catalyst industry are more highly concentrated than those of the whole
industry.  In fact, it is not unusual for one firm to account for more than
half of the sales in a particular market and for three firms to account for
all the sales.  The leading catalyst producers are Engelhard, American
Cyanamid, W.R. Grace, Air Products & Chemicals, UOP, Matthey Bishop, and
Filtrol.

     Because of the secrecy surrounding the process for making new catalysts,
details about the introduction of new catalysts are sketchy.  There have been
instances of a successful new product quickly dominating a market because the
producer demonstrates the capabilities of the new catalyst to customers who
must have it in order to remain low cost producers of their products.
Zeolite, for example, introduced in 1962, accounted for 90 percent of all
catalysts used in catalytic cracking in 1972.li./  it is clear to ICF that
innovation is high (see Appendix E).
OTHER CHEMICAL PRODUCTS

     Other chemical products consist of portions of the industry that are not
classified elsewhere because they have unique characteristics barring them
from any of the other segments.  Sub-segments of this segment are cellulosic
man-made fibers, plastics and sanitary goods, paints and allied products,
adhesives and sealants, explosives, printing ink, carbon black, and salts and
essential oils, NEC.  Most of the components are not very innovative, though
some develop new formulations using new chemicals developed in other
industries.  For example, a new surfactant may be used for new formulations in
the polishes and sanitary goods industry.  Two industries, adhesives and
sealants, and printing ink, do exhibit large amounts of innovation in our
opinion (see Appendix E).

     Adhesives are bonding agents that hold substrates together.  Sealants are
used to fill gaps or joints and to waterproof surfaces.  In the last 10 years
a number of important products in this area have come under suspicion because
of safety and health concerns.  One of the intermediates used in making almost
all epoxy adhesives is a suspected carcinogen, and workplace emissions of this
intermediate are being regulated by the Occupational Safety and Health
    31/"Catalysts I:  A $600 Million Market in Cars and Refineries,"
Chemical Week, March 28, 1979, p. 50.

    li/Chemical Week, November 1, 1972.

-------
                                     - 74 -


Administration (OSHA).  Formaldehyde emissions from the urea formaldehyde
adhesives utilized in construction have also come under government scrutiny.
Concern about the carcinogenity of materials and the high levels of energy
needed to produce many existing adhesives have prompted a search for new
products.  And a number of new products have recently been introduced.

     The adhesives and sealants industry is very unconcentrated.  In 1977
there were 750 manufacturers, with the top 50 accounting for 68 percent of
industry sales.  The large companies usually have a wide product line; the
smaller companies specialize in product or market areas.  The three most
important producers are General Electric, Borden, and Dow.!!/

     The printing ink industry has long had a very low degree of innovation,
but recently a new process and new products have been developed.  Previously,
inks dried on the paper as a solvent evaporated.  But now ink is being "cured"
with infrared heat.  Unlike most firms in the industry, the firms developing
the new technology and products are not simply blending existing
ingredients—they are developing new chemicals.

     There are two types of cellulosic man-made fibers — rayon and cellulose
acetate.  Both have been produced for more than half a century and there
appears to be no chemical innovation in this industry.  Autex, Couralds, and
Akzona are the leading producers of rayon, and Eastman Kodak and Celanese are
the leading producers of acetates.

     Cleaning and polishing compounds include bleaches, scouring cleaners,
alkaline and acid cleaning compounds, disinfectants, and waxes.  This industry
is highly diffuse:  there are over 1,000 manufacturers and fewer than 30
operate nationwide.  Some of the leading firms are Economics Laboratory, S.C.
Johnson, Chemed,  and National Chemsearch.  Most of the firms produce a few of
the product types included in this industry.3j>/  innovation in this area
involves the development of new formulations and mixtures.  The industry does
make use of new chemicals which are developed elsewhere (primarily
surfactants), but no chemical innovation occurs in the industry itself.

     The paint industry is not expected to give rise to much innovation in the
long run.  New paints are new mixtures of existing products or of new products
developed in other industries.  We are currently witnessing a shift from
solvent-based to  water-based paints,  but innovation associated with this shift
is viewed as short-term.   The industry is relatively unconcentrated, although
concentration has been increasing.  The top eight producers had 33 percent of
the market in 1972, and 44 percent of the market in 1977.   There were over
    31/Meegan,  Kline Guide,  pp.  134-135.

    l§/Ibid.

-------
                                      -  75 -
 1,200  paint manufacturers  in  1977,  a decrease of about  20 percent  in the last
 15  years.  Leading  producers  include Sherwin Williams,  DuPont, PPG, SCM, and
 Mobil. IV

     The explosives industry  is  a mature one, and production is highly
 concentrated.   The  three largest producers  (DuPont, Hercules, and  Atlas
 Powder) produced  66 percent of the  industry output in 1977, and the top eight
 firms  produced  83 percent..38/ There is relatively little fundamental
 innovation in this  industry.  The innovations that have occurred have been
 process innovation  and  relatively minor improvements and refinements which
 increase safety.

     Carbon black is a  single element, carbon.  There is no innovation in this
 industry.
 INNOVATIVE  SEGMENTS  IDENTIFIED

     From the  summaries and the detailed analyses provided in Appendix E, it
 is  apparent that historically only a few sub-segments are highly oriented
 toward new  chemical  entity introduction.  These are catalysts, surfactants,
 and miscellaneous  synthetic organic chemicals.  In addition, there are several
 sub-segments in which new chemical entities are frequently produced.  These
 are cyclic  intermediates, rubber-processing chemicals, plasticizers, adhesives
 and sealants and plastics and resins.  Finally, two segments which have
 recently become much more innovative are industrial inorganic chemicals, NEC
 and printing ink.

     As shown  in Exhibit 6-1, this list differs from ADL's in two aspects.
 First, there are some segments ADL did not consider highly innovative that ICF
 did.  Second,  ICF  did not consider some segments innovative that ADL did.  The
 segments considered  innovative by ICF, but not by ADL, were cyclic
 intermediates, rubber processing chemicals, plasticizers and adhesives and
 sealants.39/
    22/Meegan, Kline Guide; U.S. Department of Commerce, 1978 U.S.
Industrial Outlook  (Washington, D.C.:  Government Printing Office, 1978); U.S.
Department of Commerce, 1972 Census of Manufacturers; and U.S. Department of
Commerce, 1977 Census of Manufacturers.

    ;3§/Meegan, Kline Guide, p. 149.

    39/ICF also thought printing ink and some elements of "salts, essential
oils, and other chemical preparations, NEC" were innovative.  They were not
included in the analysis because (1) new product development in the elements
of "salts, essential oils, and other chemical preparations, NEC" is limited to
a very small portion of the industry and is not important to success in the
business; and (2)  the burst of innovation in the printing ink industry is the
result of a recent major breakthrough in the underlying technology; this surge
of innovation will probably be short-lived and the industry will no doubt
return to its historically low level of innovation.

-------
                                     - 76 -
                                   EXHIBIT  6-1

                          INNOVATIVE CHEMICAL SEGMENTS
            ADLa/by

       Surfactants
       Catalysts
       Miscellaneous Synthetic Organics
       Industrial inorganics, NEC
       Plastics and resins
       Synthetic rubber

       Organic fiber, non-cellulosic
       Toilet preparations and perfumes
       Soaps and detergents
ICF and ICF Industry Experts

 Surfactants
 Catalysts
 Miscellaneous Synthetic Organics
 Industrial inorganics, NEC
 Plastics and resins
 Synthetic organic dyes and
 pigments
 Cyclic intermediates
 Rubber processing chemicals
 Plasticizers, adhesives and
 sealants
^/SOURCE:    Arthur D. Little Incorporated, Impact of TSCA Proposed
             Premanufacturing Notification Requirements (Cambridge, Mass.:
             December 1978), p. III-4.

^./Segments considered highly or moderately innovative by ADL.
     The segments considered innovative by ADL, but not by ICF, were soaps and
detergents, synthetic organic dyes and pigments, toilet preparations and
perfumes, synthetic rubber, and non-cellulosic organic fibers.  Soaps and
detergents, non-cellulosic organic fibers,and synthetic rubber all produce
many new products each year; but these new products are usually just different
formulations of existing products, or they incorporate new chemical entities
produced by other chemical industry segments (surfactants, plasticizers,
rubber-processing chemicals).  There is some current innovative activity in
the synthetic organic dyes and pigment segment, and EPA has received a
comparatively significant number of section 5 notice submissions-for chemicals
in this segment.  However, much of the innovation involves mixtures of
existing chemicals, not subject to the notice requirements; and the number of
section 5 notice submissions received under an as yet undefined, highly
uncertain regulatory program cannot be used to draw conclusions about
long-term trends.  Finally, toilet preparations are not subject to TSCA.

     Competitors in Innovative Segments.  The three most innovative
subsegments—catalysts, surfactants, and miscellaneous synthetic organic
chemicals—are all characterized by a few large companies, with relatively
small market shares and with numerous small firms.  The specific firms clearly
active in these markets are:  Engelhard Minerals & Chemicals,  American

-------
                                     - 77 -
Cyanamid, W.R. Grace, Air Products, UOP, Matthey Bishop, Filtrol, Kewanee
Industries, Union Carbide, Girdler, Nalco, all the major oil companies,
Monsanto, Witco, Proctor & Gamble, Borg-Warner, GAP Corporation, Jefferson
Chemical, Rohm & Haas, Schenectady Chemical, BASF Wyandotte, Diamond Shamrock,
ICI, Stepan, Alcolac, Emery Industries, Glyco, Miranol, Henkel, Longa,
Miliraaster Onyx, Mona, Quaker Chemical, Lever Brothers, Colgate-Palmolive, and
DuPont.  In addition, hundreds of smaller companies are very active in  the
surfactant and miscellaneous synthetic organic chemicals segments.

     Each of these companies takes a different view of the importance of
innovation to its survivals.  For some, the introduction of new chemical
entities is the "lifeblood" of the firm, while for others, new chemicals are a
necessary adjunct to the basic product line.  Nevertheless, all chemical firms
appear to be innovative in at least one segment.

     The size of firms is an important factor in identifying competition in
these segments.  Therefore, with the aid of industry experts, ICF developed
Table 6-1 which shows, by sales volume and segment, the number of firms  and
their expected gross margins on new chemicals.

-------
                                                                     TABLE  6-1

         ESTIMATED GROSS MARGIN ON SALES EXPECTED FROM THE HIGHLY INNOVATIVE SEGMENTS   AND  ESTIMATED NUMBER OF  COMPANIES  IN EACH SEGMENT
 Annual
 Sales
 Volume
of Firms
                                         Synthetic
                                          Organic
Cyclic         Rubber                    Chemicals,
                                            NEC
                                                                                                Adhesives
                                                                                               & Sealants
                          Salts  &
                           Other
           Industrial     Chemical
 Printing   Inorganic,     Prepara-
   Ink          NEC	tions,  NEC

0-3m


3-10m

10-200m


>200m

Total *
Firms in
Segment^/
35 /
/
y' 3S%>
20 /
/ 30%
10 X
X
/ 25%
10 X
/
/ 20%

75
10 /
/

10 ''
, 50%
10 /
/
/ 40%
t
20 /
/
/ 30%
*

50
SCO X
/
/ 40%
120 /
S
X 40%
150 x
/'
x 40%
s
30 /
/
, 40%a/

300
0 /
/
/ 35%
/i
5 /
/ 30%
15 ^
/
/ 25%
20 /
/
/ 20%

40
X1
5 /
/
/ 35%
20 /
/ 30%
20 /
x /
/ 25%
20 /
/
/ 25%

65
500 x
/
/ 35%
300 /
/ 30%
150 /
/
/ 25%
50 /
/
/ 25%

1,000
20 X
/
/ 35%
40
/
X 30%
100 X
/'
/ 25%
50 X
/
/ 20%

210
0 X
//
/ 35%
5 /X
/ 30%
10 /
/
/ 25%
5
/
/ 20%

20
150 /
X
X 50%
150 /
/ 50%
175 x X
/
/ 40%
/
50 x
/
/ 30%

525
50 /
/ X
50%
150 /
50%
150 /
X
50%
50 '
/
/ 501

400
                                                                                                                                                       00
                                                                                                                                                       I
                                                                                                            KEY
 a/High R&D expenditures.

 b/Only manufacturers  of new  chemical  entities.

SOURCE:  Consensus opinions of panel of industry experts assembled by  ICF  Incorporated,
         June 1980.
# of companies
that carry out
chemical reactions.
                       Gross margin
                          on sales.

-------
                                      -  79  -
                                    CHAPTER 7

                              MEASURING THE  IMPACT
      In  Chapter  2 we proposed a methodology to analyze and measure the impact
 of  the premanufacture notice rules.  The methodology required that we develop
 data  on  production, sales and use, price, substitutability, financial
 condition,  foreign  trade, and engineering, for various segments of the
 chemical industry.  From these data we were to develop assessments of the
 market concentration, degree of integration, competitiveness, capital
 structure,  and demand and supply determinants.   (These assessments are shown
 in  Appendix E.)  The methodology then required the development of a demand
 forecast and a complementary supply forecast.  The costs of the section 5
 notice requirements would then be applied and changes in demand and supply
 would be measured.

      The steps outlined in the methodology were:

      1.   Identify  the number of new chemicals introduced in a segment
          annually,

      2.   Identify  the costs of section 5 notice requirements for an
          individual new chemical product,

      3.   Perform discounted cash flow analysis of a representative sample of
          new chemicals with and without the costs,

      4.   Calculate the percentage of new chemicals that would not be
          introduced because of the section 5 notice costs, and

      5.   Translate the percentage reduction in new chemicals introduced into
          an economic impact measured in terms of GNP,  unemployment, balance
          of trade, and other similar factors.
     In performing the tasks, several theoretical and empirical problems
developed.

Step 1

     The first problem was that previous estimates of the number of new
chemical entities introduced annually were not broken down in sufficient
detail to allow segment-specific projections.  This meant that any analysis
would have to focus on broader segments in which there are usually some
subsegments that are highly innovative and others that are not.  Consequently,
the analysis would be less precise than had been desired.

-------
                                     - 80 -
Step 2

     As explained in Chapter 5, the costs of the section 5 notice requirements
that are most significant are the ones that cannot be quantified.  These were
the costs of uncertainty concerning restrictive action, trade secret
disclosure, and delay stemming from requests for additional information.

Step 3

     Without representative segment-specific data and without fully
quantifiable costs, any application of discounted cash flow techniques was
limited.

     Discounted cash flow analysis (DCF)  is a technique the business community
uses variously to evaluate investment opportunities.  The concept requires
that annual cash flow returns from a project be discounted at the marginal
cost of capital to yield a value from which the required investment is
subtracted.  The difference is called the net present value.  If the net
present value is greater than zero, the project is profitable.

     ADL used DCF techniques in its preliminary paper published in December,
1978.   (See Appendix C for a summary.)  ICF was able to develop an alternative
set of  results (also presented in Appendix C)  using additional data and taking
into account depreciation costs and the after-tax cost of the section 5 notice
— two  factors ADL did not consider.  Also, ICF was able to perform some
sensitivity analysis that was not done by ADL.  The sensitivity to discount
rates — one of the several sensitivity analyses performed — was the
analytical tool used to learn how uncertainty costs might be considered in
corporate decision-making.  As Appendix C shows, adding the factors ADL
neglected yields very different results.

Step 4

     Without DCF to predict how many new chemicals will be kept off the market
by section 5 notice requirements, an analyst can only measure this decrease of
relying on estimates from independent industry experts or participants in the
segments.  The independent industry experts ICF consulted felt it was not
possible to measure the decrease ex ante.  Many industry participants have
suggested widely different estimates of the decrease.  For example, the
Adhesives Manufacturer's Association noted in response to requests for
information from ICF that only one new chemical entity had been offered to its
sample of members during the past year and that this was much lower than in
the past.  On the other hand, if section 5 notices submitted to date were an
indication of the rate of innovation, then high polymer producers have shown
no sign of a decrease in innovative activity.   Finally, some industry experts
and participants maintain that a shift is taking place from "unsafe" to "safe"
chemicals and that during this transition there will be a temporary lull in
innovative activity.

     In light of the varying best guesses and without the ability to apply DCF
techniques to a representative sample of new chemicals, ICF cannot predict a
quantified level of reduction in new chemical introductions.

-------
                                     - 81 -
Step 5

     If the necessary data were available and if costs linked to uncertainty
could be accurately figured, it would be possible to estimate the number of
chemicals introduced in the past that would not have been introduced because
of the impact of Section 5 notice requirements.  But it would still not be
possible to measure quantitatively the broader economic impacts because, as
explained in Appendix B and in the next chapter, there is no theoretical or
empirical basis, without making unfounded assumptions, for translating changes
in the introduction process of new products into quantitative economic models.

Summary

     Major shortcomings were caused by the inability of formal analytical
techniques to cope with non-quantifiable costs and by the unusual points of
impact of such costs.  As a result, the only way to consider the economic
impact of the section 5 regulations is to make a qualitative assessment of the
likely impact of a reduction in the number of new chemicals introduced in the
innovative segments.  This assessment is presented in Chapters 8 and 9.

     It is important to note here, however, that the availability of data on
historical innovation patterns by industry segment would have made the
qualitative assessments less general.  Such information is not available to us
at this time.

-------
                                     - 83 -
                                   CHAPTER  8

                  MODELS  OF  INNOVATION  IN THE CHEMICAL  INDUSTRY
     To measure qualitatively the impact of the section 5 notice program
requires identifying how the innovation process within a firm works and what
impact the innovation process has on the firm and the economy.  This chapter
discusses the available literature on the innovation process and narrows the
definition of innovation to include only new product development.  We then
describe three models of the innovation process based on empirical data and a
fourth model that incorporates all of the activities that are part of the
innovation process and that may be affected by the notice requirement,  in
conducting the literature review, ICF drew heavily upon a preliminary working
paper done by ADL which is summarized in Appendix B.  This was augmented by
our own review of public materials and by a conference on chemical innovation
at which five industry experts provided additional thoughts.  (Appendix B
expands on the discussion in this chapter and assesses the feasibility of
developing computer models to measure the impact of the section 5 notice
program.)
SUMMARY OF LITERATURE

     There are four primary bodies of pertinent literature:

     (1)  The formal micro-economic literature on industrial innovation.
Neoclassical growth theory has struggled for many years to deal with
technological change,  in the 1960's economists developed a means of
incorporating technical change in the neoclassical growth model as the "theory
of induced innovation"—but in a very formal manner of little use in this
task.  The theory does not help us because, first, it deals with what we  would
call "process innovation"—changes in the prices and quantities of input
factors for given outputs.40/  New product development generally falls
completely outside these theories.

     Another problem is that until recently the theory has been developed
largely at a macroeconomic level.  Also, there seems to be little theory  in
the economics literature on the role of uncertainty in the innovation
process—either at the firm level, industry level, or macroeconomic level,  it
    40/Even here, there is much confusion and apparently no clear way
empirically to separate movement along an isoquant (factor substitution)  and
shifts in the isoquant (technical change).  Because of the difficulty in
separating the two  (in either case, the optimal factor ratio changes in favor
of more use of the less expensive factor), several economists have recommended
abandoning the distinction entirely.

-------
                                     - 84 -


is possible to distinguish between "technical uncertainty"  (the uncertainty
that the scientific and engineering development process will produce something
of potential use)  and "market uncertainty" (the uncertainty about the degree
of demand for a product in the future and about how actions of competitors may
affect payoffs).  It is also theoretically possible to model "technical
uncertainty" through a probability cash flow model, "market uncertainty"
through a game-theoretic approach, and the two together through incorporation
into a portfolio-choice capital budgeting model for R&D expenditures.  But
there is almost no body of microeconomic theory to rely upon, as there is in
areas of industry and firm behavior.

    It is worth noting that economists are now moving  into  this field in an
attempt to extend the current body of theory.   Nevertheless,  at the present
time, economic modeling of product innovation (quite apart  from the even more
difficult issue of section 5 notice impact on that innovation)  places us at a
frontier of economic theory without tested or accepted approaches upon which
to rely.

     (2)  Empirical studies of innovation.  There are extensive studies, gener-
ally at the industry level, which attempt to analyze and understand the
industrial research and development process.   These range from studies which
are quite analytical and use formal models and actual  historical data (usually
time-series or pooled time-series/cross sectional multiple  regression analysis)
to qualitative case studies,  industries covered are primarily drugs, chemi-
cals, glass, steel, or petroleum.   Attempts are made to relate the amount of
R&D expenditures to a wide variety of firm and industry parameters and to
relate "outputs" such as sales of new products to amounts of  R&D expenditures.
Despite the recent work of Mansfield, Nelson,  Iverstine,  and  others,  there is
a paucity of empirical detail on the resource allocations to the various
stages of the R&D process, on how and wn* firms make R&D allocation decisions,
on how firms evaluate R&D expenditures,  or on how firms (in a game-theoretic
sense) decide what to patent or bring to market.  As discussed below (and in
the ADL survey), this literature provides some ideas about  the determinants
and decision-making in the R&D process,  but the bulk of it  is not very useful
for several reasons:

          •    It is based upon old data (1954-1959) and thus upon
               structural and societal relationships and forces which
               arguably no longer pertain.

          •    it is conflicting in many of its findings.

          •    It is often based upon questionable data and
               surrogates for innovation (due to lack  of useful data)
               and at rather high levels of aggregation,  even at the
               large company level,  there are many very different
               "businesses"  being  aggregated.

          •    There are few studies  of  the chemical industry and few
               significant conclusions from these studies.

-------
                                      - 85 -
      (3)   Specific  studies  on the  effects  of  government  regulation on
 innovation.   There  have been reasonably  extensive  studies  of  the  impacts of
 the  1962  Amendments to the  Food, Drug  and  Cosmetic Act on  innovation in the
 pharmaceutical industry and less extensive studies on the  effects of the
 Federal Insecticide,  Fungicide, and  Rodenticide  Act (FIFRA) on  the pesticides
 innovation process.  As discussed  below, although  these  studies highlight some
 areas  of  interest to TSCA,  there is  little of use  here,  either:

           •     the  studies  are generally post hoc  analyses with
                single equation approaches  or  international
                comparisons,  neither  of which  is  particularly  helpful
                to TSCA's need to develop a method  to assess impacts
                ex ante;

           •     the  studies  deal with regulations very different from
                notice requirements (and  different  from TSCA more
                generally);  or

           •     the  studies  are data  constrained, in some cases  use
                questionable  surrogates for innovation, fail to
                separate the  regulations  themselves from concurrent
                societal forces, and  provide little data on how
                regulations  have affected R&D  decision-making  by
                firms.

      (4)   Studies from the  "business literature".   There is a growing litera-
 ture on R&D,  innovation,  and technical change arising from managerial sources,
 which  provides some useful empirical insights.   The literature  ranges from
 studies of R&D strategy  to  the extensive work done on capital budgeting,
 decision-making under  uncertainty  (particularly  Bayesian statistics), and
 portfolio/risk theory.   Although these provide useful grounds upon which to
 model  certain  business decision-making, much  of  this literature is too
 abstract  and complex  to  be applied to  the  variety  of common business
 decisions:  there remains a  significant gap between the case  studies of
 individual firms  and  a formal analysis of  a business decision made under
 uncertainty.   Yet,  the need  to understand  the effects of notice requirements
 on the chemical industry  remains,  but  the  literature discussed  previously in
 (1),  (2),  and  (3) is of  little use.  In conclusion, if we  are going to learn
 how notice requirements  affect the decisions  made  by the chemical industry, we
 must ascertain first how and why chemical  industry officials  allocate R&D
 resources, assess probabilities, and make  decisions about  development and
 commercialization.
LITERATURE REVIEW

     We have noted in the summary above that the existing literature has
little to offer in developing a formal approach to analyze the impacts of
notice requirements on the chemical industry.  Because of this, and because
the literature review in Appenxix B is generally thorough, this section is a
brief summary.  Numbered citations in the text refer to the bibliography in
Appendix B.

-------
                                     - 86 -


     There are four areas of the literature:

      (1)  Microeconomic Theory of Industrial Innovation,

      (2)  Empirical Studies of Innovation,

      (3)  Specific Studies of the Effects of Government Regulation
          on  Innovation, and

      (4)  "Business" Literature.

 Microeconomic Theory of Industrial Innovation

      In  this  category, the review by ADL is more than adequate; therefore, the
 conclusions documented in Appendix B are summarized and restated:

          •    The theory of induced innovation that is incorporated
               into neoclassical growth models addresses only
               process innovation.

          •    Research and development costs are seldom even
               included in models; when they are, there are serious
               difficulties.  Economists argue about whether or not
               a firm can choose the direction technical innovation
               will take; some believe that to admit that firms can
               control the rate or direction of technical change
               means abandoning competitive models.

          •    Most past efforts have been in a macroeconomic
               framework.

          •    Existing models do not deal either with technical
               uncertainty or market uncertainty or with how
               allocation decisions would (or should)  change with
               changes in these factors.   TSCA regulation is
               expected to affect both technical uncertainty (by
               increasing the number of technical criteria a product
               must respond to)  and market uncertainty (by adding
               costs, increasing delays,  weakening competitive
               position, reducing payoffs, etc.); there is no
               systematic means to deal with these regulatory
               effects in these models.

Empirical Studies of Innovation

     Conclusions in these studies are often contradictory.   The question of
how diversification affects innovative activity is a good example of such a
contradictory finding.   Grabowski finds a positive effect,  Scherer finds no
systematic effect,  and  Weiss finds a negative effect.   Even when there are
findings—that R&D  intensity, as measured by R&D employment regressed against
sales,  tends  to rise  with firm size in the chemical industry—there are no
straightforward implications for the task at hand.

-------
                                     - 87 -
     Here are some broad findings from this literature:

          •    At the firm, industry, and economy level, there is
               high correlation between R&D expense and economic
               growth/ productivity.  The ADL report cites R&D
               intensity as a "source" of growth (although it notes
               that causality may run in both directions); more
               conservatively stated, high R&D is associated with
               high growth and positive changes in productivity.

          •    Chemical industry R&D intensity appears to rise with
               firm size, yet R&D per sales dollar is higher with
               smaller firms.

          •    There is no consensus on the relationship between
               innovative activity and industry concentration or
               degree of product differentiation.

          •    Several studies suggest after-tax average returns on
               R&D capital of 15 to 20 percent.

          •    One expert (Kamien)  estimated that more than
               three-fourths of U.S. industrial R&D was directed
               toward new products rather than new processes.

Specific Studies of the Effects of Government Regulation on innovation

     ICF looked closely at the previous work in the pharmaceutical and
pesticides industries because, despite its ex post nature and regulatory
differences from TSCA, the general approach they used at least addressed
directly the relationship between regulation and innovation.  Thus, the
following section provides a little more detail than the previous two summary
sections; without doubt, TSCA will be subject to analyses like these in the
future.

     This summary of the literature is a brief guide to the effects of
regulation on innovation in the pesticide and pharmaceutical industries and
how these effects may relate to the chemical industry.  References are made to
major empirical studies in the bibliography in Appendix B.

     The pharmaceutical industry

     A number of studies have been done on the effects of the 1962 Amendments
to the Food, Drug and Cosmetic Act on innovation and R&D in the pharmaceutical
industry.  The most significant findings are summarized here in terms of the
following four hypothesized impacts of regulation on innovation.  Regulation:

     (1)  lowers return on R&D due to increased costs, risk, and
          development time associated with regulatory compliance and
          leads to a lowered rate of investment in R&D and potential
          decline in innovation;

-------
                                     - 88 -
      (2)  changes the pattern of R&D and innovation  (e.g., away from
          basic research to emulative products or from product to
          process innovation);

      (3)  changes market structure which may affect the rate and
          pattern of innovation; and

      (4)  promotes international transfer of resources and
          technology from countries with strict regulatory
          requirements to those with less strict requirements.

      Level of Innovation.  Based on Baily's study (7) showing an increase in
new drug discoveries resulting from increased R&D expenditures, several
authors have investigated the expected rate of return on pharmaceutical R&D
investment  (19, 83).  Schwartzman, for example, finds a drop in the rate of
return from 11.4 percent in 1960 to 3.3 percent in 1972 due to increased costs
posed by federal regulation.

      Hansen (38) estimates a development cost of $54 million (1976 dollars)
for a new chemical entity (NCE)  and about nine years for the NCE to reach the
market.  Many others have documented the decline in introduction of NCE's
since 1962.  (See 43, 55, 75, 91, 92.)  The exact portion of that decline
directly attributable to regulation is, however, still open to question.

      Both Baily (7) and Peltzman  (75)  find a statistically significant negative
effect of regulation on the number of new chemical entities introduced in the
U.S.  after 1962, but serious criticisms of these empirical models exist.  For
example, Baily's supply-side model ignores the effects of changes in demand,
and both Baily and Peltzman neglect to provide an adequate measure of "research
depletion" or to explore the possibility that any exhaustion of research
opportunities actually existed.

      Using a comparative analysis of the British and American cases to get at
the same issue, Grabowski, Vernon, and Thomas  (32)  find a six-fold decline in
R&D productivity in the U.S. between 1960-1961 and 1966-70, compared to a
three-fold decline in the U.K. in the same time period.  The difference between
the two countries may be part of the well-known "drug lag", suggesting that it
was the particular legal and institutional form of regulation in the United
States that has been most responsible for negative effects on innovation.

      As for the three-fold decline in the U.K. rate, it is again not possible
to distinguish the effects of increased regulation in that country from
research depletion or from changes in demand due to a different social and
medical context.

     An Arthur D.  Little paper (4) emphasizes the importance of a simultaneous
equation approach  to modeling innovation in the drug industry, rather than
single-equation models which model either the supply side or the demand side,
but not both together.   The 1962 legislation in the U.S. (and regulation in
other countries) was one aspect  of a change in attitudes directly related to
the thalidomide incident which arguably affected demand for new drugs.  On the

-------
                                     - 89 -
supply side, the possibility of exhaustion of research opportunities,  perhaps
due to gaps in scientific knowledge, is an important and controversial issue
that has not been effectively addressed in econometric studies of  this subject.

     Type of innovation.  The position taken by the FDA is that the decline in
new drugs since 1962 has been a result of the focus on new chemical
modifications of existing drugs, rather than the result of inhibiting  new
therapeutic advances.  This position is a controversial one both factually and
in its tacit assumption that society has not lost much from the decline in
slightly modified drug products.  The nature of new drug innovation is dealt
with in several sources (see 39, 75, and 82).

     Market Structure.  A great deal of literature has dealt with  questions of
market structure and innovation in the pharmaceutical and in several other
industries.  Most writers have explored the relationship of firm size  to
innovation in an attempt to assess empirically Schumpeter's hypothesis that
large firms are associated with greater inventive output (81).  (See,  for
example, 5, 12, 24, 35, 37, 59, 79, 82, and 89.)  It appears that there is
still little consensus on the validity of this hypothesis.

     Kamien and Schwartz (46) provide a useful survey of research  done on this
question which takes into account the two aspects of the "scale" issue: the
effect of firm size oh efficiency for a given size R&D facility and the
efficiency of different sized R&D facilities within a firm.  More  recent
studies have investigated the effects of rivalry on the innovation process
 (31, 47, 66).

     As in the pesticide industry, active ingredient R&D on Pharmaceuticals is
conducted by a limited number of firms (approximately 100) and is  generally
confined to the larger companies because of high development and testing costs.
One study finds that, although some increased concentration has been evident
in innovation among the larger firms since 1962, the innovational  role of the
small firms that do perform R&D has diminished only slightly  (73).

     International Transfer.  A fair amount of evidence has been amassed to
indicate that since 1972, pharmaceutical innovation has shifted from the U.S.
to countries abroad — a shift due to differential regulatory requirements.
The pharmaceutical industry has a strong multinational component,  so that a
shift in R&D and production can be easily accomplished,  it is difficult,
however, to distinguish the effects of regulation on this trend from other
cost-related factors.  (See 32, 33, and 73.)

     The Pesticide industry

     Very little literature exists on the effects of regulation on innovation
in the pesticide industry.   The primary sources of data are the industry
surveys made by the National Agricultural Chemicals Association (69).   These
surveys, however, vary from year to year in the number of firms participating
and the categories used, so that a detailed analysis of trends in innovation
is not always possible.  Some idea of recent patterns in the industry  can,
however, be gleaned from the NACA data and other sources.

-------
                                     - 90 -


     Expenditures on pesticide research and development have continued to rise
in the last decade, keeping pace with the increase in pesticide sales.  The
cost of bringing a new chemical to the market,  though,  has risen from
estimates of $5.5 to $6 million in the early 1970s to a current estimate of
$15 to $20 million (including the cost of unsuccessful compounds).   Total
elapsed time from discovery of a pesticide to marketing has also been
increasing from an average of 60 months in 1967 to 110 months in 1977.
Although the share of R&D funds spent on discovery and commercialization of
new products has been roughly constant in the last five years (near 65
percent), a much higher proportion of R&D expenditures is now devoted to
regulation-related activities.  The result is that fewer new commercial
products have been forthcoming, and it is assumed that the number of annual
new registrations of active ingredients with EPA will not exceed 10 to 15 in
the near future.

     Unfortunately, no precise linkage of the changes in pesticide  innovation
with the effects of FIFRA regulation is possible.  The pesticide industry is
relatively mature  (70) and it is generally believed that most major markets
have already been filled with effective products, though product obsolescence
is common in the industry.  On the other hand,  registration requirements have
probably exacerbated the trend toward increasing modification of existing
pesticides for new uses rather than developing  new chemical entities.

     There is, in addition, a wide variety of anecdotal evidence that direct
and indirect effects of regulation have hindered both the development of
products for minor crops and uses and innovative work in the development of
biological alternatives to chemical pesticide control,   it is believed that
the few small firms that are involved in pesticide R&D have been involved in
some of this innovative work (23).  The effects of regulation on market
structure, although limited, may therefore be indirectly injurious  to some
sources of innovation.

     The Chemical  industry

     The relevance of the experience of the pesticide and pharmaceutical
industries under pre-market regulation to the effects of TSCA on the chemical
industry is severely limited.  The methodological problems in previous studies
restrict their use in studying the chemical industry.  There are also
substantive differences among the industries.

     Pesticides and Pharmaceuticals are not typical segments of the larger
chemical industry.  The levels of R&D conducted in the industries differ
significantly,  (about 10 percent of sales in Pharmaceuticals, eight percent in
pesticides, and two to four percent for chemicals).  Although we do not have
enough data on the chemical industry to be certain, we suspect that other
important differences exist as well.  Examples are the expected value of sales
of new products, the barriers to entry into R&D, and in general, the effect of
firm size and scale of R&D facility on innovative potential (see 26 and 30) .
The comparability of the chemical industry to the pesticide and pharmaceutical
industries on two matters — the role small firms play in each and  the
research opportunities each enjoys — could also modify the effects of
regulation on the industry.

-------
                                     - 91 -
     Variations in the regulatory process that different industries undergo
may also be important.  Premanufacture notification does not require specific
testing and has an established review period of 90 days (may be extended to
180 days), unlike the FIFRA pesticide registration program and FDA's
Investigative New Drug (IND) approval system.  Similarly, the treatment of
confidential information in the regulatory process may affect not only the
rate of innovation but the type of innovation (i.e., development of either
subtle variations on existing chemicals or entirely new types of chemicals)
and the location of activity.

     Measurement of innovation appears in the literature on Pharmaceuticals as
a serious obstacle to empirical study.  The problem is aggravated in the case
of the chemical industry, where even the annual number of new chemical entities
introduced is unavailable from a single authoritative source.  Some of these
issues are dealt with in (52) , (66) , (78) , and (79) .

     For these reasons, it is not likely that the major studies done on the
pharmaceutical industry or the experience of the pesticide industry can be
easily extrapolated to predict TSCA's effect on the chemical industry.  No
doubt some of the determinants of innovation in the industries will be common.
However, it is not our expectation that the results of regulation in the
chemical industry as a whole will conform to the results in two of its rather
atypical sectors.  For example, the nature of R&D in both Pharmaceuticals and
pesticides centers on product rather than process — not so for at least the
high-volume segments of the chemical industry; most new products (active
ingredients)  in the former cases are expected to penetrate large markets and
have development costs of $20-30 million over the course of five to 10 years.

     This calls attention to the heterogeneity of the chemical industry and
the need to disaggregate the various sectors it encompasses.  It appears that
pesticides and Pharmaceuticals are at one end of the spectrum of chemical
industry subgroups in terms of innovation.  As explained later, other patterns
of innovation and of regulatory effects on innovation are more characteristic
of the chemical industry in general.

The "Business" Literature

     Apart from the extensive theoretical literature on capital budgeting,
decision analysis under uncertainty, and portfolio risk analysis, the business
literature also contains examinations of individual firms or innovations and
behavioral studies relating organization and management approaches in R&D to
successful results.  Both these approaches look carefully at actual firm
performance and have produced the following conclusions:

          •    The bulk of R&D projects in large chemical firms are
               relatively safe from a technical point of view  (59).

          •    Expected rates of return (if successful) were about
               30 percent—arguably about the same, after
               probability adjustments for some failures, as other
               capital investment (59).

-------
                                     -  92 -
               Firms devoted about nine percent of R&D funds to
               "basic  research," 45 percent to "applied research,"
               and  46  percent to "development" (60).

               Even for research-intensive industries like
               chemicals, innovations are often based upon
               technology or research derived from organizations
               other than the innovating firms (68).

               There is considerable anecdotal evidence that in
               significant parts of the chemical
               industry—especially for small firms and in the
               specialty chemicals area—a great deal of product
               innovation is closely tied to customers and the
               market  (i.e., there may be little or no "market
               uncertainty").

               There has been considerable development of the
               product life-cycle theory which may have application
               in longer-term modeling of new chemical products.

               In case studies of successful new product innovations
               with major impacts, factors cited (49, 76)  as
               significant  (ceteris paribus)  in promoting
               innovations are:
                                                                Effect on
                                                                Innovation
                    firm size                                        +
                    relative R&D spending:  firm to industry         +
                    industry growth rate                             +
                    stability of industry in terms of
                      structural products
                    degree of involvement of R&D groups              +
                      with potential users
                    degree of status, authority, and involvement     +
                      of top R&D manager with the project
                    in-house initial support for the innovation      +
                      on commercial grounds
                    relative sales effort for the new product,       +
                      including publicity/advertising
MODELS OF INNOVATION

     The ICF analysis identified four models of innovation in the chemical
industry.  Three of these models are typical of different size firms.  The
fourth model, which is a theoretical model, was developed on the basis of the
literature and the other three models.  The models explicitly incorporate
different types of R&D decision-making as the basis for modeling the
innovation process.

-------
                                     - 93 -
Three Models Based on Firm Size

     During the meeting of industry experts convened by ICF, considerable
attention was directed to the  importance of firm size in understanding new
product  introduction decision-making and, therefore, section 5 notice
requirement impacts.  Numerous schemes for categorizing firms by annual sales
volume have been proposed.  SOCMA, in a presentation to EPA staff, defined a
small firm to be one with annual sales of $30 million or less.  However, their
categories were not designed to look specifically at new product introduction
decision-making.  The industry experts described four categories of firms
based on annual sales volume which were targeted at understanding innovation.
The four sizes were:

     1.   $0 to $3 million,

     2.   $3 to $10 million,

     3.   $10 to $200 million, and

     4.   greater than $200 million.

     Characteristics of Firms  in Each Category.  The smallest firms include
the industry entrepreneurs.  These firms usually are started as university
spin-offs or by employees of large firms who decide they want to try it on
their own.  Several hundred firms start this way annually and most of them
fail.  If after a few years the firm has not reached the $3 million level, it
will most likely cease to exist.

     The second category is the $3 to $10 million firms; these are the
entrepreneurs who have become successful.  They, like the smaller companies,
typically have one key person who is the "research and development,"
"marketing," and "operations" executive.

     The third size category of firm is the $10 to $200 million operation.  In
this category will be found companies which have become too large to have only
one decision-maker.  They are beginning to develop organizational divisions
with special tasks (R&D, sales, etc.).  The mix of product sales volumes in
these companies will vary from $2,000 for some new chemicals to several
hundred  thousand dollars or more for most mature products.  Occasionally these
firms will have some products that sell over a million dollars a year.  These
firms are frequently characterized by defensive marketing strategies for new
chemicals — strategies that are based on fear of competition from industry
giants and that may have consequences for section 5 notice requirements.

     The companies with a sales volume greater than 200 million a year are the
industry giants.  These companies are usually organized into divisions or
departments based on product markets.  They usually have a small basic
research unit at the corporate level and numerous applied research units
within the operating divisions.  The basic research group at headquarters
works on chemicals that might not be commercialized for 20 years.  The applied
groups are always seeking new products for current markets.  Many of these
companies also have commercial development units which link events in the

-------
                                     - 94 -
marketplace to R&D projects that should be funded and thereby identify new
business opportunities.  These groups usually consist of both sales and R&D
personnel.  By interacting with the customers to discover their needs, the
group is able to identify what characteristics a new chemical must have to be
successful,  using this information, the applied R&D unit in the division then
creates such a chemical.

     New Product Development Models,  in these large companies, formal
mechanisms allocate resources to R&D and formal processes choose among R&D
projects,  in many of the industry giants the decision-making committee meets
many times a year.  At these meetings they assess the status of the current
projects from a technical viewpoint.  They review the economic and financial
aspects of the projects, and then they decide which projects to continue to
fund in what amounts.  Formal financial analysis plays a much greater role in
the process in this size company than it does in smaller companies because the
financial analysis is the means by which all sectors of the company
communicate.

     In the $10 to $200 million companies, decisions on allocating resources
to R&D and on new products are made informally.  Usually these firms have an
executive in charge of R&D projects who meets with an Executive commmittee to
make decisions about  (1) how large an R&D budget to have; (2) what R&D
projects to fund; and  (3) what projects to commercialize.  This committee does
not usually base its decisions on formal financial analysis, because such an
analysis is not necessary for communication.  These firms are not large enough
to have commercial development groups or to afford basic research.  Almost all
new chemical introductions result from a talented researcher or from a
customer's specific request.

     In the smallest firms (under $10 million in annual sales), the new
chemical introduction process is usually based on the entrepreneur's technical
skill.   The entrepreneur does not have any formal processes for deciding how
to allocate resources,  in the course of managing daily operations, he makes
these decisions implicitly.  Nevertheless, the repartition of time among
sales,  operations, and research reflects the relative returns expected from
these efforts.  Thus, the entrepreneur indirectly makes resource allocation
decisions.

     Theoretical Model.  Each of these empirical models can be accommodated in
a theoretical model which is presented in Exhibit 8-1.  The model identifies
five blocks of activity in the new chemical development process.  These
activities are:   (1)  resource allocation to R&D; (2)  R&D allocation to
specific products; (3) creative research process; (4)  commercial development;
and (5)  market introduction and response.

     It appears evident from the diverse literature addressing this process
that each block of activity in the theoretical model may be distinct and may
be influenced by different factors.  Further, the nature of the activity will
vary widely depending upon which segment of the chemical industry is being
considered and by the size of the firm within that segment.

-------
                                    - 95 -




                                 EXHIBIT  8-1

                              INNOVATION PROCESS
         BLOCK 1
   Resource Allocation
(Capital,  Personnel Plans)
                 BLOCK 2
          R&D Allocation (Basic,
          Applied,  Development:
             Product/Process)
                               BLOCK 3
                    Creative Research Process
                    (Ideas,  Plans,  New,  Non-Tested
                         Products/Processes)
                                     BLOCK 4
                              Commercial Development
                              (Products/Processes)
                                              BLOCK 5
                                        Market Introduction
                                            and Response
                                                                  Outputs

                                                                • Sales
                                                                • Profits
                                                                • Substitution

-------
                                     - 96 -
Allocation of R&D Inputs  (Block 1).  The firm decides to allocate dollars,
people, office space, contract monies, and other resources to research and
development.  Some of these allocations may be influenced by existing R&D
projects or expenditures; may be required by specific market forces; may be a
result of a formal decision process or a rule-of-thumb; may be done at various
levels of explicitness; and may be on an annual or project budget.  But the
firm has a cost of capital and alternative allocation possibilities (capital
investment, dividends, hiring, etc.).  In a formal or informal manner, a
decision maker in the firm is assessing the potential return from R&D in
relation to other uses of capital.

Allocation of R&D Resources to Projects (Block 2).  Possibly in combination
with Block 1, on an incremental or total basis, somebody in the firm sets
objectives, defines projects, and allocates resources to projects.  These
projects may be long-term or short-term; "basic",  "applied" or "development"
research, aimed at new products or processes or at cost reduction on existing
processes; focused at a specific market opportunity ("demand-pull")  or just an
interesting product quality ("science-push").

The Research Process  (Block 3).  According to some organization and management
rules or procedures, the project research proceeds.   The activity could range
from a scientist alone at a bench for months, to close work with a potential
user, to work under contract at a university; it could be cost engineering
rather than "science"; it could be managed in a wide variety of ways;  it could
be under intense time pressure or very loose; it could take days or years.

Commercial Development (Block 4).   After the research process,  a process
change or a product is created.  A decision is made as to whether and  how it
should be brought to market.  Assessments will typically be made of the
potential market, the likely cost of production,  the possible market price,
what the competition will do,  the probability of success,  timing, the  firm's
capital availability in relation to other opportunities,  whether it can or
should be patented (if it has not been in Block 3),  and how it  should  be
introduced or promoted.

Market Introduction and Response (Block 5).   Initially, and over what  may be a
very long product (or process)  life cycle,  the product is  produced and sold.
Sales may go up,  costs may decrease as volumes increase,  a similar but
inferior product may be driven from the market, prices may rise or fall
depending upon demand and competition,  and a host  of impacts may be traced
throughout the industry  and economy.

-------
                                      -  97  -
                                    CHAPTER 9

          QUALITATIVE ANALYSIS OF  IMPACT ON NEW CHEMICAL INTRODUCTIONS
     In Chapter  8 we  identified five blocks of activity important to the
 innovation process.  Earlier we had identified three types of costs that
 chemical companies face.  These are (1) direct out-of-pocket costs; (2) delays
 in the introduction  of new chemicals; and  (3) uncertainty regarding direct
 costs, length of delay, possible trade secret disclosure, and possible use
 restrictions.   According to independent industry experts, testimony in the
 public record by affected interest groups, and the literature concerning the
 introduction of new  products, economic impacts of section 5 notice require-
 ments will result from these kinds of costs.  This chapter summarizes the
 effects of proposed  section 5 notice requirements on the introduction of new
 chemicals.
IMPACT ON COMMERCIALIZATION ACTIVITIES

    The five blocks of activities associated with new product introduction and
identified in Chapter 8 were resource allocation, R&D allocation, creative
research process, commercial development, and market introduction and
response.  Notice requirements will increase the expected costs associated
with Blocks 1 and 2.  These greater costs will decrease expected returns from
these activities, which will probably cause fewer dollars to be allocated to
these blocks.  Block 3 is presently being affected because companies are
probably beginning to rule out certain toxic families of substances as
potential products.  Similarly, the out-of-pocket costs and uncertainty would
clearly affect decisions about what to bring to market  (Block 4).  Impacts on
Block 5 would be felt if notice data made public allowed competitors to bring
alternative products to market more quickly.

    The effect of the direct costs and uncertainties on Blocks 3 and 5 will
not vary with the size of the firm.  All firms can be expected to move toward
innovation in "safe" chemicals.  What constitutes a "safe" chemical depends on
the state of chemical knowledge and EPA administrative practices with regard
to various chemicals.  All firms will also be affected by having some notice
information publicly-available which would allow competitors to bring
alternative products to market more quickly.  The effects on Blocks 1, 2, and
4, will differ however,  according to company size.

Large Companies

    Independent experts have estimated that the major firms will not
substantially change the direction or depth of their new product programs
because of the costs of notice filing, so long as those costs are only a few
thousand dollars.  Yet,  industry experts have also said that the imposition of
the notice filing cost will, in aggregate, cause industry giants to reconsider
the value of some R&D projects within new product programs  (Block 3 impact).

-------
                                     -  98 -
 Similarly,  the  uncertainty about EPA actions may cause at least a  temporary
 hiatus  in some  R&D programs.  In sum, for the large companies, independent
 industry experts  have  stated that the uncertainty costs will have  a greater
 impact  than the cost of filing the notice.  One expert, a retired  executive of
 an  industry giant, believed that (1) the cost of filing the notice was
 de  minimus,  and (2) the large companies were already moving away from research
 in  potential toxic fields, so that the effect of the final notice  regulations
 on  these companies would be nil.  This view of uncertainty costs is not shared
 by  others.   As  pointed out in Appendix C, it could be argued that, for very
 large volume-low  profit or very high profit-small volume products  (products
 typical of  the  industry giants), notice filing costs are of little
 significance to the decision process.

 Mid-Size Companies

     The impacts on Block 1, 2, and 4 for mid-size companies ($10 to $200 mil-
 lion annual sales) will be substantially greater.  The key differences between
 this size firm  and the giants are the expected sales volume of particular new
 products and the  access to expertise about the government program.  Both these
 differences  will  result in greater impacts on these smaller companies.

     In  the  ICF  meeting on chemical innovation, participants characterized
 these firms  as  being fairly uninformed about regulatory and legal matters.
 Without expertise in regulatory matters, they would likely choose to take
 actions that minimized their exposure to the threats posed by regulation.
 Thus, they  should be expected to (1) steer their new product development away
 from suspected  chemicals, and (2)  minimize their regulatory and legal costs by
 simply  not marketing any chemical on which EPA makes a request for additional
 data.

     The second  difference which will cause these medium-size companies consid-
 erable  problems is the volume of sales of their new chemicals.  Smaller
 chemical companies often obtain a portion of a large company's market when
 they introduce  new chemicals.  Small companies without extensive legal
 expertise may be unable to protect confidential information from being
 disclosed.   Public notices may reveal information about the companies and
 chemicals.   The notice process will therefore provide large chemical companies
 with an inexpensive way to track their small rivals' new product development.
 This knowledge  will inevitably result in a competitive advantage for the
 largest companies.  There is a second way in which this difference will hurt
 smaller companies.  Their lower volume sales of new chemicals generate lower
 profits (though not lower profit margins)  with which to absorb notice filing
 costs and uncertainty.   (In Appendix C considerably more analysis of this
 aspect  is provided.)

     During the meeting  of industry experts,  a characteristic capital budgeting
model for R&D commercial development decisions in this size firm was
 presented.   It highlighted how the section 5 notice cost would affect
 decision-making on new  chemicals.   The model was an equation (shown below)
 which could be used to  produce a value (PN)  for each possible project under
 consideration by management.   The projects receiving the highest PN'S are the
 first to receive funding in any year.

-------
                                     - 99 -
    The equation was:

                  T x RC x RT x RpMN x  [V x  (Price - Cost RM) - LUD - TS]
      PN = FC  x
                                  CRES + CCD + CPL + CPMN
where:

PN       =  Project number.

FC       =  Volume factor.

T        =  Life cycle of products in years.

RC       =  Probability of commercial success  (0 to 1) .

RT       =  Probability of technical success (0 to 1)  .

RPMN     =  Probability of chemical not being questioned by EPA.

V        =  Sales volume in pounds.

Price    =  Price per pound.

Cost RM  =  Cost of raw materials per pound at annual sales volumes.

LUD      =  Labor, utilities, and depreciation at annual sales volumes.

TS       =  Cost of technical service necessary on sales.

CRES     =  Cost of research to develop chemical.

CQQ      =  cost of commercial development.

CPL      =  Cost of plant and equipment (if any) .

         =  Cost of filing notice.
     As the equation reflects, prior to TSCA, the key concerns were the
expected dollar volume, the probabilities of technical and commercial success,
the gross margin on sales, and the costs of development.  The equation would
not have had the Rp^m and CPMN entries.  The section 5 notice requirements
affect this equation by adding the additional uncertainty reflected as a
probability of clearing the EPA review (RPMN) and bY addition of the cost of
filing a notice (CPMN) .  Both of these effectively reduce the value of PN
thus making all investments less attractive.  (The magnitude of the reduction
is discussed in Appendix C.)

Small Companies

     The very smallest firms in the industry (those with less than $10 million
in annual sales) may feel these same two differences — lower expected profits
on new chemicals introduced, and lack of regulatory expertise — even more than

-------
                                     -  100  -


the medium size firms.  Their small managerial staff and legal capability will
cause their direct out-of-pocket and uncertainty costs per new chemical to  be
greater.

Summary

     According to the industry experts, the process of developing new
chemicals within a firm varies most directly with the size of the firm.
Furthermore, the direct costs of the notice program and the uncertainties
about release of confidential information and a possible EPA challenge of the
chemical will affect smaller firms more than larger firms.  As firm size
increases, the cost of the notice filing becomes less important than the
uncertainty about EPA's decision.  These factors will be felt most in
decisions to allocate funds to research and development, in choosing which
projects to fund, and in beginning commercial development.  In Appendix C a
financial analysis of the extremely limited, and possibly misleading,
available data is provided.  In the next section we discuss how the impact  on
individual firm innovation will affect the chemical industry.
IMPLICATIONS FOR THE CHEMICAL INDUSTRY

     The implications of the effects identified above vary by chemical
segment.  The segment-specific factors that will cause the impact to vary are

     •    innovativeness;

     •    structure of the segment;

     •    sales growth patterns;

     •    product mix (annual sales volume); and

     •    possible restrictive action.


Innovativeness

     More inventive firms are likely to introduce more new chemical compounds
than non-innovative firms.  As a result, they will be more severely impacted
by section 5 notice requirements.

Structure of the Segment

     The industry experts identified four categories of firms based on annual
sales volume.  These categories were described in Chapter 8 and may be briefly
characterized as follows

     •    $0 to $3 million—industry entrepreneurs;

     •    $3 to $10 million—successful entrepreneurs;

-------
                                     - 101 -
      •    $10 to $200 million—firms beginning to develop
           organizational divisions with special tasks;  and

      •    greater than $200 million—industry giants.

      As noted earlier, in regard to notice requirements,  the structure  of  the
 segment is particularly important in evaluating firms  in  the $10  to $200
 million category.  These firms often will not bring  to  market a new chemical
 with tremendous potential market value because they  fear  that the industry
 giants will enter the business and capture so large  share of the  market that
 the smaller firms'  investment will be lost.

      Additionally,  if a segment possesses a  large number  of  small competitors
 and a handful of large competitors who have  only  small  market shares, it is
 likely that the disproportionate impact of section 5 notice  costs on smaller
 companies  will result in increased concentration  in  the future.   This could
 hinder economic efficiency in the segment.

 Sales Growth Patterns

      Expectation concerning  rate of sales growth  and lifespan of  products  will
 directly affect evaluation of a new product.   Will it gain broad  enough
 acceptance and stay on the market long enough to  absorb the  additional  costs?
 Where expected sales volumes and lifespan are radically different—for
 example, organic dyes with small volumes and short lifespans versus inorganic
 acids with large volumes and long lifespans—the  ability  to  absorb the  costs
 will differ substantially.   Generally,  shorter lifespan or lower  expected
 growth products will be more significantly impacted.

 Product Mix (Annual Sales Volume)

      The product mix is measured in terms of annual  sales volume  of the
 individual products of individual firms.   There are  various  mixes — large
 number  of  small volume products,  a few high  volume products,  or some
 combination.   If the future  product mix is predicted to remain the  same as
 today's, the  ability of a firm  to adequately absorb  front-end costs is
 dependent  on  the volume and  profit margin of products sold.   Generally, firms
 which expect  to introduce many  small volume  chemicals may be more affected
 than  those which expect to introduce a  few high volume  chemical products.

      Product  mix varies across  all sectors.   It is more "general"  than  say,
 innovation, and  the  effects  of  product  mix will be more widely felt than any
 other  factor.   To assess  the  relative  impact on firms with different product
 mixes,  we  examined  individual company  data provided  in  response to  the
 reproposal  in  44 Federal  Register  59764.  Exhibit  9-1 gives  a representative
 sample  of  sales  distribution by  volume  as estimated  by  CSMA  and as  reported by
 four companies.  Clearly,  smaller  volume  products  dominate the product  mix,
 even  for the  larger  firms.   If we  assume  this  distribution is externally
 valid,  then on  the  average,  the  larger  firm's  process of  developing products,
 but their  not  financial  viability,  might  be  affected just as severely as
 smaller firms.   (Our  independent  chemical  industry experts questioned the
 external validity of  this  data  because  it did  not  include enough  large  volume
chemicals.)

-------
                                     -  102  -
     As a result of notice cost imposition, there may be fewer chemicals
introduced, and the "consolidation of product lines into high volume products"
strategy may be taken by many firms.  This will lead to a lessening of new
products and could lead to tiering, where the likelihood of introduction of
competing substances decreases and/or profit margins increase.  However, this
may or may not cause a lessening of overall sales growth and profitability for
the sector as a whole.
                                    EXHIBIT  9-1

                 SALES DISTRIBUTION BY POUNDS OF PRODUCT PER YEAR
                            (in percent of total  sales)
                                              Pound Volume
Annual Sales
Company ($ Millions)
Reilly Tar
Pennwalt
Nalco
Dow Corning
CSMA
50-100
480
500
500+
—
Under
IK Ibs. 1-10K Ibs.
75
34
5
7
(
8
18
15
15
70
Over
10-100K Ibs. 100-250K Ibs. 250K Ibs.
( 17 )
14.5 2.2 31.3
34 21 25
33 ( 45 )
) ( 30 )
Source:  Responses from individual firms to 44 FR 59764, October 16, 1979.
Possible Restrictive Action

     Firms will refrain from developing new chemical products in areas where
production and/or use of the chemical is currently subject to government
restriction.  And they may refrain where they think a product likely to be a
candidate for future regulation.   Chemical companies, particularly small firms
(less than $10 million), prefer to avoid operation in a highly regulated
segment due to time delays, costs of dealing with the government, their
inexperience with the government, and the uncertainty resulting from operation
in a regulated area.
SEGMENT ANALYSIS

     These segment-specific factors form the framework for discussing the
likely effect the section 5 notice requirements will have on the innovative
segments identified in Chapter 6—segments in which new product development is
expected to be of continued importance to success.

     These segments are

     •    Surfactants;

     •    Miscellaneous Synthetic Organic Chemicals, NEC;

-------
                                     - 103 -


     •    Cyclic  Intermediates;

     •    Catalysts;

     •    Rubber  Processing Chemicals;

     •    Plasticizers;

     •    Adhesives and Sealants; and

     •    Industrial Inorganic Chemicals, NEC; and

     •    Plastics and Resins.

In the subsequent paragraphs, we discuss what the likely impact on each
segment will be on the basis of the structure of the segment, the sales growth
pattern for new products in the segment, the product mix of the companies in
the segment, and  the likelihood that new chemicals in the segment would face
restrictive action.

Surfactants

     ICF found that the surfactant industry contains three different categories
of producer:  raw material suppliers, producers with large volume captive uses,
and producers selling to others.  The first two categories are dominated by
large companies.  Conoco, Monsanto, Union Carbide, Witco, Proctor & Gamble,
Lever Brothers, Colgate-Palmolive, Shell Chemical, Rohm & Haas, GAP Corpora-
tion, and Borg-Warner are the major producers.  The third category contains
two large producers (Diamond Shamrock and ICI) and a host of producers that
fall into the $10 to $200 million size category.4V  The industry experts
believe that this segment is characterized by firms in all categories with the
majority having total sales less than $3 million.

     Direct out-of-pocket costs imposed by section 5 notice regulations may
affect the surfactant industry.  Most new chemicals in this segment involve
marginal improvements on existing chemical products.  This is an area of
fairly active entrance and departure, and it is difficult to corner the
market.  It usually takes companies several years to recoup R&D costs.

     Currently, most surfactants are derived from linear alkylbenzene
sulfonates (LAS)  which is not a completely biodegradeable structure and is
banned in some areas.   The industry is seeking a new process to replace LAS
which is completely biodegradeable and is an equally effective compound.
Environmental regulations of phosphate levels in water also affect the
surfactant industry.   Both of these regulations are well established and the
surfactant industry has operated in a regulated environment for several
years.   Because the direction of regulation is clear, the surfactant industry
is less likely to be affected by future regulatory action than the other
innovative segments.
    ii/Meegan, Kline Guide, p. 166.

-------
                                    - 104 -
Catalysts

     The catalyst industry includes approximately 10 industry giants as well
as firms with operations less than $200 million.H/  Direct dollar
out-of-pocket costs associated with section 5 notice regulations are not
expected to pose a problem for this segment.   Catalyst firms tend to introduce
a few high volume products,   unlike most new chemical compounds, generally
when a new, improved catalyst is discovered it will penetrate the market
rapidly.  A new and superior catalyst will  command the market until a better
catalyst is discovered.

     Although direct costs should not pose  a problem,  uncertainty regarding
trade secret disclosure  is of great concern to the catalyst industry.
Catalyst formulations are probably the most highly guarded  trade secrets.
Section 5 notice regulations could potentially disrupt this industry if
information regarding new formulations were revealed as a result of the notice
process.

Cyclic Intermediates

     Cyclic intermediates are produced by firms in all sales categories above
$3 million.  The market  structure varies depending on the specific chemical
product.  Fundamental, high volume compounds may  have more  than 10 major
producers, whereas special low volume compounds usually have only one..IV

     Since new cyclic intermediate compounds differ only marginally from
existing compounds, it may take several years for a new compound to gain a
comfortable market share.  As a result, direct out-of-pocket costs due to
section 5 regulations will be important for this  segment.   Furthermore, the
large number of firms in the $10 to $200 million  category suggests that
section 5 notice regulations may adversely  affect the ability of these firms
to gain the necessary market share required to recoup their initial investment.

Miscellaneous Synthetic  Organic Chemicals,  NEC

     The miscellaneous synthetic organic chemical industry  encompasses more
than 1,000 firms, far more than any of the  other  innovative segments.   Most  of
the firms in this segment have operations less than $10 million.  Thus,
section 5 notice regulations will most severely affect smaller firms through
its impact on this segment.   The extent of  competition varies by chemical, and
the leading manufacturer often differs from chemical to chemical.

     Direct out-of-pocket costs imposed by  section 5 notice regulations will
be an important factor for this segment.  The segment is extremely active,
    12/ibid.,  p.  147.

    Af/U.S.  International Trade Commission,  Synthetic  Organic  Chemicals,
1977,  (Washington,  D.C.:   Government Printing  Office).

-------
                                     - 105 -
with a great deal of product entrance and departure.  New products are usually
distinguished by only marginal improvements over existing products.  Thus, it
is difficult to corner the market.  It takes several years to recoup R&D
costs, despite the fact that start-up costs tend to be low relative to other
segments.  Because of the diverse nature of this segment, more large volume
products are introduced here than in other segments.  The firms introducing
these large volume products will be affected less by section 5 notice
regulations than the other firms.  Overall, small-volume producers in this
segment will be significantly disadvantaged by the section 5 rules.

Industrial Inorganic Chemicals, NEC

     The industrial inorganic chemical, NEC segment is diverse, much like the
miscellaneous synthetic organic chemicals, NEC segment.  Although not as large
a segment as the miscellaneous synthetic organics, the industry structure is
quite similar, with the majority of firms having operations of less than $10
million.Al/  The extent of competition varies by chemical, as does the
potential effect of section 5 notice regulations.

     Direct and out-of-pocket costs imposed by regulations will be important
factors for the segment.  Most new chemicals in this segment involve marginal
improvements to existing chemical products.  This is an area of fairly active
entrance and departure, and it is difficult for firms to corner the market.
Firms require several years of sales to recoup R&D costs for a new product.

Rubber Processing Chemicals

     The rubber processing chemical industry is fairly concentrated, with
Goodyear, Goodrich, and Uniroyal responsible for approximately 50 percent of
total production.  Most firms have operations over $10 million.  Some $3 to
$10 million firms are active, but no entrepreneurs are found in this
segment..!5/  The advantage that generally results from a high degree of
market concentration is minimized for the rubber processing chemical companies
because their purchasers, synthetic rubber producers, are also highly
concentrated.

     Direct out-of-pocket costs imposed by section 5 notice regulations will
affect the rubber processing industry but to a lesser degree than the
miscellaneous synthetic organic chemicals, NEC segment, since fewer small
firms produce rubber processing chemicals.  Most new rubber processing
chemicals are developed to satisfy a specific design feature that is desired
by a rubber manufacturer.  They are products that are marginal improvements on
existing compounds.  It takes several years for rubber processing chemical
companies to recoup their R&D costs, and these products may have short
lifespans.46/
    Al/From meeting between ICF and industry experts, June 3, 1980.

    ijj/Meegan, Kline Guide.

    £6/From meeting between ICF and industry experts, June 3, 1980.

-------
                                    - 106 -
Plasticizers

     The plasticizer industry is very similar to the rubber processing
industry, but it has more companies and is not as highly concentrated.  It
includes a few entrepreneurial firms, but the majority are spread evenly among
the remaining categories.

     Direct out-of-pocket costs may be a problem for this segment.  Many small
volume sales are involved, and new plasticizers tend to consist of marginal
improvements on existing compounds to meet specific properties desired by
plastics manufacturers.

Adhesives and Sealants

     Adhesives and sealants is a diverse industry that includes firms in all
sales categories but particularly those with $10 to $200 million
operations.AZ/  New product development in this segment is oriented towards
the introduction of many small volume chemicals.  R&D is generally in response
to a desire for an adhesive or sealant with slightly different properties than
currently available, and so a new product is often similar to an existing
chemical.  Thus, the product may require several years to obtain a share of
the market and to provide a return on the initial investment.  This slow
growth suggests that direct out-of-pocket costs imposed by section 5 notice
regulations will influence the adhesive and sealant segment.

Plastic Materials and Resins

     Basic plastic materials are produced in large volumes.  As a result, the
industry is dominated by large companies with entrepreneurial firms virtually
eliminated from the market.

     Direct out-of-pocket costs associated with section 5 notification
requirements may be an important factor for the segment.  Many small volume
chemicals are constantly introduced as basic plastics are modified to meet the
specific requirements of an endless array of applications.  Since most new
chemicals differ only marginally from existing compounds, it is difficult for
them to capture a major  market share.
SUMMARY

     This examination of the nine innovative segments of the chemical industry
has reemphasized the importance of the costs identified in Chapters 4 and 5.

Direct Out-of-Pocket Costs

     Section 5 notice requirements will impose direct costs which will affect
new product introduction in all innovative segments, except for the catalysts
    -LZ/Meegan, Kline Guide,  p.  134.

-------
                                     -  107  -
industry, because many small volume chemicals are involved which often provide
only marginal improvements to existing chemicals and which require several
years to recoup their R&D costs.  New catalysts tend to be a few large volume
chemicals with extremely rapid market penetration, thus direct out-of-pocket
costs imposed by section 5 notice regulations should not adversely impact the
catalyst industry.

Trade Secret Disclosure

     Uncertainty and additional costs associated with possible trade secret
disclosure could have a major impact on the catalyst industry, but does not
appear to be a major problem for the other segments.  Of course, small volume-
small company producers may be tracked by large companies through the Federal
Register which may lead to trade secret disclosure in all segments.

Delay in the Introduction of New Chemicals

     Uncertainty and additional costs associated with delays in the
introduction of new chemicals may have a slightly different effect on each
innovative segment.  Segments with a higher proportion of small firms—
surfactants; miscellaneous synthetic organic chemicals, NEC; cyclic
intermediates; and industrial inorganic chemicals, NEC—may have more
difficulty bearing the additional costs resulting from delays in the
introduction of a new chemical.
MACROECONOMIC EFFECTS

     Macroeconomic effects of the section 5 notice regulations derive from the
microeconomic effects discussed above.  The manifestations of the effects will
appear in changes in the rate of growth of the economy, changes in the
industry concentration, and changes in the balance of trade.  In the
subsequent paragraphs, each of these is discussed.

Rate of Growth

     The rate of growth of the economy will be affected by the segment changes
analyzed above.  The effects will be both short run and long run.  The
regulations will increase the economic costs of producing new chemicals.  This
will initially cause fewer new chemicals to be introduced into commerce,
resulting in a potential for decreased growth.

     To the extent that new chemicals increase efficiency and enhance
productivity, the rate of growth of the economy initially will be further
lessened by secondary impacts.  It is highly probable that this will be  the
case because several of the affected segments  (miscellaneous synthetic organic
chemicals, NEC; industrial inorganic chemicals, NEC; catalysts; and cyclic
intermediates) make products used in the manufacture of other products.
Denied new, less costly raw materials, secondary products will not develop as
rapidly as they have in the past.

-------
                                     - 108 -
     Another way to understand the impact on the rate of growth of the economy
is in terms of the theoretical models.  Economic growth according to most
theorists is a function of the level of investment.  New chemical
introductions result from investment in research and development activities.
We can predict that the decreased investment connected with a drop in the rate
of new chemical introductions will cause a drop in the rate of general
economic growth.

     In the longer run, it would be expected that the decreased number of new
chemicals introduced will cause the relative costs of the factors of
production in chemical-consuming industries to shift.  In some cases the shift
will be to non-chemical alternatives.  It should be expected that this will
spur investment in these industries which will offset to some extent the
decreased investment in the chemical sector.

     Because data do not allow for ex ante predictions about the number of new
chemicals not introduced after the regulations become final, the short-run and
long-run effects cannot be measured quantitatively.A§/

Industry Concentration

     The second general area of effects will be in industry concentration.  As
we explained in Chapter 8 and earlier in this chapter, the affected segments
will generally become more concentrated.  Greater concentration will occur as
regulation prevents the growth and perhaps even the viability of smaller firms.

     As explained previously, those firms most affected by the rules will be
those with less than $200 million annual sales.  The higher costs of bringing
a new chemical to market affect the entrepreneurs (the less than $3 million
annual sales companies) in two ways.  First, the direct and uncertainty costs
associated with section 5 notices will cause some relatively small number of
these firms to fail.  Second, the recognition by the capital markets that
section 5 rules reduce potential returns will result in less capital available
to these companies.  Nevertheless, Because entrepreneurs typically are
motivated by factors of which cost considerations are only a small part, it is
doubtful that section 5 regulations will greatly reduce the level of
entrepreneurial activity.
     A§/Some regulation experts today argue that the above logic is based on
the false assumption that the rate of new chemical introductions will drop.
They argue that the rate will not drop because companies will redirect their
R&D to chemicals more likely to do less harm to the environment.  Although ICF
concurs that there will be this shift, ICF believes that these experts fail to
recognize a critical fact—the new chemical that would have provided a lower
cost, or more efficient input into the economy is an opportunity for economic
growth foregone.  Furthermore, at the present time the greatest magnitude of
innovation occurs in those parts of the industry where the greatest reward can
be obtained from innovation.  Thus, even if the rate of new chemical
introductions does not drop but shifts among kinds of chemicals, there will
still be a lessening of economic growth due to these foregone opportunities.

-------
                                     - 109 -
     For  the  $3 million to $200 million annual sales companies, the effects on
 industry  concentration will be greater.  Because these companies are more
 professionally managed than the entrepreneurs, they are more risk averse.
 Risk aversion implies the necessity to weigh uncertainties highly before
 committing funds to a project.  These firms face greater uncertainty
 concerning the new regulations due to their lesser access to expertise about
 government regulations, and their product mix and research efforts usually are
 not very  broad.  We, therefore, can expect these firms to introduce fewer
 chemicals and thereby reduce their share of market in the future.

     The  recipients of this share of market will probably be the major
 chemical  companies (the over $200 million annual sales).  These firms have
 greater access to expertise about government regulation, broader (in terms of
 types of  chemicals) research and development programs, and greater access to
 capital.  Thus, they should prosper in comparison to their smaller competitors
 under the regulations.

 Balance of Trade Effects

     The  effects of section 5 notice requirements on the U.S. balance of trade
 depends considerably on the relationship between the U.S. regulatory burden
 and the regulatory burden in other major chemical-producing countries.  At one
 extreme,  if the regulations applied to toxic chemicals in the United States
 are the same  as in the other major-producing countries,  one would expect no
 changes in the balance of trade in chemicals.  If the regulatory burden were
 greater in the U.S., one would expect a shift in the balance of trade in
 chemicals away from the U.S.

     Most foreign notification programs (with the exception of Japan)  are
 pre-market rather than pre-manufacture programs.  Further, notification under
 the European  Directive is not due to take effect until March 1982.
 Nevertheless, in most cases, the information requirements are or will be
 similar.  In  fact, the testing requirements in foreign countries are much
 greater than  in the U.S., so that useful information about new chemicals can
 be expected from the Foreign Manufacturers/Suppliers Form.  On the other hand,
 small volume  chemicals (under one ton annual production)  do not require
 notification  under EEC guidelines.  (For further information see the importer
 contact issue paper.)

     If the burden is greater in the U.S.  (a questionable assumption at this
 time), some innovations which would have occurred in the U.S. will not.
 Sooner or later these innovations will occur in other countries and be adopted
 overseas.  To the extent that use of these chemicals in other countries
 displaces imports from the U.S., the U.S.  balance of trade in chemicals will
 suffer.   In addition,  imports of these chemicals into the U.S.  may displace
 domestic production.

     Although it is true that importers of new chemicals also face a U.S.
 regulatory burden, the burden per successful innovation will be less for
 imports than domestically produced chemicals.   The reason for this is that
notices must be filed for all new domestic chemicals including many which will
ultimately be unsuccessful;  but many of the unsuccessful chemicals developed
abroad will have failed before any U.S.  regulatory burden is imposed.

-------
                                     -  110  -
     A precise quantitative estimate of the impact of notice requirements on
the balance of trade cannot be made,  uncertainties about chemical regulations
in other countries, not to mention uncertainties about the impact of section 5
on U.S. chemical innovation, would make any such estimate of little value.
CONCLUSION

     ICF concludes from the foregoing analysis that the rate of introduction
of new chemicals will decrease in the short-term because of the imposition of
the section 5 notice.  Much of the decrease will occur in the highly toxid
chemicals of wide human and environmental exposure.

     Some of the decrease will occur in other chemicals, however,  not because
EPA takes restrictive actions, but because of uncertainty.  This uncertainty
will stem from the very existence of TSCA, not just section 5.  Chemical
companies will not know and cannot be expected to know exactly how EPA will
regulate chemical substances under section 5.  Without extensive legal and
government expertise — something many chemical companies cannot afford —
uncertainty about EPA's actions concerning new chemicals will be a significant
obstacle in the new chemical introduction process.

     In an uncertain environment product managers act to minimize  uncer-
tainty.  For many smaller chemical companies this could mean decreased R&D and
decreased new chemical entity development.  To the extent that these companies
are in segments where new products are important for corporate survival,
marginal small companies may be very vulnerable to failure.   For other small
companies this will mean a shift of R&D into what it perceives as  being "safe"
chemicals.

     For larger companies the expected response will generally be  a shift to
"safe" chemicals.  These companies, with their better access to information
about EPA behavior, should be less uncertain than the smaller companies.
However, until the EPA has developed a pattern of response to section 5
notices, these companies can be expected to behave much like their smaller
competitors.  Furthermore, the transition from R&D and new chemical
development in the current fields to the "safe" fields will not be without
cost.

     As this report makes clear,  the costs of the section 5 notice regulation
are not quantitatively measurable.  We have, instead, identified the most
vulnerable segments, the kinds of costs they will experience, and  the relation
of these costs to alternative regulations under consideration.  We have also
discussed the ways in which chemical firms will be affected and the relative
economic impact based on economic factors important to the success and failure
of these firms.  So, without quantifying, we have produced an economic impact
analysis of the effects of section 5 notice costs.

-------
    -111-
  PART  II




ISSUE PAPERS

-------
                                      -113-
                                  INTRODUCTION


     EPA asked ICF to analyze their regulatory alternatives and the
consequences of those alternatives for nine aspects of section 5 notice
requirements.  Those nine aspects are:

     1.   Confidentiality

     2.   Customer Contact

     3.   Definition of Importers

     4.   Importer Contact of Foreign Manufacturers/Suppliers

     5.   Exporters

     6.   Supplemental Reporting

     7.   Insufficient Submissions

     8.   Processor Reporting

     9.   Possession or Control

     For each of these aspects, specific options available to EPA were
provided.  An evaluation framework was then developed to allow analysis of
each option relative to other options.  This evaluation framework was
quantified where possible.  Data for the analyses came from:  public comments
on EPA proposed section 5 notice requirements, hearing records, previous
economic analyses, and ICF chemical segment analyses.

     The analyses of these nine aspects of the section 5 notice requirements
were used to generate the Comprehensive Program Options analysis in Chapter 5
of Part I of this report.

-------
                                       -115-


                                    CHAPTER 1

                                 CONFIDENTIALITY
      Valid  but conflicting  interests  surround  the  issue and  treatment of
 confidentiality under  section  5  notice  information.  On the  one hand,
 industry's  desire  to maintain  the  confidentiality  of business and trade
 secrets  is  based on  the  competitive advantage  it derives  from such secrecy.
 On  the other  hand, the free flow of information about chemicals should help
 obviate  harmful exposures and  possible  abuses.  To accommodate these
 interests,  EPA must  establish  a  mechanism  to define a legitimate
 confidentiality claim.

      The confidentiality provision of section  5 imposes some potentially heavy
 economic burdens on  the  chemical industry.  The following discussion
 describes,  first,  the  confidentiality options  that EPA is now considering.
 These options have four  elements:

      •    the method of  asserting  and substantiating confidentiality
          claims;

      •    the timing of  substantiation;

      •    the requirements  for generic masking; and

      •    the disclosure of chemical  identity  as part of a health
          and safety study.

      A subsequent discussion analyzes the  relative economic  impact of the
 various  options.

      Finally,  the objective of the confidentality  policy and the specification
 of  the options are studied.

 SPECIFICATION OF EPA OPTIONS

      EPA has  specified a set of  options for the several subissues which must
 be  addressed  in establishing an  approach for the treatment of confidential
 business  information.  These alternatives  are discussed in the following order:

      •    Assertion  and Substantiation Options;

      •    Generic Masking Options; and

      •    Options for Disclosing Chemical  Identity as Part of a
          Health and Safety  Study.

Assertion and  Substantiation Options

      Four options for asserting  and substantiating a claim of confidentiality
were  evaluated  (see  Exhibit  1-1).  Each option includes:

-------
                                     -  116  -
     •    a method of asserting and substantiating a claim of
          confidentiality when the notice is submitted; and

     •    a time frame for substantiation.

There are three situations in which substantiation would be required:!./

     •    when EPA receives a request under the Freedom of
          Information Act (FOIA)  for information that has been
          claimed confidential;

     •    if EPA independently decides that it is "likely" that a
          request to disclose the information will "eventually" be
          received; and

     •    when EPA independently decides to determine if information
          is confidential, regardless of whether a request for the
          information has been received.

     Option 1;  No Substantiation with Submission.  Items deemed to be
confidential business information would have to be clearly designated either
by a checkmark, red circle, or some other indicator of the submitter's
choice.  To substantiate the claim, the submitter would then respond to
questions prepared by EPA and contained in a letter to the submitter.  These
questions would concern the nature and type of confidentiality claim..£/
Submittal of the notice and substantiation need not be concurrent.

     Option 2;  Some Substantiation with Submission.  The January 10
proposal^/ requires that items asserted to be confidential be marked
item-by-item.  The same substantiation questions as Option 1 are required  for
the same areas of confidentiality.  However, the timing of substantiation  is
different.  Instead of substantiating everything at a later date,
confidentiality assertions for both chemical identity and information in a
health and safety study must be substantiated when the notice is submitted.

     Option 3;  Considerable Substantiation with Submission.  This option  is
really option 2 with the additional requirement that use data which is claimed
to be confidential must also be substantiated when the notice is submitted.
     i/These situations are specified in EPA regulations, 40 CFR Section
2-204(a).

     2/These categories and the number of questions under each  are:
chemical identity (nine questions); other information in health and  safety
study (eleven questions); manufacturer's identity  (four questions);  production
volume (six questions); use data  (nine questions); physical and chemical  data
(thirteen questions); and miscellaneous  (eight questions).

     I/The proposal is at 44 Federal Register 2242, e_t seq.

-------
                                                     EXHIBIT 1-1

                                             ASSERTING AND SUBSTANTIATING
                          Option 1

                             No
                       Substantiation
                       With Submission
                              Option 2

                                Some
                           Substantiation
                           With Submission
                             Option 3

                           Considerable
                          Substantiation
                          With Submission
                              Option 4

                              Complete
                            Substantiation
                            With Submission
Assertion
Method
   Assertion
   Assertion
   Item-by-Itern
»  Assertion
 - Item-Dy-Item
Assertion
Linkage and Categories
When
   Not EPA prescribed
   Substantiation
   only upon EPA
   request
   Substantiation
   with submission
   Chem ID
   H&SS data
   Remaining confi-
   dentiality claims
   substantiated upon
   EPA request
   Substantiation
   with submission
   Chem ID
   H&SS data
   Categories of use
   Remaining confi-
   dentiality claims
   substantiated
   upon request
All Confidentiality
claims substantiated
with submission
Method of
Substantiation
Series of questions
Series of questions
Series of questions    Limited questions

-------
                                     -118-


This procedure would let EPA know whether or not the information could be
published in the section 5(d)(2) notice.

     Option 4;  Complete Substantiation with Submission.!./  A claim of
confidentiality is asserted by categorizing information under one of seven
general groupings:

     •    Category A:  Manufacturer's Identity;

     •    Category B:  Specific Chemical Identity:

     •    Category C:  Production Volume;

     •    Category D:  use Data;

     •    Category E:  Process Information;

     •    Category M:  Proportions of a Mixture (only for use in
          health and safety study); and

     •    Category F:  Other Information.

     Boxes are provided beside each item of information, and the appropriate
letters are placed in the box to indicate the confidentiality category in
which the information falls.  Some information is automatically "linked" to or
included in a category, and nothing needs to be done beyond indicating that
the information is claimed confidential.Jj/  For other information, the
"linkage" is not automatic and must be established by explaining how
disclosure of the information would reveal confidential information about the
particular category.

     Substantiation of the confidentiality claims occurs as follows:Ji/

     •    There is a confidentiality certification which is to be
          signed by the appropriate individual.  This individual
          certifies that:

               The submitter has taken and will continue to take
               measures to protect the confidentiality of the
               information;
     i/The reproposal is at 44 Federal Register 59764 ejt seq.

     JL/A11 items which are automatically linked can be claimed confidential
by checking only one box—i.e., the claim does not have to be asserted
individually for each item of information.

     .6/The list of substantiation questions under this option is in a latter
section of this report.

-------
                                     -  119  -
               The information claimed confidential has generally
               not been "reasonably obtainable" by outsiders;

          —   The information is not publicly available elsewhere;
               and

               Disclosure would cause "substantial harm" to the
               submitter's competitive position.

     •    Manufacturer's identity is substantiated by signing the
          certification statement.  There are no other questions to
          be answered.

     •    Specific chemical identity is substantiated by answering a
          series of eight questions.

     •    For production volume, use data, process information, and
          proportions of a mixture, one substantiation question must
          be answered if manufacturer's identity has been claimed
          confidential and the submitter is asserting additional
          confidentiality claims.  For example, the submitter must
          explain why and how disclosure of the particular
          information would hurt its competitive position.  The same
          procedure is followed when chemical identity has been
          claimed confidential.  If neither manufacturer's identity
          nor chemical identity has been claimed confidential, and
          the submitter asserts additional confidentiality claims,
          no substantiation questions need be answered for the four
          categories mentioned above.

     •    For information which does not fall into one of the six
          specific categories, the submitter must answer a series of
          six questions to substantiate the confidentiality
          assertion.

     The substantiation questions must be answered only once, regardless of
how many times information is claimed confidential under a particular
category, except for "other information."  For that category, the six
questions must be answered each time information is claimed.  All
substantiation must be submitted with the notice.

Generic Masking Options

     The second sub-issue of confidentiality is the development of generic
information.  In some instances when asserting confidentiality, the submitter
would be required to propose a generic name or generic information to  replace
the confidential information.  The three options for substantiating generic
information differ in the amount of generic information they require as shown
in Exhibit 1-2.

     Option 1;  Some Generic Information. . The January 10 proposal required
the submitter to provide some generic information:

-------
                                      -120-
     •     If chemical  identity is claimed confidential, a generic
           name must  be provided.  The submitter is encouraged to
           seek prior EPA approval of the generic name before
           submitting it.  The generic name is to be  "only as generic
           as necessary" to protect the confidential  chemical
           identity,  and "should reveal to the maximum extent
           possible toxicologically significant aspects of the
           molecular  structure".  The submitter also  is encouraged to
           explain why  additional specificity would reveal
           confidential business information.  An appendix to the
           January 10 proposal gives guidance for creating the
           generic name.  It is possible that the submitter will have
           to develop more than one generic name before obtaining EPA
           approval.

     •     If use data  is claimed confidential, the submitter is
           required to  provide a generic use description supplemented
           with the likely exposures to humans or to  the environment.

     Option 2;  Considerable Generic Information.  This option has a generic
 scheme  encompassing  four classes of information:

     •     If chemical  identity is claimed confidential, the
           submitter  is to provide EPA with three generic names if_ a
           prior agreement (arrived at during pre-notice
           communication) has not been made with EPA  regarding an
           appropriate  generic substitution.2/

     •     If use data  is claimed confidential, the submitter is
           expected to  provide a generic use descriptors developed in
           accordance with a scheme in which the submitter picks a
           discriptor generic in each of six categories of use
           characteristics.

               degree  of containment;
               level of environmental release;
               type  of population exposed;
               type  of environmental release;
               type  of human contact; and
               average frequency of human contact.
     2/These names are to be developed in accordance with the instructions
in Appendix II of the January 10 proposal.  If none of the three names are
satisfactory to EPA, EPA will propose a generic name.  If EPA's generic name
is not satisfactory to the submitter, the submitter will submit a fourth
generic name and explain why EPA's candidate would reveal CBI.   If the
submitter's fourth term is unacceptable, EPA will publish its choice.

-------
                                       -121-
           If  the  submitter  does  not  use  this  scheme,  it must explain
           why use of the  scheme  would  reveal  CBI  and  provide an
           alternative generic  name.  If  the submitter fails to
           provide a generic name,  or provides a generic name that is
           too generic,  EPA  will  develop  and disclose  generic
           information.

      •    If  disclosure of  the physical  and chemical  properties of
           the new chemical  substance would reveal CBI, the submitter
           can use EPA-supplied ranges  to report the value of these
           properties.   The  ranges  are  provided for five properties:

                vapor pressure;
                density;
                solubility;
           —    melting  point;  and
                boiling  point/sublimation point.

           If  the  ranges supplied by  EPA  are not used,  the submitter
           must explain  why  ranges  would  reveal CBI.   The submitter
           would have to provide  alternative generic information for
           these properties.  If  other  chemical and physical
           properties are  claimed confidential, submitters would
           supply  their  own  ranges.

      •    if  manufacturer's identity is  claimed confidential, the
           submitter must  provide a generic description incorporating
           three categories  of  characteristics:

                general  geographic  location of the company;
                size of  the  company,  in total annual sales; and
           —    type of  company by  Standard Industrial Classification
                Code (three-digit).

           EPA has provided  a list  of these characteristics.  If the
           three-digit SIC reveals  the manufacturer's  identity when
           combined with its  two  other  characteristics, the submitter
          may provide the two-digit  SIC  with an explanation of why
           the three-digit code is  too  specific.

      For any  of these generic  requirements, the submitter may explain why the
generic information cannot  be developed.  EPA will decide whether this is
sufficient justification  for not providing generic information.

      Option 3;  No Generic  Information.  A final alternative is to not require
generic information.  When  information is confidential, no generic information
would  be provided  to the  public.

Options for Disclosing Chemical  Identity as Part of a Health and Safety Study

      If a health  and safety study  is submitted two subissues must be decided:

-------
                                     -122-
     •    when the chemical identity will be disclosed; and

     •    how the chemical identity should be added to the inventory.

     Timing Options.   Disclosure of chemical identity, as a concernittant of a
health and safety study, might take place at any of the following moments:

     •    when the notice is submitted;

     •    when manufacture begins;  and

     •    when the new chemical substance is distributed in commerce.

     Inventory Options

     There are two forms in which the chemical identity can be added to the
inventory:

          •    by generic name; or

          •    by specific chemical name.



                                  EXHIBIT 1-2

                                GENERIC OPTIONS
     Option 1                   Option 2                          Option 3
   Sane generic               Considerable                      No Generic
   Information             Generic Information                 Information
   Provide generic         •  Provide generic              •  No generic
   information for:            information for:                 information
                                                              must be submitted
   (1)  Chem.  ID               (1)  Chem.  ID
   (2)  Chem Use               (2)  Chem Use
                              (3)  Manf.  ID
                              (4)  Chem & Phy.
                                  Properties

-------
                                      -123-
ANALYSIS OF  ISSUES

     This section analyzes  the  issues  in  terms of the costs they impose on the
the  submitter of the notice.  A general dicussion of the kinds of costs
involved precedes the analysis  of how  costs vary by option.  The costs
attributed to the various options were derived from sources including industry
comments on  the proposal and reproposal,  EPA comments on the proposal and
reproposal,  and ADL estimates derived  for the reproposal..§/
Kinds of Costs

     Exhibit 1-3 lists the three of kinds of costs that are imposed directly
on submitters.  In the subsequent paragraphs each is addressed.
                                   EXHIBIT 1-3

                              CONFIDENTIALITY COSTS


Submitter;  Cost 1.  Disclosure of trade secrets due to the amount and nature
                     of notice information made public.

            Cost 2.  Out of pocket expenditures due to procedural and
                     administrative requirements.

            Cost 3.  Uncertainty about what kinds of information EPA will
                     require and how much of that information might be
                     disclosed.
     ^/Comments of the Manufacturing Chemists Association on EPA Proposed
Regulations for Premanufacture Notification Under Section 5 of TSCA
(hereafter:  MCA proposal comments.  Note;  MCA subsequently become CMA),
March 26, 1979; Comments of the Chemical Manufacturers Association on EPA's
Reproposal of Forms and Rules for the Submission of Premanufacture Notices
Under Section 5 of the Toxic Substances Control Act (hereafter:  CMA
reproposal comments),  November 30, 1979; Other industry comments including
those of the Synthetic Organic Chemical Specialities Manufacturers Association
(CSMA);  Comments by Environment Defense Fund (EOF), Environmental Action, and
National Resources Defense Council (NRDC); EPA's Interim Policy (44 Federal
Register 28564-28572); EPA's Reproposal  (44 Federal Register 59764-59783);
Industry Interviews by Arthur D. Little  (ADL Interviews); and Arthur D.
Little,  Inc., Estimated Costs for Preparation and Submission of Reproposed
Premanufacture Notice Form, (Cambridge, Mass.:  Arthur D. Little, September
1979)—hereafter, ADL Study.

-------
                                      -124-
     Cost 1;  Disclosure of Trade Secrets.  The confidentiality policy will
affect the probability of disclosing trade secrets.  Trade secrets are a very
important factor in the chemical industry and are sometimes critical to the
new chemical introduction process.  This is particularly true as the demand
for many chemicals is very price-elastic. Because a multitude of substitutes
may exist, there may be little difference between the successful new chemical
and the new chemical which fails.  Every competitive advantage, therefore,
becomes significant to the individual firm.   The existence of a trade secret
in an individual firm is a source of competitive advantage.

     Trade secrets are not limited only to the chemical identity of a new
chemical.  The process by which an intermediate chemical is manufactured could
be far more significant than its identity.  A manufacturer's identity and
location may reveal information about the potential market for the new
chemical to a critical competitor.  Therefore, the information required in the
notice may include items that the submitter would not otherwise reveal.

     The distinction between what will and will not lead to the disclosure (if
not directly, then indirectly)  of CBI may frequently be a point of contention
between EPA and the submitter;  the cost of potentially reduced or destroyed
competitive advantage must be included in the evaluation of confidentiality
issues and options.

     Cost 2;  Out of Pocket Expenditures.  These are the procedural or
administrative costs which the  submitter must absorb in complying with the
confidentiality requirements of section 5.  The largest component is the
personnel required to complete  forms (i.e., answering the substantiation
questions)  and to furnish information (i.e., generic information)  required by
EPA's confidentiality procedures.  These costs include the actual time
required to fill out the forms  (i.e., the purely procedural tasks), time spent
in consultation with EPA representatives, and time spent on internal
consultation.

     Cost 3;  uncertainty About EPA Decisions.  The submitter cannot ignore
the potentially adverse situations which may arise from an unfavorable (in the
submitter's view)  determination by EPA concerning the confidentiality claims
asserted.  Therefore, preparations may have to be made for each such situation
or at least for the most likely situations.  Such preparations divert some of
the submitter's resources that  could be used for other purposes if this
contingency planning or preparation were not required.

Cost Associated with Options on How to Assert and Substantiate

     Previously, we identified  the four options under consideration for
asserting and substantiating confidentiality claims.  The four options repre-
sented combinations of two sub-issues:  (1) method and (2)  timing.  First we

-------
                                     -  125  -
discuss the relative costs of the two "method" options.  They are summarized
in Exhibit 1-4, following the discussion below.

Specification of Options

     The two options EPA could use as the basis for its assertion and
substantiation requirements were mentioned previously as being "item-by-item"
or "categories and linkages."  They are discussed in greater detail here.

     Option 1.  Assert item by item and substantiate by responding to a series
of questions.  The assertion is straightforward; it probably would require
that the submitter check a box (printed on the PMN form) beside information
that is claimed to be confidential and mark clearly any confidential material
contained in the notice as an attachment.  The series of substantiation
questions would either be a part of the form  (or the instructions for such a
form) or would be in a form letter addressed to the submitter.  HOW the
questions are presented depends upon the timing of substantiation.

     Option 2.  Assert by categories and linkages; substantiate by making a
general certification  (which would cover manufacturer's identity),I/
answering a series of questions for specific chemical identity and "other"
information and answering no more than two questions each for use data,
process, mixture, and production information.iO/

Relative Costs of Options 1 and 2

     The relative costs of these two options are discussed below.  Exhibit 1-4
summarizes these costs.

     Both options entail similar probabilities of disclosing the  submitter's
trade secrets.  Under option 2 the submitter may feel that it is  incurring a
higher risk of revealing CBI because it has to provide  information about the
linkage of CBI to specific categories as well as substantiate its claim of
confidentiality for that information,  in fact, the two options do not require
the  submission of significantly different information and do not  involve costs
     I/This certification would  also cover confidentiality claims  for  use
data, production  volume, and process information  if_ neither manufacturer's
identity nor chemical  identity have been claimed  confidential.

     IP/These  questions must be  answered only  if  manufacturer's  identity or
chemical identity is claimed and held confidential.

-------
                                    - 126 -
that can be attributed to either option.   All material,  including substan-
tiation and linkage material,  submitted under either option is afforded the
same protection if CBI is involved.
     Each option presents different administrative requirements for the
submitter which in total are similar .jj/  The substance of what the
submitter is required to submit to EPA does not differ significantly between
the two options.  However, option 2 imposes greater assertion requirements and
lesser substantiation requirements.  The assertion requirements are greater
because the linkage and category approach requires more attention and effort
than the item-by-item checkoff.  However, the magnitude of the burden is
minimized because the linkage questions only require a narrative explanation
of how disclosure of the information would reveal confidential information
about a particular category. 13/  industry comments have indicated a
     ii/EPA has indicated that the substantiation itself, can be claimed
confidential.   (interim policy 44 Federal Register 28568, May 15, 1979) .
industry's concern is that adequate protection is not provided for the linkage
and substantiation material.  (See CMA reproposal comments, pp. 119-120, 170.
At p. 119, another threat to CBi is suggested:  "... the question format
proposed by EPA will be burdensome for submitters, and its complexity and
attendant expense may deter companies from asserting legitimate trade secret
claims."  Failure to assert legitimate trade secret claims may be due to
factors other than just substantiation requirements.  Because this failure
cannot be isolated as due solely to TSCA regulations, it is not included as a
cost in the discussion.)
           the ADL study, the range of time (in essence, the out-of-pocket
cost) required for asserting confidentiality under the reproposal was esti-
mated to be 2 to 24 hours (p. 52) and the range for developing the linkage and
substantiation material was estimated to be 12 to 100 hours  (p. 52) .  The
study also noted that, in comparing the proposal's and reproposal' s substan-
tiation requirements, it was "reasonable to assume that greater company-by-
company variations in effort expended on the second step in  the confidential-
ity process — developing substantiations — would occur in the  absence of EPA's
current (i.e., the reproposal)  confidentiality provisions.   Under these
conditions, the range of time required could conceivably be  much broader than
the range estimated above."  (p. 51, note).  The wider range  would exist
because the submitter's assertion and substantiation under the proposal would
be less structured than under the reproposal.

     i^/ Industry has claimed that the linkage system is "complex and confus-
ing" and that it "will require a considerable expenditure of time and effort
by PMN submitters. ..."  CMA reproposal comments, p. 99.   The industry also
has indicated a preference for the proposal's checkoff method of assertion
(MCA  (subsequently CMA) comments on the proposal, p. 299; and CMA comments on
the reproposal, p. 101) .  The ADL study estimated that actual form preparation
would take two to 16 hours, and review of the completed form would take two to
20 hours.  The total time required by the reproposal1 s confidentiality process
was estimated to be 18 to 160 hours  (p. 52) .  NO time estimates were made
based on the proposal's requirements.

-------
                                       -127-
preference for narrative responses instead of specific responses to a series
of questions.il/  The substantiation requirements are significantly less for
option  2 than option 1.15/ because option 2 has only two long series of
questions  (for specific chemical identity and for "other" information) instead
of several series associated with option 1; four categories have no more than
two questions to be answered, and manufacturer's identity does not involve any
substantiation questions.  Thus, the additional burden of asserting a
confidentiality claim under option 2 is counterbalanced by the reduced burden
of substantiating the claim under option 2.

     Although the substance requirements are similar, option 2 may involve
greater out of pocket costs.  There are two reasons for these increased costs:

     •    Option 2 requires that the submitter explain its
          confidentiality claims by providing information with two
          different orientations—establishing links between
          information and confidentiality categories and
          substantiating confidentiality claims.  The submitter's
          efforts to provide these two different kinds of
          information may be sufficiently different to require
          greater out of pocket expenses than option 1.

     •    The assertion requirement of option 2 that the linkage to
          a confidential category be shown must be satisfied each
          time confidentiality is claimed for information not
          automatically linked to a category.  If there is a large
          quantity of non-automatically linked information, the
          assertion burden could become quite significant.if/

     The submitter's uncertainty is relatively greater under option 1.  Option
2 explicitly requires the submitter to demonstrate that information claimed
confidential is related to the category under which the claim is made.IV
This requirement, which clearly identifies the basis for EPA's assessment of
the validity of a confidentiality claim, enables the submitter to address those
points directly relevant to EPA's determination.  Option 1, while requiring
the submitter to provide the same general substantive material as option 2,
     il/CMA comments on the reproposal, p. 103.

     i§/EPA indicated (p. 59774, reproposal)  one reason for limiting to two
the number of substantiation questions for categories of information was "to
lessen the need for multiple confidentiality claims."

     i§/The timing of substantiation has a bearing on the administrative or
out-of-pocket costs.  This is discussed in the section "When to Substantiate".

     iZ/Although the industry has indicated that it considers this approach
"complex and confusing" (CMA reproposal comments, p. 99), EPA considers the
approach to be a clarification of what the Agency needs to make a decision
about confidentiality (EPA comments on the reproposal, 44 Federal Register
59774-59775).

-------
                                      -128-
does not directly link claims to that information which is most relevant to
EPA's consideration of confidentiality claims.  Therefore, under option 2 the
submitter may have less uncertainty as to whether or not its claim will-
address points relevant to EPA's consideration of a confidentiality claim.


                                  EXHIBIT 1-4

 THE RELATIVE COSTS OF OPTIONS FOR ASSERTING AND SUBSTANTIATING CONFIDENTIALITY
                                        Option 1
                                    (Check-off/Series
                                      of Questions)
     Option 2
(Linkage/Reproposal
    Questions)
   Submitter Cost 1
   ((P) disclose trade secret)

   Submitter Cost 2
   (Out of pocket costs)

   Submitter Cost 3
   (Uncertainty)
   The option with the higher number (i.e., 2)  has the higher relative cost.
Where there are no significant differences in the costs imposed by the
options, they are equally ranked.

Source:  MCA (subsequently CMA)  proposal comments; CMA reproposal comments;
         ADL Study;  EPA reproposal comments.
WHEN TO SUBSTANTIATE

     The cost associated with the requirement of submitters to substantiate
presents clearly defined alternatives with well-defined costs.  The four
options are discussed at length and their relative costs are summarized in
Exhibit 1-5, following the discussions on specification and relative costs.

Specification of Options

     There are four timing options that are presented for analysis.

     Option 1 requires substantiation only upon request from EPA.  The
submitter could assert confidentiality without substantiating the assertion
until requested.
     M/EPA's general confidentiality regulations define three such situa-
tions.  See footnote 2, supra, and accompanying text.

-------
                                      -129-
     Option  2  requests  that confidentiality claims  for chemical  identity and
 health  and safety  study data  be  substantiated when  the notice  is submitted;
 all  other information is  substantiated  (e.g., manufacturer's identity, use
 data, production volume)  upon request.

     Option  3  requires  that substantiation for  the  minimum amount of material
 be included  in a section  5(d)(2)  notice  (i.e.,  substantiate confidentiality
 claims  for specific  chemical  identity, use data,  and  information in a health
 and  safety study); to be  submitted when  the notice  is submitted; all other
 information  (i.e., manufacturer's identity, production volume, physical and
 chemical properties  data,  and other  information)  is to be submitted upon
 request.

     Option  4  requires  substantiation for all confidentiality  claims to be
 submitted with the notice.  The  rationale for this  option is that information
 should  be made available  as soon as  possible.   Thus,  instead of  waiting for an
 FOIA request,  confidentiality would  be determined when the notice is received,
 and  everything determined not to be  confidential  would be publicly disclosed
 (either through publication or by being  placed  in the public file).

 Relative Costs of  the Options

     The relative costs of these four options are discussed below.   Exhibit
 1-5  summarizes these costs.

     The cost  to the submitter in terms  of the  probability that  trade secrets
 will be disclosed is minimized by option 1 and maximized by option 4.
 Although protection  is  available for substantiation which is itself CBI,
 mistaken disclosure  is  still  a possibility.  The  harm posed by this situation
 (i.e.,  the damage if such an  event occurred multiplied by the probability of
 its  occurrence) is not  likely to be great, but  it increases as EPA is given
 more confidential information in support of the claims.  Therefore, when
 substantiation is provided only  upon request, less  information is in EPA's
 possession and less  probability  exists for disclosure of CBI.  Since option 1
 represents the situation where the least amount of  information is initially
 provided to EPA, it  is  the least expensive to the submitter in terms of
 probable disclosure  of  CBI.   Application of this  analysis indicates that
 option 4, which supplies EPA  with the most information initially, is the most
 costly of the  four options.   In  the long run, if  information claimed
 confidential in a notice  is subject to many requests  for disclosure, it may be
 that more information will have  to be submitted to  EPA under those options
 which require  less substantiation initially.ii/   However, since the latter
 situation is speculative and  the  former  (i.e., what is required to be
 substantiated  when the  notice  is  submitted)  is certain, option 1 is least
 costly, and options  2,  3, and  4  represent successively higher costs.
     ii/This is possible because piecemeal substantiation can be item-
specific, while substantiation submitted with the notice is intended to cover
all claims under a particular confidentiality category.

-------
                                      -130-




                                  EXHIBIT 1-5


            RELATIVE  COSTS OF THREE OPTIONS FOR  TIMING SUBSTANTIATION
Option 2 Option 3
(Chemical identity (5(d) (2)
and health & information
Option 1 safety data with the Option 4
(Nothing with the notice, the (Everything
with the notice, rest rest on with the
notice) on request) request) Notice)
1
1
4
2
2
3
3
3
2
4
4
1
Submitter Cost !£/
( (P) disclose trade secret)

Submitter Cost 2
(Out-of-pocket costs)

Submitter Cost 3
(Uncertainty)
fL/The option with the highest rank (i.e., 4) has the highest relative cost.

Source:  CMA reproposal comments, EPA reproposal comments, ADL Study,
         Environmental Action comments, EDF Comments, SOCMA comments.
     The out of pocket costs to the submitter are minimized by option 1 and
maximized by option 4.2jQ/  The initial substantiation effort by the
submitter requires the expenditure of administrative resources.  The less
substantiation that is required, the lower are the administrative costs
incurred by the submitter.  Since option 1 requires no substantiation to be
submitted with the notice, it substantially reduces (relative to the other
options) the submitter's burden in providing the notice to EPA.  Similarly,
since option 4 requires substantiation of every confidentiality claim to be
submitted with the notice, the option substantially increases (relative to the
     JL2/Industry has emphasized the potential for greater out-of-pocket
costs due to a requirement that substantiation be submitted with the notice
(CMA reproposal comments, pp. 119, 123-124, and 129).  On the other hand,
based on the assumption of fairly frequent FOIA requests, EPA has pointed out
the disadvantages of piecemeal substantiation (44 Federal Register 59775).
The ADL study indicated that the time cost of developing substantiation is
directly related to the amount of information which has to be substantiated
(pp. 50-51).

-------
                                       -131-
other  three options)  the  submitter's  initial  administrative burden  in
providing EPA with  the notice.  These initial relative costs are certain,
since  the notice must be  submitted  to EPA.  Any  subsequent relative costs,
however, may be quite different.

     If  there are many requests for substantiation  after  the notice is
submitted under option 1,  the submitter would have  to return to the
substantiation question in a piecemeal fashion.  Substantiating a claim
piece-meal fashion  after  satisfying most of the  notice requirements would be
less efficient than submitting substantiation along with  the notice. The loss
of efficiency occurs  because the submitter must  refamiliarize itself with the
relevant factors in the notice and  because some  duplication of administrative
effort will be necessary.   In this  situation,  it is not clear that one option
would  keep administrative costs lower than the other three options.2!/
Therefore, balancing  the  certain relative administrative  costs of providing
substantiation with the notice against the uncertain relative administrative
costs  of providing  substantiation upon request,  leads us  to conclude that
option 1 is the least costly in terms of the  submitter's  administrative
burden, and options 2, 3,  and 4 represent successively higher costs.

     The costs imposed by  the submitter's uncertainty are increased by Option
1 and  decreased by  Option  4.  The submitter's  uncertainty in this context is
derived from two sources.   First, if  substantiation is not provided with the
notice, the submitter does  not know when (or  if) subsequent substantiation
will be required; therefore, the submitter does  not know whether its claims of
confidentiality (for  which  no substantiation  has been submitted) will be
challenged. Second, when  less than  all the confidentiality claims are
substantiated, the  submitter may not  fully realize the inter-relation of the
confidentiality claims^/  and may either claim too much confidential (which
would  result in unnecessary administrative expenses) or not claim enough
confidential (which could  lead to unnecessary  disclosure of CBI).  As more
confidentiality claims are  substantiated coincidentally,  these uncertainty
costs  would be lessened.   Therefore,  option 1  represents  the greatest
uncertainty costs,  and options 2, 3,  and 4 represent successively lower
uncertainty costs.
GENERIC INFORMATION

     By supplying generic information, submitters are trying to protect trade
secrets while allowing EPA to disseminate to the public information about new
            would depend on the volume of subsequent requests and the
efficiency difference between substantiating when other notice requirements
are being met, and substantiating after the notice has been submitted.  The
volume of requests for disclosure has been the subject of some experience
under the current interim policy.

    22/See 44 Federal Register 59775.

-------
                                     -132-
chemicals introduced into the commercial market.   The less generic the
information, the less protection the submitter has for its trade secrets.
Yet, the more generic the information,  the less use such information has for
public evaluation of a new chemical substance.  EPA has suggested that generic
information can be specific enough to give meaningful information to the
public, yet broad enough to protect the submitter's confidential interest in
the specific information behind the generic mask (see, for example, 44 Federal
Register 59777).  The relative costs of the three options are summarized in
Exhibit 1-6, following the discussion below.

Specification of Options

     There are three generic options described below.

     Option 1 requires generic information that is only used to determine
chemical identity and categories of use.  Only one generic name for chemical
identity must be submitted, and the submitter is encouraged to obtain EPA
approval before submission.  The rule is that the generic name is to be "only
as generic as necessary to protect the confidential identity of the particular
chemical substance".  The generic description for categories of use must be
supplemented by the likely exposures to humans or to the environment and
toxicologically significant information.

     Option 2 requires the submitter to provide generic information for
chemical identity, categories of use, manufacturer's identity, and physical
and chemical properties.  For chemical identity,  the submitter must initially
provide EPA with three generic names.2_3/

     Option 3 requires no generic information to be submitted.  If an item of
information was confidential, nothing would be made public.

Relative Costs of the Options

     The three options present a spectrum of options with well-defined costs.
     ^I/This procedure can be avoided if the submitter opts to obtain EPA
approval of a generic name prior to submission of the notice.

-------
                                      -133-



                                   EXHIBIT 1-6

                  RELATIVE COSTS OF OPTIONS FOR GENERIC MASKING
                                                   Option  2
                                                    (chemical  ID,
                                                   chemical use,
                                                   manufacturer's
                                                   ID, and
                                                   physical/
                                                   chemical
Option 1
(chemical ID
and
                                    chemical use)  properties)
Option 3
(no generic
information)
Submitter Cost 1
((P) disclose trade secret)

Submitter Cost 2
(Out of pocket costs)

Submitter Cost 3£L/
(Uncertainty)
2
2
2
3
3
3
1
1
1
£/The option with the highest number  (i.e., 3) has the highest relative cost.
Source:  CMA reproposal comments, MCA  (subsequently CMA) proposal comments,
         CSMA comments, SOCMA comments, comments by Monsanto and FMC Corp.
         EDF comments, EPA reproposal comments.

-------
                                    -  134  -
     Option 3 imposes the lowest relative cost in terms of the probability
that The submitter's trade secrets will be disclosed.jj7Option 3 requires
the least amount of information from the submitter for public disclosure.  As
a general rule,  the probability of trade secret disclosure has a direct
relation to the amount of information which is publicly disclosed.  Since
option 3 requires no generic information, it imposes the least amount of costs
in these terms;  options 1 and 2 are relatively more costly.  Option 1,
however, requires only two items of generic information while option 2
requires four.  In addition, option 1 enables the submitter to develop the
generic categories of use information according to its own scheme which is
responsive to the submitter's concerns about trade secret disclosure.  Option
2 provides the same opportunity, but requires an additional explanation of why
EPA's scheme is not adequate (providing such an explanation would also involve
additional administrative costs for the submitter).  Under both options 1 and
2, there is a possibility that the initial section 5(d)(2) notice (published 5
days after receipt of the notice)  will have to be amended if its generic
chemical identity is determined to be too broad.  Combining the initial
generic name published and the amended version of the generic name may reveal
sufficient information to unmask the specific chemical identity.  Of the three
options, therefore, option 3 imposes the least relative costs, and options 1
and 2 impose successively higher relative costs in terms of the probability
that trade secrets will be disclosed.  (See also, MCA (subsequently CMA)
comments on the proposal, p. 319.)

     Option 3 involves the lowest relative costs for the submitter in terms of
out-of-pocket expenses.  Because option 3 requires the submitter to develop
and submit no generic information,  it represents a clearly lower relative out-
of-pocket cost.   Another savings is the absence of negotiations with EPA
concerning the adequacy of particular generic information.!!/  It is more
difficult to determine which of the other two options is least costly.
     ±j!/Industry has made clear its position that supplying any generic
information for public disclosure will compromise the submitter's confidential
business information.  The following examples from CMA's comments on the
reproposal indicate industry's attitude:

     •    It is a faulty premise "that useful generic data can be
          disclosed to the public without itself revealing the
          underlying confidential information." (p. 151).

     •    "The purported benefit of the (generic)  scheme ... is
          highly dubious and in any event cannot be achieved without
          a serious risk of compromising submitter's rights to
          confidentiality," (p. 156).

     •    "Although disclosure of these range data would be of
          questionable value to the public, they may prove extremely
          valuable to the submitter's competitors." (p. 161).

     M/See CMA's commments on the reproposal, pp. 104, 151, and 155.

-------
                                      -135-
because the two are not directly comparable.  In particular, option 2 permits
three of the four generic descriptions to be constructed from a "laundry list"
of characteristics or ranges compiled by EPA — a process far easier than
developing generic information "from the ground up" as required by option 1.
However, option 2 also gives submitters the alternative of using their own
schemes (with proper justification) to develop this same information.  The
submitter will use the EPA scheme only if it believes the results will be
satisfactory from its viewpoint.  The relative costs of the two options will
depend upon its decision.

     Unlike option 2, option 1 does not require submission of generic
information for manufacturer's identity or physical and chemical properties;
this represents a clear time savings for option 1 in comparison to option 2.
Option 1's provision for generic categories of use information, however,
requires development solely by the submitter, and thus seems more burdensome
than option 2's "laundry list" alternative.  (Note:  if the submitter opts for
its own scheme for developing the generic categories of use information, there
is little difference between options 1 and 2 on this point.)  Similarly,
neither of the two option's requirements for generic chemical identity
presents clear-cut administrative cost savings over the other option.  Option
1 requires the submission of only one generic chemical identity, but obtaining
EPA's prior approval of the choice is encouraged.  Thus, the submitter is not
formally limited in the number of generic names it may have to develop.
Option 2 requires the submitter to develop at least 3, but no more than four
generic chemical identities.  Under option 2, the submitter also has the
alternative of obtaining EPA's prior approval for a single generic chemical
identity and avoiding the requirement for initial submission of three generic
names (under this alternative, option 2 is identical to option 1).

     In sum, option 1 requires less information than option 2; where the two
options overlap (i.e., for chemical identity and categories of use), it is not
clear which one will, in practice, require more time of the submitter.  By
requiring less information, option 1 appears to require less time than option
2 and should represent a relatively lower out-of-pocket cost for the
submitter.26/  Therefore, option 3 is the least costly option in terms of
the submitter's out-of-pocket costs, and option 2 is the most costly.

     Option 3 imposes the least costs in terms of submitter uncertainty about
EPA actions.  EPA decides whether generic information is adequate (i.e.,
           should be noted that option 2 presents a situation in which the
submitter may opt for higher costs of one kind in order to avoid higher costs
of another kind.  In particular, the EPA-supplied scheme for chemical use,
manufacturer's identity, and physical and chemical properties may represent
time cost savings because the submitter does not have to design its own frame-
work.  However, in the submitter's opinion, the EPA scheme may not provide as
much protection for trade secrets as a generic description developed indepen-
dently by the submitter.  Although it would involve higher administrative costs
to develop the independent description, the submitter may choose to incur that
cost in order to avoid the much greater cost imposed by endangering the
confidentiality of trade secrets.

-------
                                      -136-
whether it is as specific as possible without revealing CBI).   The submitter,
therefore, may be uncertain whether its perception of what is  adequate
coincides with EPA's perception.  Since option 3 involves no submission of
generic information, the question of EPA's acceptance of the information does
not arise.  Option 1 involves only two kinds of generic information, while
option 2 involves four; option 1, therefore, should involve less uncertainty
for the submitter relative to option 2.J_7/  Option 2, however, does lessen
substantially the costs associated with uncertainty due to the requirement for
additional generic information.  Under option 2, EPA has provided a framework
for developing all four items of generic information and has indicated that if
this framework is used, the information developed will generally be acceptable
to EPA.
CHEMICAL IDENTITY AS PART OF HEALTH AND SAFETY STUDY

     This issue is one of particular sensitivity to submitters.  Two sets of
options are discussed; the first is the timing of disclosure for chemical
identity that is part of a health and safety study,H/ and the second is how
chemical identity will be added to the inventory.  The relative costs of the
two sets of options are summarized in Exhibits 1-7 and 1-8 below.

Chemical Identity Disclosure Timing Options

     If a notice includes chemical identity which will be disclosed because of
the factors discussed above, EPA has three options on when to disclose this
information.

     Option 1 requires disclosure of the chemical identity when the notice is
submitted.  Thus, the specific chemical identity will be known at least 90
(and as many as 180) days before the submitter can begin its commercial
manufacture.

     Option 2 requires disclosure when manufacture begins which prevents
disclosure during the period between submission of the notice, its approval,
and the commencement of manufacture.

     Option 3 requires disclosure when the chemical is distributed in
commerce.  Distribution occurs some time after manufacture begins, but it is
not apparent that this is a significant difference from option 2.  The
submitter will not be commercially manufacturing the chemical without
    AZ/Under both options 1 and 2 the submitter can seek pre-notice
consultation with EPA.  This reduces uncertainty about whether the generic
information submitted with the notice will be adequate.  However, there still
is uncertainty during the pre-submission stage, and this uncertainty also is
relatively higher for option 2 than for option 1.

    2j§/If chemical identity is not a part of the health and safety study,
the issue does not arise.

-------
                                      -137-
                                   EXHIBIT 1-7

           THE RELATIVE COSTS OP OPTIONS FOR TIMING THE DISCLOSURE OF
             CHEMICAL  IDENTITY AS PART OF A HEALTH AND SAFETY STUDY
Submitter Cost 1
((P) disclose
 trade secret)

Submitter Cost 2
(Out-of-pocket
 cost)

Submitter Cost 3
(Uncertainty)
                            Option 1

                          (disclose when
                          PMN submitted)
   Option  2
(disclose  when
 manufacture
    begins)
   Option 3
(disclose when
distributed in
    commerce)
A "1" indicates the least costly alternative.  Where there are no significant
differences in the costs imposed by the options, they are equally ranked.
Source:  MCA (subsequently CMA)  proposal comments; Environmental Action
         comments; EOF comments; NRDC comments; CMA reproposal comments; EPA
         comments accompanying the Interim Policy and the Reproposal.

-------
                                      -138-
                                   EXHIBIT 1-8

           THE  RELATIVE  COSTS OF OPTIONS  FOR PLACING CHEMICAL IDENTITY
            (CONTAINED IN  A  HEALTH  AND  SAFETY STUDY)  ON THE INVENTORY
                                        Option 1
                                   (Place on inventory
                                    by specific name)
     Option 2
(Place on inventory
  by  generic name)
Submitter Cost 1
((P)  disclose trade secret)

Submitter Cost 2
(Out-of-pocket cost)

Submitter Cost 3
(Uncertainty)
A "1" indicates the least costly alternative.  Where there are no significant
differences in the costs imposed by the options, they are equally ranked.

-------
                                      -139-
assurances that it has a market for the chemical.  Thus, if the manufacturing
process is relatively short, it may be a matter of weeks between the
commencement of manufacture and the distribution in commerce.  On the other
hand, if the manufacturing and distribution process is relatively lengthy,
there may be a significant time difference between the commencement of
manufacture and the distribution in commerce.

The Relative Costs of the Timing Options

     Each of these options imposes different costs on the submitter.  Exhibit
1-8 summarizes the relative costs.

     Option 1 provides the earliest disclosure.  The earlier disclosure
imposes a great burden on the submitter because its specific chemical identity
is revealed before commercial manufacture has begun.H/  Earlier disclosure
increases the submitter's costs of maintaining trade secrets.  For example, if
a trade secret's primary importance lies in giving the submitter an early
start on its competition, the earlier the trade secret (in this instance, the
chemical identity) is disclosed, the greater the harm.  Other costs for the
submitters are not significantly affected by this set of options.

     Option 2 is the mid-ground of the three options.  Information is revealed
more slowly than option 1, but more quickly than option 3.  Similarly, option
2 prevents possible disclosure of trade secrets for a longer period than does
option 1, but not as long as option 3 (although, as discussed above, the
length of the interim between 2 and 3 is not clear).

     Option 3 imposes the lowest costs for submitters.  Option 3 protects the
chemical identity from disclosure longer than the other two options and, thus,
postpones possible trade secret disclosure.  Furthermore, option 3 may provide
the submitter with all the competitive advantage it had initially planned.

Inventory Options

     The final set of options is the addition of chemical identity (that is
disclosed as part of a health and safety study)  on the inventory.  There are
two options:

     •    addition to the inventory by specific chemical name, or

     •    addition to the inventory by generic chemical name.

Exhibit 1-9 summarizes the relative costs.
     H/The industry repeatedly emphasizes the importance of chemical
identity confidentiality.  See, e.g., MCA comments in the proposal (p.  319),
and CMA comments on the reproposal (p. 138).   It is especially important in
the pre-commercial stages.  See, e.g., ADL's  Industry Interviews.

-------
                                      -140-
     The inventory options do not affect the timing options considered above;
they will still occur. 3Q/  The rationale for the option to add the chemical
to the inventory by generic name is that most of the submitter's competitors,
as a matter of course, will maintain an up-to-date inventory.  If the specific
chemical identity was placed on the inventory, it would be readily available
to the submitter's competitors.  On the other hand, adding only the generic
name to the inventory would require competitors to consult the public record
to discover the specific chemical name (i.e., public disclosure, which is
separate from addition to the inventory,  is made by placing the specific
chemical identity in the public file, not by publishing it as an individual
document).  The generic option, in effect,  does not prevent competitors from
obtaining the specific chemical identity; it merely makes the competitor's
ability to gain the information slightly more difficult.   At best, this
protection gives the submitter a little extra time to keep the specific
chemical identity obscured from competitors.
     12/Note, however,  that there will be an inconsistency if the chemical
identity is added to the inventory when commercial manufacture begins and
public disclosure is not made until commercial distribution,   in such a
situation,  delaying public disclosure until distribution has  little effect.

-------
                                      -141-


                                    CHAPTER 2

                                CUSTOMER CONTACT
     The thorough evaluation of the environmental risks of producing any new
chemical may require EPA to obtain information from those who purchase the
chemical.  Information sought could include data on exposure of workers during
processing, environmental release, disposal, and use by the general
population.  To obtain this information, EPA has proposed several alternative
rules.  In this paper, we assess the relative costs to industry of these
alternatives on the basis of information in the public record.

     In the following sections, we will:

     •    specify the options under consideration,

     •    specify the types of costs incurred by industry, both
          submitters and customers, and

     •    estimate the relative burden to industry of each option
          for each type of cost.

    It is difficult to discuss customer contact in isolation from the other
issues.  Both supplemental reporting and confidentiality play a role in the
discussion of customer contact.  In our discussion, we assume that the Agency
may obtain supplemental information and that confidentiality provisions will
protect confidential data from disclosure.   The interactions among the issues
are discussed in the section on comprehensive program options.
CUSTOMER CONTACT OPTIONS

    The options proposed by EPA are:

     •    Option 1;  January 10 Proposal.  The submitter must
          contact all customers in writing, unless he believes that
          the information is already contained in the notice.  If he
          believes that contacting all customers would be
          duplicative, he can contact a representative sample.  The
          submitter must request each person contacted to complete a
          Processing and Consumer user form.  The persons who must
          be contacted are:

               firm customers—persons party to a contract to obtain
               the substance from the submitter;

               likely customers—persons who have contacted the
               submitter and indicated an interest in obtaining the
               substance;

-------
                           -142-
—  • Potential customers—persons who have obtained a
     sample of the substance and who have indicated an
     interest in purchasing it; and

     Possible customers—persons whom the submitter has
     contacted or intends to contact and whom the
     submitter firmly believes will purchase the substance
     during the first three years of commercial production.

The submitter must provide EPA with a list of the names of
the customers as well as with any customer contact forms
returned to him (the customers are not required to respond
and they may respond either to EPA or to the submitter).

Option 2;  October 16 Proposal.  Under option 2, the
submitter would have to indicate in the notice the number
of customers who have made a firm commitment to purchase
the substance for a category of use unknown to him and the
percentage of estimated production that such customers are
expected to purchase during the first three years of
production.  EPA may subsequently ask the submitter to
voluntarily provide the agency with the names and
addresses of customers; EPA can then contact customers
directly.  A variation on this option is to require the
submitter to indicate on the notice the number of all
expected customers whose use of the product is unknown to
him, not just those who have made a firm commitment to
purchase the substance, and the percentage of estimated
production that such customers are expected to purchase
during the first three years of production.

Option 3;  The submitter would have to indicate in the
notice the estimated number of customers for all
categories of use, including unknown categories, as well
as an estimate of production volume expected for each of
those uses for the first three years.  EPA may ask the
submitter to voluntarily provide the agency with the names
and addresses of customers.  Or EPA may require that a
customer list be furnished, under the supplemental
reporting provisions.  EPA can then contact customers
directly.

Option 4;  In addition to the information provided under
option 2, the submitter is required to provide EPA with a
list of the names of potential customers.  The customers
can be divided into four categories, as in option 1.

Option 5;  Elimination of  all customer contact
provisions.  The submitter need not contact customers,
provide a customer list, or estimate the number of
customers in various categories of use.

-------
                                     -143-
Evaluation Framework

     In this section, we evaluate the direct costs to industry of each option.
Option 5, elimination of all customer contact provisions is used as the base-
line from which to assess the other options.  Under options 1 and 4, the
effect of using different definitions of "customer" will be assessed.   These
definitions of customer will be:

     (A)   firm customers only;
     (B)   firm and likely customers;
     (C)   firm, likely, and potential customers;
     (D)   firm, likely, potential, and possible customers.

The direct costs to the industry are seen as the following:

     •    cost of compiling customer lists;
          direct cost of contacting customers;
          delay in the introduction of new chemicals;
          uncertainty;
          cost to customers of providing information; and
          reduction of demand for new chemicals.

     Because each cost will vary depending on the behavior of submitters and
EPA, the nature of other section 5 rules, and the nature of specific chemical
markets, it is not reasonable to specify the exact costs attributable to the
customer contact provisions.  It is reasonable, however, to specify relative
costs of each option to industry under each type of cost, comparing each
option against the baseline (option 5).  The result is a matrix of options and
costs, shown in Exhibit 2-1.  The justification for the entries in the matrix
is given in the rest of this section.
COST OF COMPILING CUSTOMER LISTS

     The cost to a chemical producer of compiling a customer list depends on
EPA's definition of "customer".  The four categories of customers are listed
under option 1, and one or more of those categories comprises the customers
referred to under the other options.  Therefore, in determining the cost of
compiling the customer list, we examine the special characteristics of each of
these categories of customers.

     The cost of compiling a customer list is largely determined by the
internal recordkeeping practices of chemical firms and how many customers are
to be contacted.  For example, if chemical firms normally compiled lists of
customers for each product and the firm's definition of customers were the
same as EPA's definition, the additional cost of compiling and submitting a
customer list to EPA would involve only minor administrative and postage
expenses.  Or, if information on each customer for each chemical were kept on
a separate card or sheet, compiling a customer list would be only slightly
more expensive than the previous case.  The firm would need only assemble the
separate documents and type a list.  If the documents were not kept in some

-------
                                      -144-
                                   EXHIBIT 2-1

     RELATIVE  COSTS  OF  CUSTOMER  CONTACT OPTIONS 1-4,  AS COMPARED TO OPTION 5
Option

Customer^/
Definition
B
                                                                              D
Cost to Industry

Compiling Customer
  List

Direct Cost

Delays

Uncertainty^/

Cost to Customers

Reduction in Demand
2
2
2
3
3
3
3
2-3
2
3
3
3
3
2-3
2
3
3
3
4
3
2
3
3
4
2
1
2
3
2
2
2
1
2
3
2
2
2
1
2
3
2
2
3
1
2
3
2
2
3
1
2
3
2
2
4
1
2
3
2
3
1 = Negligible
2 = Low
3 = Medium
4 = High

£/The four definitions of customer are:

       (A)  firm customers only;
       (B)  firm and likely customers;
       (C)  firm, likely, and potential customers;
       (D)  firm, likely, potential, and possible customers.

b_/ = Initial costs—should decrease over time.

-------
                                      -145-
 central location, but instead were scattered among a number of salespeople,
 the cost would increase, especially if the documents were located in different
 plants.  (The firm would have to assemble the documents in some central
 location.)   Furthermore, if there were no written records specifying exactly
 who is a customer for each chemical,  it would be necessary for that person
 responsible for completing the notice to contact each salesperson and elicit
 the information in writing.

      Information about firm customers—those who have signed contracts—will
 be listed on some written document.  A customer list may not exist and, depend-
 ing on the time when the notice is filed, the documents containing the names
 of customers may not be gathered in a central location.  But a written record
 should exist.  A written record should also exist for likely customers—those
 who have obtained samples—though the record may not indicate if the person  is
 interested in purchasing the chemical, and the records may not be kept in  one
 central location.  Information about  potential customers—those who have
 contacted the submitter and indicated an interest in obtaining the substance—
 and about possible customers—those whom the submitter has contacted or
 intends to contact and whom the submitter firmly believes will purchase the
 substance during the first three years of commercial production—may not be
 kept in written form. '

      The total cost of compiling a customer list cannot be found merely by
 adding the  costs associated with the  current method  of internal record-
 keeping.   A firm might find it less costly to change its method of record-
 keeping to  produce the type of records required by section 5.   It is unlikely
 that we will be able to estimate the  cost of compiling a customer list if
 internal recordkeeping is altered,  because we would  need to  know more about
 chemical companies'  data management systems before we  could make such an
 estimate.   Therefore,  we will  assume  that internal recordkeeping is  not
 altered.  We would note,  however,  that the cost of compiling  a  customer list,
 based on current recordkeeping practices,  is an upper  bound of  the actual
 costs,  because firms might conceivably alter their recordkeeping procedures
 and lower  their costs.

      We should also note that  the  firm has some control over  the cost of
 providing a  list,  because the  decision that someone  is a  likely,  potential, or
 possible  customer  is,  to a certain  extent,  a subjective judgment,  especially
 when possible  customers  are  involved.   An examination  of  the  language  defining
 possible customers—"whom the  submitter firmly believes will purchase  the
 substance during  the first  three years of  commercial production"—shows the
 amount  of discretion available  to  the  submitter  in determining  who are
 possible customers.

      Under options  2 and  3, EPA may ask  the  submitter  to voluntarily provide
 the  names of customers in  those cases  in which  the chemical is  under detailed
 review; EPA  estimates  a  small  number of  such cases.H/  under options  1
           Federal Register 59765  (October 16, 1979).  EPA anticipates that
it will be able to decide that most chemicals require no further regulation on
the basis of the information provided in the notice and a literature search
conducted by EPA staff.

-------
                                      -146-
and 4, complete lists of customers are to be provided in all cases.  Because
the lists are required in four or five times as many cases under options 1 and
4 than under options 2 and 3, the costs associated with the latter options
should be substantially smaller.

     The costs of options 1 and 4 depend greatly on just who is included in
the list of customers and on the  way in which records are kept in the chemical
industry.  We estimate that if only firm customers are included, the relative
cost is likely to be small, because written records must be kept on customers
with signed contracts.  If likely customers are added, the cost could increase
substantially, since it is less probable that a written record of these
customers would exist.  The addition of potential customers should not
increase the cost much since some written record of these customers should
exist.  However, the addition of  potential customers could make the relative
cost quite large, because the firm might well have to piece together infor-
mation from a large number of its employees in order to compile such a list.

Direct Cost of Contacting Customers

     Only under option 1 does the submitter bear the cost of contacting
customers.  Under the other options, the cost of contacting customers is borne
by EPA.  With any recordkeeping system, under option 1 the costs, consisting
of postage and the administrative burden, will be virtually proportional to
the number of customers contacted.  If the number of customers is large and
the producer believes that the information they provide will be duplicative,
the producer can contact a sample.  This should reduce somewhat the cost of
customer contact.

     Under options 2, 3, and 4 the contact cost is no different from the
baseline, because EPA bears the cost.  Under option 1, the costs to the
submitter vary with the number of customers contacted.  We therefore expect
direct costs of contacting customers to be low if only firm customers are
included, low to medium if likely and potential customers are included, and
medium if all customers are included.

     Arthur D. Little, Inc. estimated the cost of completing the notice form.
However, we can not extract from  those estimates the cost of the information
that must be submitted under the options 2, 3, and 4.  In the ADL estimates,
the cost of questions on customer use is grouped wth other questions and the
costs cannot be disaggregated.

Delays in the Introduction of New Chemicals

     There are two ways in which  the customer contact provisions could cause a
delay:

     •    The process of obtaining information from customers
          results in an extension of the 90-day notice period.  As
          noted above, customer information is usually important
          only when the substance is under detailed review; in these
          cases, the notice period will most likely be extended
          anyway.  It is possible that in some cases which are not

-------
                                     -147-
          under detailed review, customer information might be
          important and delays could occur while the information is
          acquired.  But analysis of the relevant options (options
          2, 3, and 4) suggests that delays should occur very
          infrequently.  Therefore, in these cases, the customer
          contact provisions will most likely not cause any delays
          in the introduction of a chemical.

     •    When little is known about the chemical, customers may be
          contacted to determine if detailed review is necessary.
          In these cases, if customers respond slowly, the review
          period might be extended and the introduction of the new
          chemical delayed.  Similarly, customers might conceivably
          supply information that would subject the chemical to
          detailed review.

     As stated above, options 2, 3, and 4 would result in only a small chance
of delays in new chemical introduction; the overwhelming number of customer
contacts would involve chemicals subjected to detailed review, for which the
notification period would already have been extended.  Under option 1, the
customers would be contacted routinely so the process itself would not result
in delays.  If customers did not provide information under option 1 voluntar-
ily, EPA might seek the information under the supplemental reporting provi-
sions, and the use of these provisions might cause an extension of the notice
period and a delay in the introduction of a new chemical.  But again,
supplemental information will only be requested in those cases in which the
substance comes under detailed review.  The total cost, therefore, is likely
to be small.

Uncertainty

     In addition to the costs which may actually be imposed upon chemical pro-
ducers and consumers, uncertainty about how EPA will implement whatever option
is eventually chosen and uncertainty about what those costs will be, will
impose costs on chemical producers.  Because of this uncertainty, firms may
devote more or fewer resources to the section 5 notice process than are
required, and they make take actions that result in unnecessary delays.  Some
of this uncertainty is inevitable and will diminish as the Agency and industry
gain experience with the Act and with each other's actions related to the
Act.  But some residual uncertainty and associated negative consequences will
remain.

     With the exception of option 5, which contains no provisions for customer
contact, each of the options allows EPA to retain a great deal of discretion
in the degree to which it will contact customers and, therefore, in the costs
it imposes on the industry.   We estimate that the initial costs of uncertainty
should be medium under all options, but should decrease over time.  However,
it should be noted that the estimate of the level of uncertainty costs is
itself quite uncertain.

-------
                                      -148-
Cost to Customers

     When customers choose to provide information to EPA rather than forego
use of the chemical, they will incur costs in providing information to EPA.
The costs include more than just the time and resources necessary to provide
EPA with the answers to its questions; the costs also include the risk that
important information will become available to competitors.   We have assumed
that the same confidentiality provisions that apply to information obtained
from producers will also apply to information obtained from  customers.  The
confidentiality provisions are analyzed in another section.

     As noted above, under options 2, 3, and 4, EPA anticipates that customers
will be contacted only if the chemical is under detailed review,  which should
occur relatively infrequently.  In those cases, the costs should  be small.
Under option 1, customers are contacted and EPA can require  customers who did
not respond voluntarily to provide information to EPA.   Because customers will
be providing information to EPA in many cases under option 1 (the number
depends on the extent of voluntary responses by customers)  in which they would
not provide information under the other options, the cost of option 1 will  be
greater.

Reduction of Demand for New Chemicals

     The costs incurred by customers may result in a downward shift of the
demand curve for new chemicals.  Under the customer contact  provisions,  the
customer has the option of refusing to provide the information voluntarily.
However, the possibility exists that if the information is not provided
voluntarily, EPA will make worst case assumptions,.3V leading to
restrictions on the production of the chemical.  Thus,  the firm must report to
EPA if it is identified as a customer.  The need to provide  this  information
will act as a deterrent to firms with an interest in the new chemical; they
will be dissuaded from buying it.  The result will be a loss of profits  and
sales for the producer of the new chemical.

     It is important to note that a potential customer  cannot be  compelled  by
EPA to provide information during the notice period.  Although the customer is
not required to provide information, he is faced with a choice:  either
provide the information requested or risk unfavorable activity.  Both actions
impose costs on the customer; either he must incur the costs of providing
information to EPA or forego the chance to use a new, presumably  more
efficient chemical.  The customer's choice should depend on  his valuation of
the chemical.  If he believes the value of the chemical outweighs the cost  of
providing information to EPA, he will pay for providing information; if  not,
he may forego use of the new chemical.  Therefore, there will be  a downward
shift in the demand curve.  Consequently, a decrease in the  introduction of
     21/44 Federal Register 2244 (January 10,  1979).

-------
                                       -149-
new chemicals  in  those cases  in which customers are contacted will occur.  But
those chemicals that  the customer  foregoes are, in all likelihood, chemicals
whose expected benefits to  the customers are small.33/

     Even  if some new chemicals are not introduced because customers are
discouraged from  purchasing the chemicals by the burden of providing
information to EPA, it is not clear to what extent the chemical industry as a
whole suffers  any loss.  Certainly the firm which loses customers suffers a
loss in  sales  and income, but the  producer of  the old chemical which would
have been  replaced by the foregone innovation  does not lose sales and income.
In this  case,  section 5 serves only to redistribute sales and income among
chemical producers.   The chemical  industry as  a whole experiences a loss only
when the foregone chemical  would have replaced something other than a chemical
 (perhaps even  expanding demand because of cost reduction) or when imports take
over a market  because a new chemical is not produced in the United States.  It
is important to note  that no  producer of a new chemical will be more or less
disadvantaged  than other producer  of a new chemical, because the same
requirements apply to all new chemicals.

     There is  another possible consequence of  the customer contact provisions
which could reduce the demand for  new chemicals from the original innovating
companies.  A  customer who  is also a potential competitor may be alerted to
the development of a  new chemical; he might preempt the innovator by producing
the chemical itself instead of purchasing it from the innovator.  Such a
circumstance is only  a danger for  possible customers—those whom the producer
firmly believes will  purchase the  substance during the next three years—
because  customers in  the other three categories must, by definition, already
know about the chemical.  In  addition, by the  time customers are contacted,
non-confidential  information  on the submission is about to be placed in the
Federal_Register  and  a sanitized version of the notice made available to the
public.34/Therefore, the  danger  of preemption is increased only to the
extent that the definition  of customers includes possible customers and to the
extent that possible  customers, who are really interested in preempting the
innovator, would  not  be alerted to the new chemical as a result of the Federal
Register notice.   The extent  of the danger cannot fully be determined without
extensive information about chemical industry  innovation and marketing, but we
    .3_3/The issue of the importance of innovation to both society and the
chemical industry and the effects of section 5 on innovation is a central part
of the overall analysis of the economic effects of section 5.  A
methodological framework for this analysis has been developed and was
published in the Federal Register, and the discussion of innovation found here
is consistent with that methodology.  We recognize that additional analysis is
needed in this area.

    li/A submitter must certify on the notice form that he has contacted his
customers.  That contact must have occurred before the notice is submitted,
but it appears that contact on the same day that the notice is submitted would
not violate the rules.  The Agency must publish a Federal Register notice five
days after receipt of the form from the submitter.

-------
                                      -150-
expect it to be small,  because the number of cases in which the necessary
conditions for preemption exist will be quite small.   It is possible that
competitors will obtain useful information from the notices, but the potential
disclosure of confidential business information is a cost of the
confidentiality policy  and is not discussed here.   However, the additional
potential for disclosing confidential business information due to the customer
contact provisions should be small, because in only a few cases would
additional information  actually be made available  to competitors through the
customer contact process.

     Because the costs  imposed on customers will be greater under option 1
than under the other options, the costs of reduction in demand will be greater
under option 1 than under the other options.  In addition,  because of the
possibility that potential competitors may preempt the initial developer, the
cost of customer definition option D, under options 1 and 4, in which all four
categories of customers are included in the customer list,  will be greater
than the costs of the other customer definition options.

SUMMARY

     Under option 1, EPA requires information from customers in many cases in
which information would not be required under the  other options.
Consequently, under option 1 the costs imposed on  the industry are
substantially larger than under the other options.  The extent of the
difference depends on the definition of customer that is adopted.  Option 4
presents some of the same problems as option 1, but in a less extreme form;
customer contact is not required in all cases, but provision of a customer
list is.  Under option  5, all costs to the industry are avoided.  The
differences between options 2 and 3 are very small.  In fact, the only
difference is that under option 3, the submitter would have to indicate on the
form the estimated number of customers for all categories of use, including
unknown categories; under option 2, the number of  customers for unknown uses
only would be estimated.  The difference in cost between options 2 and 3 is
trivial because making  the additional estimate required under option 3
requires little additional time.  There is little  difference in the cost to
industry of reporting all uses as opposed to unknown uses.

-------
                                      -151-
                                    CHAPTER 3

                             DEFINITION OF IMPORTERS
     Under section 720.10(a)(2) and (b)(1) of the January 10, 1979 proposed
rules, "any person who intends to import" a new chemical substance is required
to submit notification to EPA.  Import transactions vary in the number and
type of parties involved and in the methods of importation.  Some parties
involved in the import process may be unaware of information about the
chemical substance.  There  is a variety of participants in importation, and
some of the information required on the notification form is highly
technical.  Therefore, EPA  is considering which parties should be responsible
for submitting notification on new chemical substances that are imported.

IMPORTER DEFINITION OPTIONS

     The contractor has evaluated five alternatives for designating persons
responsible for submission of section 5 notification on imported new chemical
substances:

     Option 1;  Define Importer as the Consignee.  This definition would
identify one person as responsible for notification, although the consignee
would not necessarily possess enough information to submit an adequate
notice.  The consignee is the person to whom goods are delivered, and may be
anyone involved in importing a chemical, including a retailer or broker.

     Option• 2;  Define Importer as Consignee and Impose upon Him an Obligation
to Obtain the Information Required by the Rules and Forms from Other Persons.
These other persons, including identifiable future users and processors of the
chemical, would be expected to provide information to the submitter or EPA, if
information is known or fairly easily ascertainable.  Such a system would
provide EPA with more information than might be submitted by the consignee
alone.

     Option- 3;  January 10th Definition of Importer—Section 720.2.
"Importer" means any person who imports a chemical substance or a chemical
substance that is part of a mixture or article into the Customs Territory of
the united States.  The definition of "importer"  includes the person
primarily liable for the payment of any duties on the merchandise or an
authorized agent acting on his behalf.  "Importer" also includes as
appropriate:

     •    the consignee;
     •    the importer of record;
     •    the actual owner if an actual owner's declaration and
          superseding bond has been filed in accordance with 19 CFR
          141.20; or
     •    the transferee .  . .".

-------
                                      -152-
This definition is derived from the definition of importers used by the U.S.
Customs Service which defines importer as "the person primarily liable for the
payment of any duties on the merchandise or an authorized agent acting on his
behalf." (19 CFR 101.1 (k)) .  According to this definition, each of the
persons listed would be responsible for ensuring that notification is
submitted to EPA; a single person with responsibility for submission is not
identified.

     Option 4;  Define Importer as "The Person Who Imports, or Knowingly
Causes to be Imported a Chemical Substance".  The person ordering a chemical
substance in a domestic transaction would knowingly cause the substance to be
imported if he places an order knowing it will result in the importation of
the substance.  This definition is similar to one adopted by the Customs
Service for reporting requirements and was suggested by the American Importers
Association in their comments on the January 10th proposal.

     Although this definition would theoretically require the most
knowledgeable person to submit notification, in reality it might prove
difficult for both the Agency and the parties involved to identify the single
submitter responsible for notification.  This definition also might not reach
the most knowledgeable person.

     Option 5;  Clarify January 10th Definition of Importer in the Rule Itself
and in Other Relevant Documents to Indicate that EPA Will Only Allow the
"Principal Importer" Involved in Import to Submit the Section 5 Notice.  The
"principal importer" is the person who, knowing the new chemical substance
will be imported, selects the identity and the total amount of the chemical
substance to be imported.  The "principal importer" is also likely to be the
person who will process, use, or distribute the chemical.

     This clarification would aim at getting the most knowledgeable person to
submit notification and is intended to reach a single individual.  EPA would
further identify the responsible submitter by providing examples and
instructions for different types of import transactions.

Evaluation Framework

     The five options are evaluated on the basis of costs to industry.  These
costs are seen as the following:

     •    Costs of Notice Submission, including out-of-pocket costs
          of filling out the notice form, and bargaining, delay, and
          costs associated with uncertainty when no single submitter
          is easily identifiable.

     •    Post-Submission Costs:  dollar and delay costs of
          clarifying supplemental information, potential for trade
          secret disclosure, and action under section 5(e).

-------
                                      -153-
     Exhibit  3-1  shows  a matrix of  the options and their relative costs and is
discussed more  fully  below.   It is  assumed  that the notification form
submitted for all five  options is the October 16th reproposed form.
                                   EXHIBIT 3-1

                                COSTS TO INDUSTRY
                       Option  1

                        narrow
Coverage

Cost of
Notice Submission:
Out of  Pocket              1

Delay,  bargaining          1

Uncertainty  (liability)    0
Option 2    Option 3    Option 4    Option 5

 broad       broadest    broad       narrow
                                      2

                                      2

                                      2
                2

                2

                3
1

1

2
1

1

1
 Post-Submission Cost

 Supplementary Reporting   3

 Section  5(e) Action       3

 Trade Secret Disclosure   1
                                      2

                                      2

                                      3
                2

                2

                2
2

2

1
1

1

1
Note:  O = no cost, 1 = lowest cost, 2 = next lowest cost, etc.
COVERAGE

     As outlined in Exhibit 3-1, coverage refers to the number of people that
would likely be involved in the preparation and submission of the notice
form.  Option 3 requires each of the parties included in the definition of
importer to be responsible for submission and, therefore, has the broadest
coverage.  Options 2 and 4 can include more than one individual submitter, if
(under option 2) the importer must contact more knowledgeable persons in order
to complete the form, or if (option 4) uncertainty exists as to who knowingly
caused a chemical to be imported.  Both options 1 and 5 are more specific in
pinpointing a single submitter and, therefore, have the lowest coverage.

     Many of the options evaluated here do not specify exactly how many people
would be involved in the notification process or whom these parties would

-------
                                     -154-
be.  It is difficult therefore,  to estimate coverage as well as costs.   Under
any of the options, the behavior of the parties involved cannot be predicted
with certainty.  For example,  under option 1 the submitter is the consignee of
the imported chemical.   However, if the consignee is a broker and does  not
receive assistance from the customer in filling out the form, the broker will
refuse to handle new imported  chemicals.   The consignee/submitter would then
become someone more knowledgeable about the chemical,  and option 1 might
resemble option 4 or 5.

     In terms of the actual entities or people involved in import
transactions, it is useful to  characterize participants as follows:

     •    brokers who clear merchandise through Customs.   Some
          brokers act also as  freight forwarders,  arranging for
          transportation and insurance of the merchandise;

     •    trading companies (including manufacturers'  repre-
          sentatives)  generally  of small size and varying
          familiarity with the products they handle; and

     •    domestic chemical companies who may import new chemicals
          for use in manufacturing and processing,  or to distribute
          them to customers.

     The primary distinction among these entities in terms of reporting
requirements is between importers that act as expediters or agents of the
import process, and importers  that intend to process,  use, or distribute the
imported chemical themselves.  The latter presumably would have the technical
knowledge to submit a section  5  notice.  The former type of importer could not
submit an adequate notification; rarely,  in fact,  would this type of importer
even be aware or able to determine that chemical shipments in a consignment
are "new."

     Methods of import vary.  Most of the largest chemical companies have
separate importing departments and act as their own importers.  Other large
companies employ brokers to handle the imported goods.  If they are regular
importers, they likely will have established relationships with a custom house
broker.  Data were not available on what proportion of companies clear  their
own merchandise compared to those that use brokers.  Cost factors, the  degree
of oversight of merchandise entering the country,  and the distance from the
port of entry to the company plant, may all be factors in the decision  to use
a broker.

     Factual information is not  available for small and medium-sized domestic
chemical companies.  They may  not have importing experience and often may rely
on larger chemical companies to  import the chemicals they need.  Occasional
importers are likely to use brokers.

-------
                                       -155-


      Trading  companies  fall  somewhere  between  brokers and chemical companies
 in  terms  of the  amount  of  knowledge  they  would have  about a new chemical
 import.   Trading companies are  usually small companies  that specialize in
 buying and selling  goods.  At times  they  match up  known buyers and sellers, at
 times they find  either  a product or  a  market.   Trading  companies do not use
 the chemicals they  import, nor  would they usually  develop a wide distribution
 system for selling  a chemical.  The  extent of  their  knowledge about a chemical
 would vary,   it  is  possible, for example,  that in  some  cases they would know
 the chemical  identity of an  imported substance, while in other cases they
 might not.

      As far as new  chemicals are concerned, it is  unlikely that a trading
 company would be involved  unless it  had firm customer commitments.  Trading
 companies are usually high-turnover, low  inventory operations, and would not
 be  able to afford to import  and warehouse new  chemicals while searching for
 markets among domestic  companies.

      If a trading company  does  have  customers  for  the new chemical,
 responsibility for  submitting notification to  EPA  would depend on the
 definition of importer  and the  circumstances of the  transaction.
 COSTS TO  INDUSTRY

 Costs of  Notice Submission

     The  notice form to be submitted under each of the five options is assumed
 to be the October 16th reproposed form.  The costs of completing the form are
 estimated at $900 to $8900 by  ICF on the basis of ADL's costs estimates for
 other firms.   (The domestic manufacturers' form is estimated to cost $1200 to
 $8900; in some cases the importer will omit the section on exposure at
 industrial sites, estimated to cost $300 to $2100 to complete.  Therefore, the
 lowest estimate will be $900 and the highest will be the same as for domestic
manufacturers, $8,900).

     The  out-of-pocket cost of submitting the form is assumed to be equivalent
under each option, although increased costs of gathering the information may
 shift the cost towards the high end of the estimated range for options 2 and
 3.  Similarly, under options 2 and 3, delays may occur in preparing the form
because of the need to contact other parties.  Bargaining costs may also be
incurred  as importers negotiate assistance from more knowledgeable parties.
Uncertainty costs may be present under option 5, but are likely to be larger
under options 2 and 4, where the most knowledgeable persons or the person who
knowingly caused importation may be hard to identify.  Uncertainty costs are
highest for option 3, where parties may not know all the persons who are
responsible for submission, and where some parties may be unwittingly exposed
to liability for failure to submit notification.

     Finally, there is the possibility of a continuing cost to importers of
loss of business if domestic companies reduce their level of import demand.
However,   this potential cost would not likely be significantly affected by the
choice of importer definition from among the five options.

-------
                                      -156-


Post-Submission Costs

     Potential costs which may be incurred following notice submission include
the possibility of out-of-pocket costs and delays due to supplementary
reporting requirements, section 5(c) and 5(e)  actions, and trade secret
disclosures.  The first two sources of additional costs are affected by the
choice of importer definition to the extent that additional data either
existing, 5(c), or to be created, 5(e), is needed.  It is possible that
notices submitted under option 1 will be inadequate, although option 5 notices
will contain better information.  However, it is difficult to predict the
relative consequences of choosing among the various options.

     Trade secret disclosures may be more likely to occur when several parties
are submitting information, particularly when the submission is not well
coordinated.  The various parties who may provide information directly to EPA
and the party who actually submits the notice may not be sufficiently aware of
all the confidentiality concerns of all parties involved.  This situation is
most likely to arise under options 2 and 3 where there may be some uncertainty
about which parties are involved and where a high degree of participation may
be necessary to produce an adequate notice.
SUMMARY

     Significant differences among the five options discussed occur mainly in
the extent of responsibility for notice submission and the costs of notice
submission.  These costs would weigh most heavily on less knowledgeable
individuals who would have to obtain the information required on the form from
others or negotiate for assistance from other parties.  Although some
adjustments may occur in methods of importation because of section 5
notification requirements, they are not likely to be significantly affected by
the choice of importer definition.

-------
                                       -157-


                                    CHAPTER 4

               IMPORTER CONTACT OF FOREIGN MANUFACTURERS/SUPPLIERS
     Because  of  the  nature  of  the  import  business and the limited information
 that is  available  to some importers on new chemical substances, an adequate
 section  5  notice in  many cases will require  information that can only be
 supplied by the  foreign manufacturer of a new chemical substance.  In some
 cases  it is expected that the  foreign manufacturer/supplier will assist in the
 preparation of the notice form, however this is not likely to be true in
 general.   The various types of importers will usually have different types of
 relationships with their foreign manufacturers/suppliers—their "upstream"
 contacts.
UPSTREAM CONTACT OPTIONS

     Two issues are being considered by EPA with regard to obtaining informa-
tion from foreign manufacturers and suppliers about new chemical substances
being  imported into the United States.  They are:

     •    whether mandatory contact with foreign manufacturers and
          suppliers should be imposed on importers, and

     •    the extent of information that should be requested of
          upstream contacts.

     The following four alternatives are being evaluated:

          Option 1; mandatory contact and the Foreign
          Manufacturers/Suppliers Form outlined in the January 10
          proposal;

          Option 2; mandatory contact and a Foreign Manufacturers/
          Supplier Form revised to be consistent with the forms in
          the October 16 reproposal;35/

          Option 3; mandatory contact and the request of only health
          and environmental effects data and risk assessments from
          foreign manufacturers/suppliers;

          Option 4; no mandatory contact by importer.
           forro outlined in the January 10 proposal includes the following
information which would be omitted in the revised form:  names and addresses
of related companies; some production and marketing data; chemical use
assumptions; information about foreign restrictions or bans on the production
of the chemical; and a detailed outline of submitted risk assessment data.
None of these questions are included on the October 16 reproposal of the
domestic manufacturers' form.

-------
                                      -158-
Evaluation Framework

     An evaluation of costs to industry of each option is presented in this
section using option 4, which does not impose mandatory contact, as a baseline.

     The direct costs to industry include costs to parties involved as
importers as well as costs to foreign manufacturers.  It should be noted that
although importers would be required to contact the foreign manufacturer and
supplier of the new imported chemical substance, the cooperation of the
foreign manufacturer or supplier is entirely voluntary.

     The direct costs to the industry are seen as the following:

     •    costs to importers of contacting and requesting
          information from foreign manufacturers/suppliers; and

     •    costs to foreign manufacturers/suppliers, including
          out-of-pocket costs of submitting information, and the
          possibility of trade secret disclosure.

     Exhibit 4-1 presents a chart of the relative costs to industry of each
option.
                                   EXHIBIT 4-1

                                COSTS  TO  INDUSTRY


                                 Option 1    Option 2    Option 3    Option 4

    Costs to Importers               111             0

    Costs to Foreign
      Manufacturers/Suppliers;
Submission
of
Informations/
$3
11
r
,
000-
600
$1
7
,800-
,200
$300-
1,500
                                                                         $0
                                    11,600       7,200       1,500

    Trade Secret Disclosure             3            211
a/See Exhibit 4-3 below.

Note:  For ordinal estimates, 0 = no cost, 1 = lowest cost, 2 = next lowest
       cost, etc.

-------
                                      -159-
COVERAGE

    The analysis of this issue is based on the assumption that foreign
manufacturers/suppliers will be from Japan or the European Community where
there are new chemical notification programs and where, consequently, much of
the information requested would be incremental to existing, available
information.  Before proceeding to the analysis, it is worthwhile discussing
this assumption in greater detail.

Sources of Imports

    Exhibit 4-2 shows the percentage of total value of chemical imports by
foreign origin for all chemicals and organic chemicals for 1975-1978.  As can
be seen in the final column of the table, Japan and the European Economic
Community countries account for approximately half the annual imports of all
chemicals, and about 73 percent of annual imports of organic chemicals.  In
terms of all chemicals, Canada remains the U.S.'s largest single-country
trading partner.  The percentage of imports of organic chemicals from Canada
is roughly the same as from West Germany or Britain, and about five percent
less than Japan.  There would be a serious error in the assumption used here,
if Canada, or the other countries outside Japan and the EEC which account for
15-25 percent of imports, were major sources of new chemicals.  However,
although it is difficult to document, it is widely agreed among participants
in this segment of the industry that most new chemical imports at present
originate in Western Europe and Japan.
                                   EXHIBIT 4-2

           PERCENT OF TOTAL VALUE OF IMPORTS (F.A.S.  VALUE),  1975-1978
Chemicals

1975 all chemicals
     organic

1976 all chemicals
     organic

1977 all chemicals
     organic

1978 all chemicals
     organic

Canada
23.5
7.4
27.0
10.9
26.7
13.8
West
Germany
N/A
N/A
N/A
N/A
11.0
14.2

EEC
40.2
57.2
40.6
55.7
40.3
56.2

UK
8.4
9.0
9.9
9.8
10.2
9.6

Japan
10.5
22.5
8.3
17.7
8.6
16.6

Other
25.8
12.9
24.1
15.7
24.4
13.4
28.1
12.5
12.4
13.9
43.4
54.8
11.4
11.8
 8.7
16.3
19.8
16.4
Source:   U.S. Department of Commerce, Bureau of the Census, Highlights of
          Exports and Imports, Report FT 990  (Washington, D.-C.:  Government
          Printing Office, December 1975, 1976, 1977, 1978).

-------
                                      -160-
Foreign New Chemical Notification Programs

     The assumption used here is that the Foreign Manufacturers/Suppliers Form
would be requesting information generally available and previously submitted
in foreign notification programs.  Although most foreign notification programs
(with the exception of Japan) are pre-market rather than pre-manufacture
programs, and although notification under the European Directive is not due to
take effect until March, 1982, in most cases the information requirements are
or will be similar..367  Testing requirements in foreign countries may
increase the amount of useful information about new chemicals that could be
available from foreign manufacturers.  However, small volume chemicals (under
one ton annual production)  do not require notification under EEC guidelines.

Costs to Industry

     An assessment of the costs to industry imposed by any of the four options
is made difficult by the fact that cooperation on the part of foreign
manufacturers/suppliers is voluntary.  Even an assessment of the relative
costs of each option must depend on some assumptions about the behavior of
industry parties.  For example, as shown in Exhibit 1, the cost to a foreign
manufacturer of completing the January 10th Foreign Manufacturer/Supplier Form
(option 1)  is much greater than submission of only health and safety data
(option 3).  However, a foreign manufacturer, under option 1, is free to
submit only health and safety data and not incur the larger cost.  It is
difficult to predict, however, whether the foreign manufacturer would submit
more information if asked for it, or whether the more onerous request would
discourage submission of some or possibly any information.

Costs to Importers

     Under the January 10th rule provisions regarding mandatory contact by
importers,  the direct cost to importers involves only the written request of
information from the foreign manufacturer/supplier.  In cases where the
importer does not know who the manufacturer is, he must request the supplier
to provide the identity of the manufacturer.  Foreign manufacturers and
suppliers are not required to cooperate.  But the information they supply
would reduce the likelihood of EPA imposing supplementary reporting
requirements.  The direct costs to the importer of contacting the foreign
manufacturer/supplier and certifying to that contact are negligible.   These
rule provisions are applicable to options 1, 2, and 3.
     11/EEC Council Directive of 18 September 1979,  printed in Official
Journal of the European Communities, L259, Vol.  22,  15 October 1979.   See
also:Mitre Corp., Information Required for Regulation of Toxic Substances,
Vol. 1, June 1978 (NTIS PB-288 023).

-------
                                      -161-
     Indirect costs to importers could involve foregoing possible business
because of the expectation that the foreign manufacturer will not cooperate
and adverse action will be taken against the chemical.  They could also
involve the loss of import business as a result of concern by the foreign
manufacturer of trade secret disclosure.  This concern is discussed below in
greater detail.  However, there appear to be no significant differences among
the four options in the extent to which these indirect costs are affected.
These indirect costs, which represent the possibility of loss of business, are
present whether or not mandatory contact is imposed and whatever amount of
information is requested of the foreign manufacturer.

Costs to Foreign Manufacturers/Suppliers

     Direct out-of-pocket costs are incurred by foreign manufacturers/suppliers
if they elect to submit the requested information.  These costs vary across
the four options and are outlined below in Exhibit 4-3.  The costs shown
represent the estimated costs of completing the entire notification request.
However, these costs are based on estimates for similar forms submitted by
domestic manufacturers.  Foreign manufacturers may already have much of the
information requested available because of prior submission to a foreign new
chemical notification program.  If they do, the direct costs shown here may be
overestimated.

     With regard to option 4, no mandatory contact by the importer, it is not
possible to attribute any one cost figure to the foreign manufacturer/supplier.
Certainly, in some cases, the foreign manufacturer/supplier will be asked by
the importer to provide information, and in other cases, the foreign
manufacturer may even assist in the preparation of the importer's form.  Under
this option, as well as the other three, submission of insufficient
information increases the risk of added costs and delays,  importers and
foreign manufacturers will calculate these risks and incur costs of submitting
information depending on the expected return from the new chemical substance.
For some products, for example, the U.S. market may represent the most
significant area of marketing opportunities, and the costs of providing all
requested information to EPA would be readily borne by the foreign
manufacturer.

     It would not be feasible to attempt to estimate what portion of new
chemical imports might be affected by these direct costs.  Indeed, the dollar
cost of submitting information may not be the significant burden on foreign
companies exporting to the U.S.  Rather, the confidentiality issue could be
the more powerful inhibitory factor.  There are two aspects to this issue: the
first is the general confidentiality issue of how confidential business
information will be treated under TSCA.   This issue is discussed in a separate
document.  Essentially, the burden will be on submitters to assert and
substantiate confidential business information.  Even for some confidential
business information asserted and substantiated as confidential, a "masked"
disclosure of the confidential business information will be required.  In
addition, Japan and Canada treat all submitted data as confidential, so that
fear of trade secret disclosure might inhibit imports of new chemicals from
those countries.

-------
                                                                          EXHIBIT 4-3

                                                              COSTS OF SUBMISSION OF INFORMATION
 ption 1.  Foreign Manufacturer/Supplier Form, January 10, 1979.
    Section of Foreign
Manufacturer/Supplier Form

I.A, I.B
          Information Category
  Comparable  Section of
    Proposed.  Domestic
    Manufacturer Form
(Source of  Required Hours)
                                                                                                                  Hours
General Information and Chemical Identity  I.A minus items 5c,  6,
                                           and I.B
 Clerical
($10/hour)

 3-5
Technical
($25/hour)

 4-8
Managerial
($50/hour)

 1-1
    Cost
(in  dollars)

 180 - 300
I.C
I.D
II. A
II. B
II. C
Production and Marketing Data I.C
Federal Register Notice I.D
Risk Assessment Data II. A
Risk Analysis III. A
Structure/Activity Relationships III.B
2 - 62/ 4 - 24S/ 2 - 6S/
2-4 4-16 2-4
10 - 80 16 - 120 8-20
10 - 20 10 - 80 10 - 20
10-20 10-40 5 - 10
220 - 960
220 - 640
900 - 4,800
850 - 3,200
600 - 1,700
                                                                                                                               Total Cost   S2.97Q - 11.600
 /     Production and marketing data are requested in section I-C of the foreign form.  ADL's hour estimates for section I-C of the domestic form were con-
  sidered in estimating the hours for section I-C of the foreign form.  However, from observation of the two comparable sections, it is clear that they are
  substantially different.  Although information requested in the domestic form's section is also requested in the foreign form's section, the foreign form
  requires additional information not required in the domestic form.  The domestic form only involves the reporting of new chemicals (or chemical processes)
  but the foreign manufacturer/supplier form addresses either of two cases:  (1) the reporting of new chemicals or (2) the reporting of chemicals intended
  for import into the United States which have previously been manufactured abroad but have never been introduced in the United States.  In case 1, the
  information requirements are the same as those for the domestic form, and so ADL's hour estimates are applicable.  In case 2, however, the information of
  case 1 is required plus additional information on past marketing and use abroad.

       The minimum hours that ADL attributed to production and marketing data for the domestic form were two for clerical, four for technical, and two for
  managerial labor.  Because the requirements of this section of the domestic form are the same as those under case 1 for the foreign form, and since case 1
  requires less information than case 2, these hours were used to represent the minimum hours for the foreign form's section on production and marketing
  data.   The maximum hours attributed by ADL to the domestic form's section were:  four for clerical, 16 for technical, and four for managerial.  Yet, any
  estimate of the maximum hours for the foreign form's section must exceed these hours to reflect the additional information required under case 2.  ICF
  estimated that this additional information requires an approximate 50 percent increase over the labor necessary for completing the comparable section of
  the domestic form.  Thus, maximum hours for this section were increased by 50 percent over those for the domestic form.

-------
                                                                    EXHIBIT 4-3 (continued)

                                                              COSTS OF SUBMISSION OF INFORMATION
Option 2.  Revised Information Requests of Foreign Manufacturers/Suppliers.
    Sections of Forms
      Required Under
	This Option

Revised Form:  I.A
minus items 4 and 5
          Information Category
General Information
                                             Comparable Section of
                                               Proposed Domestic
                                               Manufacturer Form
                                                                                                                   Hours
                                                                         Clerical
                                                                                     Technical
(Source of Required Hours)    ($10/hour)    (325/hour)
                                           Revised Domestic
                                           Form:  I.A
                                                                                                     0 -
                                                                                                                   0-0
Managerial
($50/hour)
                                                                                                                                 1  -
                                                                                                                                                  Cost
                                                                                                                                               (in dollars)
                                                                                                                    .50  - 410
Revised Form:  I.B
Chemical Identity
                                                                       Revised Domestic Form: I.B    0 - 2£/      2-13
                                                                                                                                                 50  -  345
 Revised Form:   I.C
                            Generic Names
                                           Revised Domestic Form: I.C    0 - l£/      0-4
                                                                                                                                 0-1
                                                                                                                                                  0  -  175
January 10 Form:  I.C
minus items 3, 4(c),
5(a) (iii) , 6 ,  7, and 8,
plus Revised Form:
I.D, item 2 only
                            Production and Marketing  Data
                                           January 10 Domestic Form:
                                           I.C minus items 3, 4 (c) ,
                                           5(a) (iii) , 6, 7, and 8,
                                           plus Revised Domestic
                                           Form: I.D, item 2 only
                                                                                                       - 2£/
                                           3 - 16£/
 1 - 3d/
                                                                                                                                                135  -  570
Revised Form:
IV.A, IV.C, and IV.D
Federal Register Notice
Revised Domestic
Form:  Part; IV
                                                                         1-2
                                                                                      1-8
                                                                                                    1-2
                                                                                                                    85  - 320
                                                                                                                                                              Co
                                                                                                                                                              I
Revised Form:  I.F
                             Risk Assesment
                                                                        Revised Domestic Form:  I.F    0 - 2£/      0-16
                                                                                                                                0-2
                                                                                                                                                  0  -  520
January 10 Form:
II.B and II.C
Risk Analysis, Structure/
Activity Relationships
January 10 Domestic
Form:  I.B and II.C
                                                                                                    20 - 40
                                                                                                                 20 - 120
                                                                                                                               15 - 30
                                                                                                                                             1,450  -  4,900
                                                                                                                               Total Cost:  $1,770  -  7.240
b/
       From  "Regulatory Alternatives,'  EPA  Working  Paper,  February  1980,  pp.  38-39.
£/     ADL did not provide  estimates  of  clerical  hours  for  these  sections.   ICF  estimated  these hours by using ratios of clerical to professional hours.
  From ADL's  time estimates for  the reproposed  domestic manufacturers  form,  the  ratio  of total minimum clerical hours to total minimum professional  hours  is
  0.23, and the  ratio  of  total maximum clerical hours to total  maximum professional  hours  is  0.13.  The minimum professional hours  (technical  anfl Tuanaqe-
  rial) in each  row  of the  table were multiplied  by 0.23 to yield an estimate  of the minimum  clerical hours.  These values were  rounded  to  the nearest  hour.

£•/     The information required  under the  Production and Marketing Data category is  information requested  in various discrete items of the  January 10 and
  revised forms.  As a result,  there  are no ADL time estimates  for this information.   Rather, ADL estimated labor requirements for the larger  cateaories that
  contain these  discrete  items.   ICF  estimated  the labor requirements  by considering each  item's required  labor as a fraction of ADL's labor estimations for
  the larger  category  containing that item.  For  the items  in  the January 10 foreign manufacturer/supplier form  (section T.C minus items 3, A (c), S(a)(iii),
  and 6-8) , ICF  estimated two to eight technical  and one to two managerial hours.  For the reproposed form's item 2 of  section I.D, ADL  had orovided an
  estimate for technical  hours of one to eight, but did not provide an estimate  for  managerial hours other than a single estimate for the entire  section.
  The managerial hours for  this  item  were  estimatd by ICF to be zero to one.  Clerical hours  for the Production and Marketing Data cateqorv were  estimated
  using ratios of clerical to professional hours   (see footnote  c/).

-------
                                     -164-
                                  EXHIBIT  4-3
                                   (continued)

                       COSTS OF SUBMISSION OF INFORMATION
Option 3.  Health and Environmental Effects Data;  Request of Foreign
           Manufacturers/Suppliers to Complete Section III.B of the
           Reproposed Domestic Manufacturer Form
        Hours:

Cost Per Hour:

         Cost:
 Clerical

 2 - 6£/

   $10

$20 - 60
  Technical

   8-40

     $25

$200 - 1,000

      Total Cost:
 Managerial

   2-8

    $50

 100 - 400

$320 - 1.460
£/     ADL did not provide estimates of clerical hours for these sections.   ICF
  estimated these hours by using ratios of clerical to professional hours.
  From AOL's time estimates for the reproposed domestic manufacturers form,
  the ratio of total minimum clerical hours to total minimum professional
  hours is 0.23, and the ratio of total maximum clerical hours to total
  maximum professional hours is 0.13.  The minimum professional hours
  (technical and managerial) in each row of the table were multiplied by 0.23
  to yield an estimate of the minimum clerical hours.  These values were
  rounded to the nearest hour.
Sources: Arthur D. Little, Inc., Estimated Costs for Preparation and
         Submission of Reproposed Premanufacture Notice Form,  prepared for the
         EPA Office of Toxic Substances, September 1979.

         Arthur D. Little, Inc., Impact of TSCA Proposed Premanufacturing
         Notification Requirements, prepared for EPA Office of Planning and
         Evaluation, December 1978.

-------
                                      -165-
                                    CHAPTER 5

                                    EXPORTERS
     This  issue pertains to the manufacture in the U.S. of new chemicals for
the sole purpose of export.  The purpose of the information requested is to
enable EPA to determine whether unreasonable risk is posed by these new
chemicals  during manufacture and transport in the U.S.
EXPORTER OPTIONS

     Three alternatives evaluated by the contractor concerning new chemicals
manufactured solely for export are the following:

          Option 1;  Exporters are required to submit a notice form
          under section 5 of TSCA.  The form would be similar to the
          exporters form in the October 16th reproposal, but would
          not include information about uses and exposures occurring
          outside the U.S.

          Option 2;  Exporters would be exempt from section 5
          authority but would be subject to reporting requirements
          under section 8(a) of TSCA.  As defined by EPA, this
          option would require exporters to submit a notice form at
          least 90 days before commencing manufacture of a new
          chemical substance.  The form again would be similar to
          the October 16th exporters form, excluding the section on
          uses and exposures outside the U.S.  This option would
          also include a small business exemption for companies with
          an annual income under $1 million.

          Option 3;  Exporters would be exempt from any reporting
          requirements for new chemicals manufactured solely for
          export.

Evaluation Framework

     In this section,  the direct costs to industry of each option are
evaluated,  using option 3 as a baseline from which to assess the other
options.  Option 3 exempts manufacturers of chemicals solely for export from
any reporting requirements and therefore poses no costs to the industry.

     The direct costs to the industry are seen as the following:

     •    Costs of Notice Submission,  including out-of-pocket costs
          of filling out notice form,  and the delay in introduction
          of new chemical;  and

-------
                                      -166-
     •    Post-submission Costs, including costs, delay  (5(c)), and
          uncertainty involved in possible supplementary reporting
          requirements, section 5(e) and 5(f) actions.

     Exhibit 5-1 shows a matrix of the options and their relative costs.  It
is discussed more fully below.
                                   EXHIBIT 5-1

                                COSTS  TO  INDUSTRY


                                   Option 1          Option 2          Option 3

Coverage                              210

Costs of Submission                   110


Potential Post-Submission Costs

Supplementary Reporting               110

Uncertainty                           100

Delay in introduction of              1                 00
  chemicals  (5(c) extension of
  Notice Period)

Sections 5(e) and 5(f) action         100

Action under other TSCA authority     111


For ordinal estimates, 0 = no cost, 1 = lowest cost, 2 = next lowest cost.



COVERAGE

     The number of people affected by reporting requirements for chemicals
manufactured solely for export varies for the different options.  Under option
3, manufacturers would be entirely exempt from reporting these chemicals.
Option 1 entails the widest coverage of exporters, differing from option 2 in
that option 2 includes a small business exemption.

     The small business exemption under option 2 would exempt exporters with
total annual sales less than $1 million from reporting requirements,  it is
difficult to estimate precisely how much this provision would reduce coverage

-------
                                      -167-
of exporters.   (Data from the Census of Manufactures are too highly aggregated
to permit an accurate assessment.)  However, most small firms appear to rely
heavily on close marketing relationships with their customers, so it is not
likely they would be heavily involved in manufacture solely for export.  The
difference in coverage between options 1 and 2 is therefore not expected to be
large.

     It is difficult to estimate the size of the market for chemicals made
solely for export.  Most such chemicals would be specialty products and many
would be custom made for particular customers.  The reasons for manufacturing
a new chemical solely for export can range from differences in U.S. and
foreign technology to a foreign firm being the only known customer.  Examples
cited by companies include (1) different uses or applications for chemicals in
Europe and the U.S., and (2)  different performance standards or environmental
and health standards for chemicals in the U.S. and some underdeveloped
countries.

     None of the companies contacted who manufacture chemicals solely for
export operate in the second category.  However, there may be companies that
specialize in chemicals for foreign markets.  In most firms, solely-for-export
chemicals would not represent a large proportion of total business.  Large
chemical companies would likely manufacture a new chemical abroad if a major
and sole foreign market were identified.


Costs to Exporters

     Costs of Notice Submission.  These costs, incurred by exporters under
options 1 and 2, involve completing the required reporting form.  The basis
for the cost estimates was the ADL work on estimating the costs for
preparation and submission of the October 16 domestic manufacturers form.37/
The minimum exporters form cost is the minimum domestic manufacturers form
cost less than the minimum cost for section II-B (industrial sites controlled
by others) and the minimum cost for section II-C (consumer and commercial user
exposure).  Because the minimum cost of II-B and II-C are both zero, the
minimum exporter's form cost is estimated to be $1,200.  The maximum
exporter's cost, $8,100, is the maximum domestic manufacturers form cost plus
the cost of completing section II-C.

     Both option 1 and option 2 would require the same form to be used; the
out-of-pocket costs of notice submission would therefore be equivalent.
Likewise,  under these two options, the forms would have to be submitted at
least 90 days in advance of manufacture.  Thus, any delays in the introduction
of a new chemical due to this requirement would be equivalent for options 1
and 2.
         rthur D. Little, Inc. Estimated Costs for Preparation and Submission
of Reproposed Premanufacture Notice Form (Cambridge, Mass.:  September 1979),
p. 38.

-------
                                      -168-
     Post-Submission Costs.  The primary difference between options 1 and 2 is
the statutory authority covering exporters.  Under option 1, exporters would
be subject to section 5 regulations, including the possibilities of an
extension of the notice period, regulation pending development of additional,
not previously existing information (section 5(e)  action), and regulation for
protection against unreasonable risks (section 5(f) action).  Under option 2
(and option 3), exporters would not be subject to actions under sections 5(e)
and 5(f).  However, under any of the three options, chemicals could be subject
to action under sections 4, 6 and 7 of TSCA.  Both options 1 and 2 would allow
EPA to require supplementary reporting of information on a new chemical
substance.  The costs associated with this possibility would be the same under
both options.

     It should be noted that the post-submission costs to exporters are not
fixed costs  (such as the costs of notice submission).  The magnitude of these
costs would depend on the probability or frequency that an extension of the
notice period or restrictive actions would be taken.   Furthermore, the
specific costs to the exporters involved in such restrictive actions would
depend on a variety of factors, such as the cost of providing information
requested by EPA, the extent to which an exporter appeals an Agency action,
etc.  These factors and the costs associated with them will vary on a
case-by-case basis for each chemical.   However, to the extent that action
under section 5 is more likely than action under other sections of TSCA, the
uncertainty over post-submission costs for a new chemical will be higher for
exporters under option 1.
SUMMARY

     Broadly considered, options 1 and 2 entail an equivalent cost to
exporters.  They differ in the authority EPA has to take action against new
chemicals once a notice has been submitted.  That difference would raise the
cost to exporters under option 1 if such actions are taken and would increase
uncertainty for exporters submitting notification.  Both options 1 and 2 allow
for the possibility that supplementary reporting will be required.  Option 3
provides no information to EPA on new chemicals manufactured solely for export
and imposes no costs on exporters.

-------
                                      -169-


                                   CHAPTER  6

                             SUPPLEMENTAL REPORTING


     In assessing a new chemical substance,  EPA may require information other
than that contained in the initial notice submission.  In the following
sections, we will:

     •    specify the options under consideration;

     •    specify the types of costs incurred by industry; and

     •    estimate the relative burden to industry of each type of
          cost.


SUPPLEMENTAL REPORTING OPTIONS

     We have analyzed three alternatives to accomplish this task.   They are:

     •    Option 1;  January 10 proposal.  Under option 1, EPA may
          require a submitter or processor of a chemical substance
          for which a notice has been submitted to report the
          following types of information:

               information which clarifies and supplements the
               information requested on the  form;
               information concerning the benefits of substances for
               various uses and the availability of substitutes for
               those uses; and
          —   information concerning the economic consequences of
               any specified regulation under the Act.

          This information can be requested  if the Agency believes
          that the information would be relevant to determine
          whether EPA should require the substance to be tested
          under section 4 of the Act, control the substance under
          section 5(e), 5(f), or 6(a), or follow-up on the substance
          under section 5(a)(2)  or 8(a).  EPA may also require the
          manufacturer or importer of an unknown reactant to report
          its identity or composition if the submitter demonstrates
          that he has attempted to obtain information concerning the
          identity of the reactant from the  manufacturer or
          importer.  In addition, the Agency may require any person
          who has possession of a health and safety study to submit
          it ". . . if the Agency believes that the study would
          assist in the evaluation of the health or environmental
          effects of the manufacture, processing, distribution in
          commerce, use, or disposal of a new chemical substance for
          which the Agency received a premanufacture notice".38/
     38/44 Federal Register 2278.

-------
                                      -170-
          Th e information that can be requested and the circum-
          stances under which the information can be requested are
          quite broad.

     •    Option 2;  October 16 proposal.   The most important
          difference between option 1 and  option 2 is that option 2
          provides a more precise delineation of the information
          which may be  required and the circumstances under which it
          may be required.  In place of the broad definitions of the
          types of information which may be required, EPA provides a
          more detailed list of information grouped into the
          categories of manufacturing,  processing or industrial use,
          distribution  in commerce, disposal, and end use.  The
          criteria which must be satisfied before information can be
          requested are also more specific.  For example, to obtain
          information concerning the risk  to human health and the
          environment resulting from environmental release, the
          Agency must find either that:  (1)  the substance may
          present a significant hazard to  human health or the
          environment,  and has a potential for release which would
          result in human or environmental exposure, or (2) data
          sufficient to determine toxicity are not available, and
          the substance has a potential for significant release
          resulting in  human or environmental exposure.  The types
          of health and safety studies and the circumstances under
          which their submission may be required, as well as the
          provisions which deal with obtaining information from
          manufacturers and importers of unknown chemicals, are no
          different from option 1.  In addition, option 2 provides
          for an appeal to the Assistant Administrator or Deputy
          Assistant Administrator for Pesticides and Toxic
          Substances and limits the period during which information
          can be obtained to the review period; under option 1, the
          period during which supplemental reports could have been
          required was  not specified.

     •    Option 3;  Delete supplemental reporting rule.  The Agency
          may still request information after the review period has
          ended under section 8(a) of the  Act.  If the agency has
          insufficient  information to determine the risk of the
          chemical, it  can regulate and obtain additional, not
          previously existing information  about the chemical under
          section 5(e)   (if the other requirements of that section
          are met) .

    Under each of the options, information could be provided voluntarily.
Only if the information were not provided  voluntarily would the formal
supplemental reporting  provisions be used.

EVALUATION FRAMEWORK

    The costs of the supplemental reporting provisions to the chemical industry
depend on the regulatory option which is chosen, the way in which EPA
implements the regulations, the manner in  which industry chooses to respond to

-------
                                       -171-
the regulations, the nature of other section 5 rules, and the nature of the
specific chemical market.  The sensitivity of the costs to the discretionary
behavior of both regulators and the regulated—behavior which is not controlled
by the provisions of the Act—necessarily reduces the precision with which
costs can be estimated.

     The uncertainty about the future behavior of the industry and EPA means
that we can only estimate the relative burden of each option to industry under
each type of cost.  The types of costs and the relative burdens attached to
them are listed in Exhibit 6-1.  The explanation for the entries in the matrix
is given in the following section.
                                   EXHIBIT 6-1

          RELATIVE COSTS OF SUPPLEMENTAL REPORTING OPTIONS 1 THROUGH 3


              Option:                      L            —            —

        Cost to Industry

        Out-of-Pocket Costs                3            2            a/

        Delays                             2            1            a/

        Section 5(e)  Action                1            1            £/

        Uncertainty                        321
        1 = lowest cost, 2 and 3 = next lowest costs, respectively.

        fL/These costs are heavily dependent on the actions of the
          submitters.  Although we cannot be sure of the relative
          magnitude of each cost component under option 3, we do know
          that the total cost of all components will be lower under
          option 3 than any of the other options.
Out-of-Pocket Costs

     The out-of-pocket costs include the administrative, clerical, and techni-
cal time needed to supply the information requested.  This cost depends heavily
on the amount of information that is requested and the frequency with which
requests are made; EPA has considerable latitude to vary the cost to itself
and the industry.  There are upper limits placed on the cost, however.  A per-
son receiving a request need only submit information that is known or reason-
ably ascertainable; "we don't know" is an acceptable response to a request for

-------
                                       -172-
infonnation.  The frequency is limited under each option to those cases which
satisfy the criteria, but within such boundaries EPA can decide in each case
whether or not to require information.  The frequency with which EPA initiates
a formal request is in turn dependent on the degree to which the industry
voluntarily cooperates with requests for information.

     Because the criteria for imposition of a supplemental reporting
requirement under option 1 are broader than under option 2, there will be at
least some cases in which information would not be available to EPA during the
PMN period under option 2, but would be available during the PMN period under
option 1.  The appeal procedure under option 2 would further reduce the amount
of information EPA could receive under option 2.  The information delivered
under option 3 will be less than under option 2, since there will be at least
some cases in which the submitter will choose not to provide information that
is requested.  Therefore, the out-of-pocket costs will be greatest under
option 1 and least under option 3.  The magnitude of those costs is heavily
dependent on the behavior of the industry and EPA.  Therefore,  the available
information does not allow us to estimate the dollar amounts.

Delay in the Introduction of New Chemicals

     A common complaint from industry as expressed in the public comments on
previous proposals is that the notification process will result in delays in
the introduction of new chemicals.  Delays are costly to chemical producers
and could result in a decline in the number of new chemicals introduced.39/
The possible sources of delay under the supplemental reporting  provisions are:

     •    An extension of the notice period in order to allow more
          time to obtain supplemental information.  This should be
          relatively insignificant since in most cases supplemental
          reporting will be associated with chemicals under detailed
          review and for which the notice period will probably  have
          already been extended.4O/

     •    An extension of the notice period in order to allow more
          time to evaluate the information obtained from
          supplemental reporting.  This should also be a minor
          problem, again because of the association between
          supplemental reporting and detailed review.
     ^I/Comments of the Manufacturing Chemists Association on EPA Proposed
Regulations for Premanufacture Notification Under Section 5 of TSCA, March 26,
1979, pp. 154-5.

     M/44 Federal Register 59765 (October 16, 1979).  EPA anticipates that
it will be able to decide that most chemicals require no further regulation on
the basis of the information provided in the notice and a literature search
conducted by EPA staff.

-------
                                      -173-
     As  noted  in  the  previous  section,  there  will  be  less  supplemental  infor-
mation obtained by  EPA under option  2  than  under option  1.  Therefore,  there
will be  fewer  delays  and  less  cost of  delay under  option 2; however,  the dif-
ferences between  the  two  options  should be  minimal.   The relative extent of
delay under option  3,  compared to options 1 and 2,  is quite dependent on
industry behavior.  Therefore,  we cannot estimate  the extent and cost of delay.

Restrictions Under  Section  5{e)
     Under  option  3,  the  refusal  by  firms  to provide information voluntarily
may  result  in  restrictions  being  placed on use of the chemical until such
information is provided under  section  5(e) of the Act.  These restrictions may
result  in losses of income  and profits, and a reduction in chemical
innovation,  and should be considered therefore as costs to industry.

     The cost  to industry of section 5(e)  restrictions may be great, but it is
important to understand that section 5(e)  actions will occur in the absence of
any  supplemental reporting  rules.  Therefore, any cost of actions taken under
section 5{e) should not be  attributed  to the supplemental reporting
provisions,  but to the section as a  whole.

Uncertainty About  EPA's Implementation of  Supplemental Reporting

     In addition to the costs  which may be imposed on the industry because of
EPA's implementation  of whatever  option is eventually chosen, the uncertainty
about how EPA  will implement that option will itself impose costs on the
industry.   Because of EPA's wide  degree of latitude under options 1 and 2,
industry may take  a considerable  period of time to learn how the Agency will
behave.  In  so doing, industry may take actions that may result in unnecessary
costs and delays.  Costs  which are incurred as a result of such uncertainty
are  properly attributable to the  regulations.

     The major difference between option 1 and option 2 lies in this component
of industry  cost—uncertainty  regarding EPA's implementation of the rules.
Option  2 more  precisely defines the  information to be requested and the
circumstances  under which it will be requested than does option 1 under
different circumstances.  This  increased specificity should make the cost of
uncertainty  under  option  2 less than under option 1.

     The difference between option 3, on the one hand, and options 1, and 2, on
the  other, lies in the ability of the  submitter under option 3 to refuse
requests for supplemental information.   Because of this ability, the submitter
must incur lower expected costs under option 3 than under the other options.
The  submitter can guarantee he does no worse under option 3 by acceding to all
requests for information.   But  if the  submitter believes that the expected
costs of supplying a piece of  information  are greater than the expected costs
of not providing the  information  (e.g., the possibility of restrictions under
5(f), the costs of delays), the submitter will, of course, choose to withhold
the  information.  Therefore, under option  3,the expected costs to industry are

-------
                                     -174-
lower than under the other options..£!/ in effect, option 3 gives the
submitter greater control over its own fate and causes less uncertainty about
EPA's behavior than do options 1 and 2.42/
SUMMARY

     In comparing options 1 and 2,  we find the costs of option 2 to industry
are the lowest.  Compared to option 1,  option 2 reduces the uncertainty for
industry, and will allow somewhat less  information to be acquired by EPA,
resulting in less delay and lower out-of-pocket costs.   Both options 1 and 2
are more costly than option 3.   Under the first two options, the Agency can
require submitters to provide certain kinds of information, while under option
3, submitters may refuse requests for information.  Therefore, the submitter
must incur lower expected costs under option 3 than under the other options.
     .li/Even though expected costs are lower under option 3, actual costs
may be greater.  In the absence of information, the submitter's predictions
about EPA's actions may be quite poor, as may his predictions of the cost of
providing information.  The submitter's actual, as opposed to expected, costs
under option 3 will be lower only if his predictions are unbiased estimators
of the actual outcome.

     .ll/This same analysis applies to the appeal procedure which is
available under option 2 but not under option 1.  The submitter will incur
costs in making the appeal unless his expected benefits exceed his expected
costs.  Therefore, adding an appeal procedure provides a net benefit to
submitters.

-------
                                      -175-



                                   CHAPTER 7

                            INSUFFICIENT SUBMISSIONS
     Section 5 of the Toxic Substance Control Act (TSCA)  requires that a notice
contain certain information.  If submissions do not contain this information,
EPA would inform the submitter of the deficiencies and provide the mechanism
by which the submitter can correct the deficiencies.

     In the following sections we:

     •    discuss the options under consideration;

     •    specify the types of costs incurred by industry;  and

     •    estimate the relative burden to industry of each  type of
          cost.

INSUFFICIENT SUBMISSION OPTIONS

     We have analyzed four alternatives to accomplish this  task.   They are:

     •    Option 1;  The January 10 proposal.  The proposal divides
          deficiencies into major and minor categories.  In the case
          of a major deficiency, the notice is invalid and  the
          notice period does not start.  The 90-day review  period
          begins after the deficiency has been corrected.   In the
          case of a minor deficiency, the notice period is  suspended
          (the notice period clock stops running)  until the
          deficiency is corrected.  If the deficiency is not
          corrected within 30 days of the notice of deficiency, the
          notice may be declared invalid.  The notice can be
          declared invalid at any time during the notice review
          period, although findings of minor deficiencies can only
          be made during the first 30 days.  If EPA discovers at any
          time, even after the expiration of the notice period, that
          a submission includes intentionally false or misleading
          statements, EPA may find that the notice was invalid from
          the time of its initial submission.  In addition, any
          production following such a submission is in violation of
          the Act.  There are no procedures for appeal of agency
          decisions.

     •    Option 2;  Addition of Appeal Procedures.   Under  this
          option, a submission containing a major deficiency is an
          incomplete submission,  if the submission is incomplete,
          the notice period clock does not begin to run until the
          deficiency has been corrected.

          Unlike option 1, the submitter can appeal any finding that
          a submission is incomplete (within 10 days of notification

-------
                                     -176-
          of the finding),  and the Agency may determine the submis-
          sion is incomplete only within 30 days of receipt of the
          notice.  The criteria for determining that a submission is
          incomplete are narrower than the grounds for determining
          major deficiencies listed under option 1.  (See Exhibit
          7-1).  If the agency finds an error, the agency asks the
          submitter to correct the deficiency, but does not suspend
          the clock.

          If EPA discovers  at any time during the notification
          period that the submission includes intentionally false or
          misleading statements,  the Agency may find that a notice
          was never submitted and the submitter is in violation of
          the Act.  After the notification period, the Agency would
          prosecute intentionally false notices under federal laws
          concerning false  and misleading information rather than
          TSCA.

          Option 3;  Deletion of insufficient submission provisions
          from the rules.   EPA could still find a submission to be
          insufficient to meet the statutory requirements, but such
          findings would be made on a case-by-case basis; there
          would be no general rule delineating the deficiencies.
          Under this option EPA would handle errors in much the same
          way as under option 2.

          Option 4;  No insufficient notice provisions.  Unlike
          option 3, in this case EPA would not reject submissions.
          EPA would obtain  additional information by attempting to
          persuade the submitter to voluntarily provide it, by
          requesting further information under the supplemental
          reporting provisions, and/or by regulating the substance
          under section 5(e)  of the Act.  These actions may require
          an extension of the notice period.
EVALUATION FRAMEWORK

     We will discuss, in turn, each of the costs to industry.   The direct
costs to industry are the following:

     •    Delay in the Introduction of New Chemicals;
     •    Restrictive Actions Under Section 5(e);
     •    Out-of-pocket Cost; and
     •    Uncertainty.

     We will quantify costs to the extent permitted by the limited information
on record.  However, because the magnitude of each type of cost will vary
depending on the behavior of submitters and EPA, the nature of other section 5
rules, and the nature of specific chemical markets, it is not  reasonable to
specify precise burdens attributable to the insufficient submission provisions.

-------
                                  -177-



                               EXHIBIT 7-1

           CRITERIA FOR FINDING SUBMISSIONS  TO  BE  INSUFFICIENT


Option 1                     Common                        Option 2

                       Failure to sign the notice
                       form.

                       Failure by an importer to
                       comply with the procedures
                       for obtaining information
                       from foreign manufacturers
                       or suppliers, in accordance
                       with Section 720.21(c).

                       Failure to provide any infor-
                       mation requested on the
                       form or indicate that
                       it is not known or reasonbly
                       ascertainable.

                       Failure to provide any infor-
                       mation required by Sections
                       5(d)(1)(B)  and  (C)  of the
                       Act,  in accordance with
                       Section 720.23.

                       Failure of a notice to include
                       the test data or other infor-
                       mation which the submitter is
                       required to submit pursuant
                       to a rule promulgated under
                       Section 4 of the Act.

                       Failure to include the spe-
                       cific chemical identity  of the
                       substance for which the  notice
                       is submitted, unless it  is
                       impossible to do so in accor-
                       dance with Section 720.20(f).

                       If EPA has listed a chemical
                       substance under section  5(b)(4)
                       of the Act, failure to submit
                       data which the submitter
                       believes show that the chemical
                       substance will not present an
                       unreasonable risk of detriment
                       to health or the environment.

-------
                                       -178-
                                  EXHIBIT 7-1
                                   (continued)

                     CRITERIA FOR FINDING MAJOR DEFICIENCIES
    Option 1

Failure to remedy any
deficiency for which
EPA issued a request for
correction under para-
graph  (a) of this
section, within 30 days
following a person's
receipt of the request.

Submittal of inten-
tionally false or mis-
leading responses to
questions on the form.

Except as specifically
authorized by Sections
720.10{a) (3) and (b) (2) ,
submittal of a notice by
someone other than either
the person who intends to
manufacture or import the
chemical substance, or
his designated agent.

Failure to comply with
the procedures for ob-
taining information from
other persons, in accor-
dance with 720.20(e),
or to certify that the
procedures have been
complied with.
Common
   Option 2

Failure to use
appropriate notice
form.
             Source:   44  Federal  Register  2272-3 (January 10,  1979);
                     guidance received from EPA.

-------
                                       -179-
 It  is  reasonable,  however, to  specify  relative burdens of each option to
 industry under  each  type of cost.  The result will be a matrix of options and
 costs  which are shown  in Exhibit  7-2.
                                   EXHIBIT 7-2

              RELATIVE COSTS OF INSUFFICIENT SUBMISSION OPTIONS 1-4


Option                    1234

Cost to  Industry

Delays                    2s./              2£/             2               2

Section  5(c) Action       1112

Out-of-Pocket Costs       2222

Uncertainty               1-2              141
 1 = Negligible
 2 = Low
 3 = Medium
 4 = High

£/The cost of option  2 should be less than the cost of option 1, even though
  both are low.
Delay in the Introduction of New Chemicals

     The most important of the costs to industry with respect to this issue is
the delay between receipt of the section 5 notice by EPA and the time at which
the submitter can manufacture or import the substance.  Chemical industry
spokesmen have stated that delays in the introduction of new chemicals will be
costly to the producers because the competitive advantage of new chemicals
will be reduced.  The result will be a reduction in the number of new
chemicals introduced, which reduces the income and profits flowing to those
companies whose new chemicals are not introduced.il/  Delays may result if
the clock is stopped under the first three options, if the notice period is
extended, or if supplemental reporting or section 5(e) actions are pursued
under option 4.
     ll/comments of the Manufacturing Chemists Association on EPA Proposed
Regulations for Premanufacture Notification Under Section 5 of TSCA, March 26,
n f\*^f\  _  I^O
1979, p. 163.

-------
                                     -  180 -
     If new chemicals are not introduced, individual firms and  society  as  a
whole may be harmed.  But there may well be little or no cost to  the  industry
as a whole.  Individual firms whose products are not introduced because of
delays in the section 5 notice process will lose income and profits,  but
income will flow to the producers of the products that would have been
displaced by the new chemicals.  Therefore, the chemical industry as  a  whole
would suffer lost sales only when a foregone new chemical would have  opened a
new market for chemicals or when the foregone chemical would be replaced by
imports.  However, to the extent that the new chemical would increase
efficiency and productivity in the chemical industry, its delay would
indirectly hurt the industry.

     Options 1 and 2 differ in the way that they treat errors.  Under option
1, the notice period clock remains stopped until errors are corrected,  while
under option 2 the clock keeps running.  Therefore, delays under  option 1  are
likely to be greater than under option 2.  The extent of the delays should not
be great in most cases—perhaps as little as two weeks—depending  on  how
quickly firms respond to requests for corrections.

     The criteria for not accepting a submission are somewhat different under
options 1 and 2.  Exhibit 1 shows the criteria for option 1 and option  2.
Under option 2, eight major deficiencies are listed, seven of which are common
to option 1.  The only one that is not listed under option 1 is failure to use
the notice form.  Eleven major deficiencies are listed under option 1—the
seven that are common to option 2 plus four others.  Those four are failure to
contact customers, failure to remedy an error (which does not apply under
option 2), submission of intentionally false or misleading responses  (which is
covered in another part of option 2), and submission of a notice  by someone
other than the manufacturer or importer.  The conditions under which  a
submission is incomplete are roughly the same under options 1 and  2,  except
that option 2 criteria are somewhat less extensive.  Therefore, more
submissions may be found incomplete under option 1 than under option  2.

     Option 2 includes two other elements that are missing from option  1:   an
appeal process and the limitation of findings of major deficiencies to  the
first thirty days after the submission is received.  The appeal process should
result in a decrease in the cost of delays to industry, since it  would  only be
used when the submitter felt the expected benefits of a successful appeal
would exceed the additional delay resulting from the appeal process.  The
thirty day limitation may also reduce the number of rejections of  submissions.

     It is difficult to compare the extent of delays under option 3 to  any of
the other options, because EPA's latitude under option 3 is so  large.   Option
3 contains no guidelines on what fails to constitute a notice and  no  statement
of the procedures EPA will follow in attempting to obtain information.   The
extent of delays under option 3 also depends on the willingness of the
industry to correct deficiencies identified by EPA.

     The possibility of delays under option 4 is also depends on  the  behavior
of the industry.  If option 4 is adopted and industry is willing  to correct
deficiencies voluntarily, there will be little delay.  If industry does not

-------
                                      -181-
respond, however, EPA could extend  the notice period, request further
information under the supplemental  reporting provisions, or, if appropriate,
regulate the substance under section  5(e) of the Act pending development and
submission to  the Agency of sufficient data to evaluate the substance's
effects.  A low  level of voluntary  cooperation on  the part of the industry
could  lead to much greater delay  under option 4 than under any other option.
If delays are  as costly as claimed  by industry in  public comments received on
previous proposals, we could expect a relatively high degree of voluntary
compliance under option 4.

     As can be seen under all  the options, the extent of delay depends greatly
on the actions of chemical firms.   If chemical firms are willing to correct
deficiencies quickly, there should  be little delay under any of the options.
The extent of  delays also depends,  to a  lesser extent, on the uncertain
actions of EPA.  Because delays in  the introduction of new chemicals depend on
the future behavior of EPA and the  industry, any cost estimate will be highly
uncertain.  Assuming that submitters generally make good faith efforts to
submit complete  notices, the delays should be short and their costs low under
all options.   In addition, option 2 should produce fewer delays than option 1.

Restrictive Actions Under Section 5(e)
     Under option  4, EPA may try to persuade submitters to voluntarily provide
 information, or EPA may use the supplemental reporting provisions to obtain
 information.   If neither of these actions  is successful in obtaining
 corrections, EPA may regulate  the substance under section 5(e) of the Act
 pending development of information, provided the requirements for action under
 5(e) are met..£!/   Restrictions placed on  the chemical under section 5(e)
 may reduce or  eliminate the market for the substance or increase the
 manufacturer's cost of production.  Note that imposing 5(e)  restrictions can
 be taken for reasons other than the existence of an insufficient submission,
 but we deal here only with the 5(e) actions taken to correct insufficiencies.

     Actions taken under section 5(e) to correct deficiencies in the notices
 occur only under option 4.  Because the frequency of 5(e)  actions is heavily
 dependent on the actions of EPA and the chemical industry, it is difficult to
 estimate the frequency with which 5(e) actions would be undertaken for this
 purpose.  The  cost to industry of each action taken could be significant.  But
 if we assume that chemical firms generally make good faith efforts to provide
 complete submissions, the total costs of 5(e)  actions in response to
 insufficient submissions should be low.
     Ai/EPA may take action under section 5(e) if the Administrator deter-
mines that the information available is insufficient to permit a reasoned
evaluation of the new chemical, and either  (1) the manufacture, processing,
distribution in commerce, use, or disposal of the substance, or any combination
of such activities, may present an unreasonable risk of injury to health or
the environment; or  (2) the new chemical will be produced in substantial quan-
tities, and either enters or may reasonably be anticipated to enter the envi-
ronment in substantial quantities or there is or may be significant or
substantial human exposure to the chemical.

-------
                                      -182-
Out-of-Pocket Costs

     Industry will incur out-of-pocket costs to correct deficiencies under all
options.  Again, the cost of correcting deficiencies depends to a large extent
on the behavior of the industry.  Assuming that chemical firms generally make
good faith efforts to provide complete submissions, the out-of-pocket costs
should be small for all options.

Uncertainty

     In addition to the costs incurred as a result of actions taken under
these provisions, industry may also incur costs as a result of uncertainty
about how EPA will implement the rules.  Because of this uncertainty, firms
may take unnecessary actions or fail to take necessary actions; they may
devote too many or too few resources to the notice process.  Some uncertainty
is inevitable no matter which option is eventually chosen, since no set of
rules can completely specify a future set of contingent actions.  But a
different degree of uncertainty is associated with each option.

     Option 3 contains the most uncertainty for industry, since no criteria
for identifying deficiencies is given.  There is, furthermore, a great range
of action open to EPA to request or require additional information.  Much less
uncertainty is associated with option 1, which provides some description of
the criteria for identifying major deficiencies as well as a clear statement
about the actions EPA can take in response to a deficiency.  Least uncertain
is option 2 which clearly delineates responses to an inadequate submission.
Option 2 also lists the omissions which make a submission inadequate.  In
addition, option 2 provides that submissions must be determined to be
insufficient within 30 days after the receipt of the material.  We estimate
that the cost of uncertainty is low for option 2, low to medium for option 1,
and high for option 3.

     Option 4 provides the least uncertainty to the submitter, since it gives
the submitter the ability to refuse requests for information.  Because the
submitter has more control over the notice under option 4 than under the other
options, its uncertainty must be less under option 4 than under the other
options.  The submitter can guarantee he does no worse under option 4 than
under the other options by simply acceding to all requests for correction of
insufficiencies.  But if the submitter believes that the expected costs of
supplying a piece of information are greater than the expected costs of not
providing the information  (e.g., the possibility of restrictions under section
5(e)), a rational submitter will choose to withhold the information.
Therefore, under option 4, the expected costs to industry are lower than under
the other options..IV
     .iVEven though expected costs are lower under option 4, actual costs
may be greater,  in the absence of information, the submitter's predictions
about  (1) EPA's actions, (2) the cost of those actions, and (3) the cost of
providing information may be quite poor.  The submitter's actual, as opposed
to expected, costs under option 3 will be lower only if his predictions are
unbiased estimators of the actual outcome.

-------
                                     -183-
SUMMARY

     As stated above, option 4 must be the least costly of all the options
because of the discretion it allows the submitter.  The differences among the
other three options  (as well as the magnitude of the difference between option
4 and the other options) are heavily dependent on the behavior of industry.
The major differences among option 1, 2, and 3 lie in the cost of
uncertainty.  Option 3 has an uncertainty cost that is much higher than option
2.  Option 2 has lower costs of uncertainty than option 1.

-------
                                      -185-
                                    CHAPTER 8

                               PROCESSOR REPORTING
INTRODUCTION

     EPA recently proposed a processor reporting rule to supplement the
section 5 rules proposed on January 10 and October 16, 1979.  The purpose of
this analysis is to assess the potential economic impact of the proposed
processor rule and its alternatives.

     Because many of the costs of processor reporting do not easily lend
themselves to quantification, the major outcome of this analysis is a ranking
of options by expected economic costs to industry.

     The analysis (incorporates previous cost estimates by Arthur D. Little,
Inc., for preparing and filing section 5 and "minimum guidance" notices.  ICF
estimates of the cost of submitting section 8 notices are also cited where
appropriate.  (See table at end of Introduction.)

     Background information on the proposed rule and alternatives was obtained
from EPA draft documents on processor reporting and communications with EPA
personnel.

     This analysis is organized in five sections.  The first section provides
background information on the proposed processor rule and states the scope of
this analysis.  Section 2 provides some information on the extent of processor
reporting impacts, e.g., how many and what types of processors are likely to
be affected by the proposed rule.  Section 3 describes the options, and
section 4 distills unique features from each option.

     The analysis of economic costs begins in section 5 with the
identification of major costs to industry and an assessment of the relative
magnitude of such costs.  The options are then ranked in terms of the
following costs:

               Out-of-pocket expenditures
               Time delays
               Disclosure of information
               Risk of adverse determination
               Number of persons covered


     This analysis is designed to assist EPA in making optimal decisions on
processor reporting by constructing an economic framework from which to
evaluate the proposed rule and its alternatives.

-------
                                      -186-
            ESTIMATED  COSTS  OF  PREPARING  AND  FILING PROCESSOR NOTICES
                Alternative Notices

                Section ,6 Notice        $1,155   to   $8,925a/
                Section 8 Notice           724   to    4,45Qb/
                Minimum Guidance         1,011   to    7,486£/
                January 10 Notice        3,700   to   42,000
-------
                                      -187-
OVERVIEW

Background

     The processor reporting rule only applies to certain, so-called "exempt,"
chemicals,   it is therefore necessary to begin analyzing the rule by
identifying  which chemicals are affected by the rule.  Section 3(2) (B)  of the
Toxic Substances Control Act excludes the following substances from coverage
by the Act,  as they are not chemical substances:

     •    any pesticide  (as defined in the Federal Insecticide,
          Fungicide, and Rodenticide Act) when manufactured,
          processed, or distributed in commerce for use as a
          pesticide,

     •    tobacco or any tobacco product,

     •    any source material, special nuclear material, or
          byproduct material  (as such terms are defined in the
          Atomic Energy Act of 1954 and regulations issued under
          such Act),

     •    any article  (e.g., firearms)  the sale of which is subject
          to the tax imposed by Section 4181 of the Internal Revenue
          code of 1954  (determined without regard to any exemptions
          from such tax provided by Section 4182 or 4221 or any
          other provision of such code), and

     •    any food, food additive, drug, cosmetic, or device (as
          such terms are defined in Section 201 of the Federal Food,
          Drug, and Cosmetic Act) when manufactured, processed, or
          distributed in commerce for use as a food, food additive,
          drug, cosmetic or device.

     Certain chemical substances are also exempted from section 5 review.
Specifically, section 5(h)(1) authorizes the EPA Administrator to grant
section 5 notice exemptions, upon application, for chemical substances
manufactured or processed for test marketing.

     Section 5(h)(3)  exempts chemical substances manufactured or processed
only in small quantities solely for the purposes of scientific experimentation
or analysis, or for the purposes of chemical research on or analysis of the
substances or another substance, including research or analysis for the
development  of a product (R&D).

     In addition, section 5(i) of TSCA defines "manufacture and process" to
mean manufacturing or processing for commercial purposes.  Thus, section 5
reporting applies to chemicals manufactured or processed for commercial
purposes.  The January 10 proposal on premanufacture notification defines
by-products  as substances manufactured without a separate commercial purpose.
Therefore, by-products are exempt from section 5 reporting.

-------
                                      -188-
     On December 23, 1977, EPA promulgated regulations that allowed the
reporting of chemical substances for the TSCA Inventory.  The rules also
allowed reporting for chemical substances manufactured, imported, or processed
for a commercial purpose since January 1, 1975.  Therefore, according to these
rules, chemical substances manufactured or imported only before January 1,
1975 and not processed after that date and reported, would not be added to the
Inventory.  Although the Inventory was initially set up to comply with
sections 8(a) and (b) of TSCA, it served other important purposes.  In
particular, chemical substances not on the Inventory after a certain date were
to be considered "new" chemicals for the purposes of section 5 reporting.

     In summary, the following chemical substances are covered by the proposed
rule under certain conditions:

     •    those not within the TSCA definition of "chemical
          substances";

     •    those manufactured for test-marketing;

     •    those manufactured for research and development;

     •    those byproducts manufactured without a separate
          commercial purpose; and

     •    those manufactured only before January 1, 1975 and not
          processed and reported after that date.

The Need for a Processor Reporting Rule

     The chemicals described above may be processed for a TSCA nonexempt
commercial purpose.   "Process" is defined in TSCA as the "preparation of a
chemical substance or mixture, after its manufacture, for distribution in
commerce—(A) in the same form or physical state as, or in a different form or
physical state from, that in which it was received by the person so preparing
such substance or mixture, or (B) as part of an article containing the
chemical substance or mixture."  Reporting is not now required for the
processing of an exempt, non-inventoried chemical for a nonexempt commercial
purpose.  Therefore, exempt substances processed for such a purpose would
enter commerce without having been reviewed.

     The rule would  modify a subsection of the January 10 proposed
regulation.   According to this subsection, persons who intend only to process
a new chemical substance for commercial purposes are not subject to
premanufacture notification requirements (Subpart B, section 720.11 (b)).

Examples of Gaps Covered by Processor Reporting Rule

     The processor reporting rule is designed to close a gap in EPA's
regulatory scheme.  The following serve as examples of instances when the
processor rule would apply.

-------
                                       -189-
     •    Chemicals Not Within Scope of TSCA Definition of "Chemical
          Substance";  A substance may not be on the TSCA Inventory
          because it is manufactured solely for a non-TSCA use.  A
          processor wants to process the substance for a nonexempt
          commercial purpose, say as a plasticizer.  Processing the
          substance creates a significant new use and changes the
          type, form, or duration of exposure.  Although the
          substance may have previously fallen under a regulatory
          authority other than TSCA, the processor rule would
          require that the substance be reported and reviewed before
          it is added to the Inventory and possibly used for any
          TSCA purpose.

     •    By-products;  The production of a chemical results in a
          by-product which is of no commercial value to the
          manufacturer.  However, a processor may later find it
          profitable to process and distribute the by-product in
          commerce.  Under the January 10 proposal, no section 5
          notice is required because the by-product is exempt and
          because processors are not subject to premanufacture
          notification requirements.  The processor rule would allow
          EPA to review the by-product for toxic effects before it
          is processed for commercial activity.

     •    Research and Development Substances;  A new substance is
          manufactured for research and development.  No section 5
          notice is required.  A processor other than the
          manufacturer may want to process and sell the R&D
          substance for a nonexempt purpose.  No section 5 notice is
          required because the January 10 proposal does not require
          section 5 notices from persons who intend only to process
          a new chemical substance for commercial purposes.  The
          processor rule would allow EPA to review changes in
          exposure to the chemical substance.

     •    Test-marketed Substances;  A new substance is manufactured
          under a test-marketing exemption.  The substance shows no
          sign of profitability in its test-marketed form.  However,
          a person other than the manufacturer recognizes an
          economic benefit in processing the substance.  Again, no
          section 5 notice is required under the January 10
          proposal.  The processor rule would require submission of
          a notice prior to processing.

Scope of Analysis

     The processor rule may impose certain costs on industry.  Our task is to
identify these possible costs and assess the extent and magnitude of their
impact.

     We approach this task first by specifying the processor reporting options
and distilling from each option unique features related to coverage, reporting

-------
                                      -190-
requirements, enforcement,  and disclosure.  Next,  we examine the expected
economic costs to industry  of processor reporting.  In order to assess the
extent of these costs, we make several assumptions based on information
provided by EPA.  Finally,  the alternatives are ranked according to their
expected impact.
EXTENT OF IMPACTS
Processors
     Chemical industry processors who will now be subject to section 5
requirements include manufacturers and importers who process substances and
"separate" processors who do not manufacture or impact the exempt chemical
substances.

     The precise number of processors falling in these ^categories is not
available.  However, according to a report prepared for EPA,1§/ over 50
percent of the chemical firms in the United States only process or formulate
chemicals and do not manufacture them, i.e., they are "separate" processors.
Assuming there are 7,000 chemical firms (excluding drug companies), at least
3,500 separate processors may be affected by a processor reporting rule.

     It is important to know not only how many processors will be affected by
the rule, but also who these processors are.  How they fill out the notice
form will reflect their knowledge of the chemical and this may in sane way
affect EPA's action on a notice as well as the kinds of assumptions the Agency
makes about a particular notice.  EPA has solicited comments from processors
on the proposed rule.  When this information becomes available, it will be
possible to determine the number of processors which will be affected.
Financial characteristics of processors may also affect analysis of the
impacts of a processor rule to the extent that a notice requirement may alter
a processor's decision to commercialize an exempt substance for a TSCA purpose.

Substances

     Another important part of the analysis is the number of notices expected
to be submitted under the rule.  How often are chemicals processed in such a
manner as to require reporting under the rule?  This question is related not
only to industry costs, but also to the costs of administrative review of
processor notices by EPA.

     Estimates of the number of exempt chemical substances are not available.
However, the following assumptions help to narrow the focus of the
analysis;47/
    .i§/Arthur D. Little, Inc.  Impact of TSCA Proposed Premanufacturing
Notification Requirements (Cambridge, Mass.:  December 1978), page II-2.

    AZ/Assumptions were obtained from Premanufacture Review Program;
Proposed Processor Requirements, 45 FR 54642, August 15, 1980.

-------
                                     -191-
     •    Substances manufactured for foods, food additives,  drugs,
          or cosmetics are not likely to be processed frequently for
          TSCA purposes due to their specialized applications and
          chemical formulas.  The same is likely to be true for
          chemicals manufactured as source materials, special
          nuclear materials, or by-product materials as defined by
          the Atomic Energy Act of 1954.

     •    Substances manufactured for R&D and test marketing  are
          likely to be processed more frequently for nonexempt
          commercial purposes than other types of exempt substances
          because they were manufactured for exploring nonexempt
          uses in the first place.

     •    Pesticide ingredients are likely to be processed for non-
          exempt commercial purposes.  This assumption is based on
          the presence of inert and active pesticide ingredients on
          the TSCA Inventory.

     Under these assumptions, greater emphasis may be placed  on R&D  chemicals,
substances with test marketing exemptions, and pesticide ingredients.   EPA
provided the following information that may be valuable in setting constraints
on the analysis;48/

     •    Of 1,452 pesticide active ingredients, 688, or 47.4
          percent are listed on the TSCA Inventory.

     •    Of 734 inert pesticide ingredients, 617, or 82.7 percent
          are on the Inventory.

     •    It is probable that not all of the pesticide ingredients
          reported for the Inventory were reported by processors.
          Some were manufactured or imported in a manner
          inconsistent with an exemption and thus were reported for
          the Inventory by manufacturers or importers.

     •    One thousand to 2,000 new R&D chemicals are manufactured
          each year.  This range cannot serve as a lower bound on
          the number of R&D substances affected by the processor
          rule because not every new R&D chemical will be processed
          for a nonexempt commercial purpose.  Secondly, this range
          cannot serve as an upper bound on the number of R&D
          substances affected each year because "old" R&D chemicals
          may also be processed for nonexempt commercial purposes.
          However, this range can serve as an upper bound on the
          number of new R&D substances potentially affected by the
          processor rule each year.
    M/Ibid.

-------
                                      -192-
OPTIONS

     Substances manufactured or imported for TSCA-exempt purposes may later be
processed for TSCA commercial uses.  Under the joint statutory authority of
TSCA Sections 5(a) (1) and 5(a) (2) , EPA has proposed to require any person who
first processes an exempt, non-inventoried substance for a nonexempt purpose
to submit a section  5 notice.  The notice would be virtually identical to a
premanufacture notice.

     For purposes of analysis, the Agency specified the following alternatives
to the proposed processor rule:

     •    January 10 Proposal—Manufacturer and importer reporting;

     •    no reporting for commercial processing of exempt
          substances;

     •    section 8 processor notification;

     •    processor rule as proposed, with exemption for one-time
          processing;

     •    combination of section 5 and section 8(a)  notices; and

     •    minimum guidance notice.

     Features of the proposed rule and its six alternatives are described in
Sections 3.1 through 3.7.

Option 1—Proposed Processor Reporting Rule

     At least 90 days before commencing first-time commercial processing for a
TSCA purpose, the processor must ensure that a section 5 notice is submitted.

     The prospective processor has three options for reporting.  He may:

     •    submit the notice himself;

     •    persuade the manufacturer or importer to submit it; or

     •    submit the notice jointly with the manufacturer or
          importer.

In any event, the processor is responsible for ensuring that the substance in
question completes a section 5 review prior to processing.

     In addition to data required  by the section 5 rules as proposed on
January 10 and October 16,  the processor must submit information concerning:

     •    the current source of  the substance; and

     •    future sources of the  substance.

-------
                                      -193-
Qption 2—January 10 Proposal

     Sections 720.10(a)(3) and  (b)(2) of the January 10 proposal require
section 5 notices from the manufacturer or importer if he distributes or uses
R&D or test-marketing substances  in a manner inconsistent with these
exemptions.  Separate processors would not be required to submit notices.

Option 3—No Reporting for Commercial Processing of Exempt Substances

     There would be no reporting prior to the processing of a chemical
substance for a nonexempt commercial purpose.

     EPA could conceivably rely on the terms of TSCA Sections 5(a)(1)(A) and
15(2) to restrict the nonexempt processing of exempt substances.

Option 4—Section 8 Processor Notice

     As an alternative to the processor reporting rule, EPA could require
preprocessing notices under the authority of TSCA Section 8.  Briefly, a
section 8 rule could have the following features:

     •    does not apply to "small processors";

     •    allows a notice to be submitted less than 90 days before
          processing;

     •    may require fewer data  than the October 16th proposed
          section 5 notice;

     •    does not result in the  substance being added to the
          inventory (as considered here); and

     •    no section 5(e) or 5(f) action could occur during the
          review period.

     With respect to the last feature, the Administrator could not issue
proposed section 5 orders to prohibit or limit the manufacture, processing,
distribution in commerce, use, or disposal of the substance in question, but
could regulate the chemical under TSCA sections 4, 6, and 7.

Option 5—Processor Rule with Exemption for One-Time Processing

     One-time processing applies  to those exempt substances processed for use
on a one-time or limited basis and processed only as an alternative  to
disposal of the substance.  Option 5 incorporates this one-time processing
exemption into the provisions of option 1.   (This exemption could be
incorporated into most of the options.)

     Under option 5, EPA is considering limited section 8 reporting only for
those substances that:

-------
                                      -194-
     0    were manufactured or imported in quantities over 500
          pounds for R&D or test marketing; and

     •    are processed for a nonexempt commercial purpose in
          quantities over 500 pounds, solely as an alternative to
          disposal.

     R&D and test-marketing substances processed for a nonexempt commercial
purpose in quantities under 500 pounds, solely as an alternative to disposal
could be processed without any section 5 or section 8 notification
requirements.

Option 6—Combination of Section 5 and Section 8(a)  Notices

     Option 6 is a combination of options 1 and 4 and consists of two parts:

     •    section  5(a)(2) notice for substances of special concern;
          and

     •    section  8(a) notice for the remainder of exempt substances
          processed for nonexempt commercial purposes.

     The section 5 rule would apply to specific chemical categories (selected
L.y EPA)  whose nonexempt commercial processing raises special concern about
toxicity or exposure.  The rule would establish which processors would report
and when they would report.

Option 7—Minimum Guidance Section 5 Notification

     This alternative would cover the same persons as the "rule" proposed in
August 1980.  It would require a 90-day advance notice and would provide
minimum guidance on information requirements.  The minimum guidance notice
takes somewhat less time to complete than a section 5 notice as proposed in
October.  This statement is based on the Arthur D. Little and ICF estimates
for the cost of submitting a section 5 notice.1V and the cost of submitting
a notice under the minimum guidance option.j>£/  The former would range from
$1,155 to $8,925 and the latter would range from $1,011 to $7,486.

     Regulatory action under sections 5(e) and 5(f)  could still be taken under
option 7.
    .ii/Arthur Q. Little, Inc., Estimated Cost for Preparation and Submission
of Reproposed Premanufacture Notice Form (Cambridge, Mass.:  September 1979).

    12/ICF Incorporated, Estimated Cost of Compliance with the Minimum
Guidance Option.

-------
                                      -195-
MAJOR DIFFERENCES AMONG ALTERNATIVES

     Comparisons between processor notice options indicate three major areas
of difference.  They are:

     *    Coverage;  Certain persons may be covered by one form of
          the processor rule and not by others.  For example,
          separate processors do not submit notices under Option 2
          (January 10 Proposal).  However, the proposed rule (option
          1) would require separate processors to submit section 5
          notices.

     •    Reporting Requirements;  Some alternatives request more
          information than others.

     •    Regulatory Authority;  The options differ in terms of
          subsequent regulatory action.  For example, under a
          section  8 rule, EPA could not regulate the production or
          distribution of an exempt substance that is processed for
          a nonexempt commercial purpose using section 5 authority.
          However, under the proposed rule, the Agency would
          regulate such a substance under TSCA Sections 5(e) and
          5(f).

     •    Time;  Some options require a notice to be submitted
          before a specified time period while other options allow
          submissions upon processing.  This feature creates
          differences across options in terms of delays experienced
          by  the processor before he is allowed to commence his
          operations.  For example, the proposed processor  reporting
          rule would represent  at least a 90-day delay for  the
          processor; the no reporting option, of course, means no
          delay, and the section  8 option could represent almost no
          time delay in processing a chemical substance.

     In  addition to these areas,  certain options may have features that
distinguish them in terms of implementation costs, confidentiality effects,
risk of  adverse determinations, etc.

     The Appendix  to this Part  summarizes differences between  options  in  terms
of  coverage,  reporting requirements, risks of disclosure, and  other  factors.


ANALYSIS OF ECONOMIC COSTS

     EPA is proposing  the processor notice rule  to  ensure that exempt
substances  undergo Agency review  prior  to commercial processing  for  a
nonexempt purpose.  Without such  a  rule, persons could process for a TSCA
commercial  purpose an  exempt substance  which was not on  the Inventory  and
which  was never  reviewed  for toxic effects.

-------
                                      -196-
     Although the January 10 proposal (720.11 Note) allows the manufacturer or
importer to request that the processor "participate in the filing of the
notice by providing information on uses and exposures, either to the
manufacturer or importer, or directly to EPA," the proposal does not establish
any mechanism for the processor to file a notice independent of the
manufacturer or importer.

     The purpose of this section is to identify the economic burdens that may
accompany promulgation of a processor notice rule.

Industry Costs

     The bulk of the economic costs of a processor rule will fall on industry
in the form of out-of-pocket costs, time delays, risks of disclosing confiden-
tial business information, and risks of adverse determinations.  This section
deals with industry costs in four parts.  The first describes the potential
costs incurred by processors on a per-notice basis.  The second part explains
why costs differ among processors, and the third part summarizes the cost of a
processor rule for the industry as a whole.  In the fourth part of this
section, the processor rule and its alternatives are ranked according to cost.

     Per-Notice Costs.  The per-notice cost to industry of filing a processor
notice directly corresponds to the cost of filing a premanufacture
notification as it applies to manufacturers and importers.  Examples of
possible industry costs include the following:

     •    Out-of-pocket expenditures;  This category involves the
          direct cost of providing information.  Out-of- pocket
          costs will vary from alternative to alternative, depending
          on the amount and type of information requested.

     •    Time Delays;  Preparing and filing a processor notice and
          awaiting EPA review may cause significant delays in
          getting a processed substance out in commerce.  The costs
          of such delays are clear.  However, it is likely that the
          length of delay will vary from alternative to alternative.

     •    Disclosure;  For various reasons, a processor may not want
          to disclose his intent to process a chemical substance.
          However, if a processor must contact the manufacturer or
          importer of a chemical substance for additional
          information to complete a notice, the processor risks
          disclosing his intent to process.  The need to contact the
          manufacturer or importer may arise when the processor
          lacks complete information on the chemical identity,
          production process, production volume, or other data
          requested in a notice form.

          The risk of disclosing an intent to process may arise
          under any one of the alternatives, except the "no
          reporting" option.

-------
                                      -197-
     •    Risk of Adverse Determinations;  Under the January 10
          proposal, separate processors were not faced with EPA
          regulatory action.  With the processor rule, exempt
          substances processed for a TSCA commercial purpose could
          be regulated under sections 5(e) and 5(f).  This added
          risk represents a cost to industry because it may
          discourage processing and lead the industry to forego
          potential earnings on the processed substance.

     Variables Influencing Notice Costs.  According to a report by ADL,JLL/
the following company-related variables could affect the cost of filing a
notice:

     •    company size;

     •    organizational structure (centralized vs. decentralized);

     •    existing information storage and retrieval methods;

     •    information availability;

     •    level of technical resources used to complete notice; and

     •    importance of claiming confidentiality.

In addition, the report cites the following chemical-substance-related
variables that could influence notice costs:

     •    apparent toxicity of chemical substance;

     •    anticipated sales or profit potential;

     •    complexity of chemical production process; and

     •    complexity of chemical composition.

Because the behavior of these variables is unique to each situation that
requires a section 5 notice, the cost of filing a notice is expected to be
different for each substance.

     Total Industry Cost.  The total cost to industry of a processor notice
rule is the sum of all per-notice costs incurred in a given period of time.
    li/Arthur D. Little, Inc., Estimated Costs for Preparation and
Submission of Reproposed Premanufacture Notice Form  (Cambridge, Mass.
September 1979).

-------
                                      -198-
     An important caveat arises in the calculation of total industry costs.
That is, the number of notices attributable to the processor rule does not
include notices that would have otherwise been submitted without the rule.
This caveat is applicable on two occasions:

     •    when the processor is the same person as the manufacturer
          or importer of an R&D or test marketed substance, and
          knows prior to manufacture that he or another will process
          the substance for a nonexempt purpose; or

     •    when the processor is not the manufacturer or importer,
          but the manufacturer or importer of an exempt substance
          would have submitted a notice because he knew of an intent
          to process the substance for a nonexempt commercial
          purpose.

     In the first instance, no new costs would be added by a processor rule.
Using the January 10 proposal as the baseline for analysis, the manufacturer
or importer of an R&D or test-marketed substance would file a notice under
sections 720.10(a) (3) and  (b) (2) because he would have had knowledge of an
intent to use his exempt substance "in a manner inconsistent with the terms of
the applicable exemption."  With a processor rule, the same person (the
importer or the manufacturer) would file a notice presumably at the same cost.

     In the second instance, the processor is distinct from the manufacturer
or importer.  However, the manufacturer or importer may still know of the
processor's intent to process an exempt R&D or test-marketed chemical for a
nonexempt purpose and would have submitted a notice under Sections
720.10(a)(3) and  (b)(2).  With a processor rule, the savings realized by the
manufacturer or importer in a transfer of reporting responsibility would
cancel the costs incurred by the separate processor, ceteris paribus, of
submitting a notice.  Thus, no new costs would be added by a processor rule in
this case either.

     Costs incurred by a processor in filing a section 5 notice would be
directly attributable to the processor notice rule as long as a notice would
not have been submitted by the manufacturer or importer in the absence of a
processor rule (e.g., in the case where the manufacturer or importer did not
have knowledge of an intent to process an exempt substance for a nonexempt
purpose at the time of manufacture).

     Ranking of Alternatives by Cost.  Ranking the proposed processor rule and
its alternatives by cost requires a few assumptions.  The assumptions are:

     •    all notices are submitted by persons of identical salaries
          and technical expertise;

     •    all notices are submitted by processors of identical
          financial characteristics, including company size and
          organization;

-------
                                     -199-
     •    all processors have the same quantity and quality of
          information regarding the substance to be processed; and

     •    time delays, uncertainty, risks, etc., affect all
          processors equally.

     Exhibit 8-1 presents the results of ranking all options by per-notice
cost and by coverage.  A "1" indicates the least costly alternative or least
coverage.  A higher rank indicates a more costly alternative or more extensive
coverage.  The ranking was based on information from a number of sources:

     •    ADL estimates in Impact of TSCA Proposed Premanufacturing
          Notification Requirements, December 1978.

     •    ADL estimates in Estimated Costs for Preparation and
          Submission of Reproposed Premanufacture Notice Form,
          September 1979.

     •    Premanufacture Review Program; Proposed Processor
          Requirements, 45 FR 54642, August 15, 1980.

     •    ICF estimates in "Cost Estimates of Alternative Processor
          Notification Requirements" in Appendix A of
          Appendix-Volume 1.

     •    ICF estimates in "Estimated Cost of Compliance with the
          Minimum Guidance Option" in Appendix A of Appendix-Volume
          1.

     The ranks were established in the following manner:

     •    Out-of-Pocket Expenditures

          Option 3 (no reporting for commercial processing) and case
          2  (January 10 proposal) of option 2 received the lowest
          rank for costs incurred in filing a notice.  In fact,
          filing costs in both cases would be zero.  Undar the
          option of no reporting, a processor would not submit a
          notice.  Under case 2 of the January proposal, the
          separate processor would not be required to submit a
          notice.

          Although a separate processor incurs zero cost under the
          January 10 proposal, a manufacturer who knows of an intent
          to process his substance for a nonexempt commercial
          purpose must submit a section 5 notice under the same
          (January 10) proposal.  This case was included in the
          ranking to compare the out-of-pocket costs incurred by the
          manufacturer under the January proposal and by the
          separate processor under the proposed processor reporting
          rule.  These options ranked among the costliest in terms
          of filling out a notice.  ADL estimated $3,700 to $42,000
          for a January 10 notice, and $1,155 to 8,925 for the
          proposed processor rule.

-------
                                     -200-
                                  EXHIBIT  8-1

         RANKING OF PROCESSOR NOTIFICATION OPTIONS BY COSTS TO INDUSTRY
Costs Per Notice
Options
1. Proposed
Processor
Notice Rule
2. January 10
Proposal3/
• Case lb/
• Case 2
3. No Reporting
4. Section 8 Notice
5. Processor Rule
With Exemption
• Section 5
• Section 8
6. Combined
Section 5 and
Section 8 Notice
• Section 5
• Section 8
7. Minimum Guidance
Notice
Out-of-Pocket Time
Expenditures Delays

4


5
1
1
2


4
2



4
2
3

4


4
1
1
2


4
2



4
2
4
Risk of Adverse
Disclosure Determination

4


3
1
1
2


4
2



4
2
3 or 4

2


2
1
1
1


2
1



2
1
2
Coverage

5


2
1
1
3


4
4



4
4
5
  A "1" indicates the least costly alternative.
£/Case 1 assumes the manufacturer has prior knowledge of an intent to process
  his substance for a nonexempt commercial purpose.   Case 2 assumes that only
  the separate processor has knowledge of such an intent.  All other options are
  ranked according to the Case 2 assumption.
        costs are incurred by the manufacturer.

-------
                                      -201-
          Option 4  (section 8 processor notice) received a rank of 2
          because, as proposed here, less information is requested
          under the section 8 rule than under the section 5 rule
          (option 1) and less time would be required to complete the
          notice.  According to ICP estimates, the cost of filing a
          section 8 notice under option 4 ranges from $724 to
          $4,450.  This compares to ADL's estimates of $1,155 to
          $8,925 for completing a section 5 notice.

          Out-of-pocket costs under options 5  (processor rule with
          exemption for one-time processing) and 6 (combination of
          section 5 and 8 notices) would be the same as under option
          4 when a section 8 notice is required (i.e. from a
          one-time processor under option 5 and from a processor
          subject to section 8 review under option 6).

          However, out-of-pocket costs under options 5 and 6 may be
          identical to out-of-pocket costs under the proposed rule
          when a section 5 notice is required  (i.e. when the
          one-time exemption is inapplicable under option 5 and when
          a processor is subject to section 5 review under option 6).

          The cost of submitting a notice with minimum guidance (ADL
          estimates of $1,011 to $7,486) is more than a section 8
          notice  ($724 to $4,450) and less than a section 5 notice
          ($1,155 to $8,925).  Thus option 4 was ranked third
          costliest.

          Time Delays

          Delays would be nonexistent under the no-reporting option
          and under case 2 of the January 10 proposal.52/

          Submitters of a section 8 notice will experience some
          delay, although it is expected that any delay will be
          shorter than under a section 5 notice for two reasons:

               A section 8 notice may be submitted less than 90 days
               before processing, but a section 5 notice must be
               submitted at least 90 days prior to processing.

          —   A section 8 notice requests less information than the
               proposed section 5 notice, thus the time required to
               complete the form is shorter.
    .52/Case 2 assumes that only the separate processor has knowledge of an
intent to process for a nonexempt commercial purpose.  Case 1 assumes that the
manufacturer has knowledge of such an intent.

-------
                            -202-
Therefore, option 3 and case 2 of option 2 were ranked as
the least costly in terms of time delays.  Option 4 was
ranked as the next costly, followed by those options
requiring a section 5 notice.

Because options 5 and 6 require either a section 5 or a
section 8 notice, each was assigned two ranks.  When the
section 5 notice is submitted, delays will be the same as
under option 1 and case 1 of the January 10 proposal.  When
the section 8 notice is submitted under option 5, delays
will be somewhat longer than under option 4 for the reasons
cited above.  When the section 8 notice applies to option
6, delays would be identical to those under option 4.

Disclosure

The risk of disclosing confidential business information
is likely to be highest under the proposed processor
rule.  In addition to facing the possible disclosure of
information as part of a notice submission, the separate
processor is faced with another risk.  That is, it may be
necessary to contact the manufacturer in order to complete
a notice, and the processor may not want the manufacturer
to know of his processing intentions.

The manufacturer in case 1 of option 2 who knows that his
substance will be processed for a nonexempt purpose must
submit a section 5 notice and may incur some risk of
disclosing information provided in the notice.  However,
because contact between a separate processor and the
manufacturer is unnecessary (since no processor reporting
is required), the risks of disclosure are less than under
the proposed processor rule.

Under a minimum guidance option, manufacturer contact may
be necessary.  However, the processor need not submit as
much information as was required in the section 5 notice
proposed in October, so he risks the disclosure of less
information.  Taken together,  these two conditions leave
the minimum guidance option with a rank similar to that
for options 1 or 2.

When section 5 notices must be submitted, options 5 and 6
rank high along with option 1 in terms of disclosure
risks,  when section 8 notices must be submitted, however,
the options rank with option 4.  Although manufacturer
contact under option 4 may pose some risk of disclosure,
it is unlikely that the risks will be as high as under
option 1 and case 1 of option 2.  Thus, the section 8
notice received the third highest rank.

Option 3 and case 2 of option 2 would pose no risk of
disclosure.

-------
                            -203-
Risk of Adverse Determination

Upon reviewing a notice, EPA may determine that the
intended activity described in the notice may present
unreasonable risks.  Under options where section 5 rules
apply, EPA can take regulatory actions against the
intended activity using the authorities of sections 5(e)
and 5(f).  Thus, risks of adverse determinations, i.e.
regulatory action, are high under these options.

On the other hand, when no provision is made for a notice
submission, EPA would have a more difficult time
identifying activities that may present unreasonable
risks, and processors would face lower risks of adverse
determinations.  This would be the case for the option of
no reporting and for case 2 of option 1, where no notice
would be filed.  Thus, these options received ranks of 1,
indicating least cost.

Submitters also face the risk of an adverse determination
if they file incomplete submissions.  That is, if EPA
determines that a notice is incomplete, then the processor
cannot begin his processing activities.  While the risk of
an adverse determination represents a cost in itself, the
cost to a processor of an adverse determination is the
foregone opportunity to process a substance for commercial
purposes.

Coverage

The total cost to industry of a processor reporting rule
is a function not only of notice costs but of coverage.
The more people covered by the rule, the higher the cost
to industry, other things being equal.

The proposed processor rule requires that all those
persons who process exempt chemicals not on the TSCA
Inventory for nonexempt commercial purposes must ensure
that section 5 notices are submitted.  Coverage is widest
under this option and also under the option requiring a
notice submitted with minimum guidance.

Coverage is also broad under option 5, although one-time
processors would be exempt from submitting section 5
notices.  Therefore, option 5 ranks just under the
proposed rule in terms of coverage.

Some uncertainty surrounds coverage under option 6.  Small
processors would be exempt from section 8 reporting.
Thus, coverage is narrower than under option 1.  However,
some processors will be subject to section 5 reporting,
and coverage under option 6 is greater than under option 4
which promises the third highest coverage.

-------
                           -204-
Under case 1 of option 2, only a few processors are
covered (i.e., the manufacturer/processor who will either
process or know of an intent to process his substance for
a nonexempt commercial purpose).

Under option 3 and case 2 of option 2,  no processors are
covered.

-------
           -205-
  APPENDIX TO CHAPTER 8
DIFFERENCES AMONG OPTIONS

-------
                                     -206-
                      APPENDIX:  DIFFERENCES AMONG OPTIONS
A.I  PROPOSED NOTICE RULE  (7) VS. JANUARY 10 PROPOSAL (2)
     Coverage;
     Reporting
     Requirements!
     Disclosure:
     Other:
Under option 2, the manufacturer or importer of an R&D or
test-marketed substance is to submit a section 5 notice as
proposed on January 10, 1979 if he distributes or uses the
exempt substance in a manner inconsistent with the
exemption.  Option 2 does not permit separate processor
reporting.  Option 1 places reporting responsibility on
the processors.  Option 1 also deals with FIFRA, FFDCA,
by-products and substances manufactured, only before
January 1, 1975 and not processed after that date.

Under option 1, the processor submits a notice virtually
identical to the manufacturer's October 16, 1979 proposed
section 5 notice.  The submitter is also required to
identify sources of the substance, describe the
manufacturing process, and estimate the amount of the
substance to be processed.

Although the processor is believed to have a good
knowledge of his processing techniques and perhaps his end
use of the substance, he may have a significantly less
complete knowledge of its health and environmental effects
and possibly the specific chemical identity.  In these
instances, the processor may find it necessary to contact
the manufacturer for further information.   This procedure
raises the issue of confidentiality both for the processor
and the manufacturer.

If the processor has a less complete knowledge of the
substance, he may face a greater risk of follow-up action
by EPA.
A.2  PROPOSED PROCESSOR NOTICE RULE (1)  VS. NO REPORTING FOR COMMERCIAL
     PROCESSING OF EXEMPT SUBSTANCES (3)
     Coverage;
     Reporting
     Requirements:
Under option 1, the processor submits a notice.  Under 3,
no one, not even the manufacturer or importer who may be
aware of an intended inconsistent use, submits a notice.
In effect, option 3 permits all processed substances to
enter commerce without EPA review.

Under option 3, no information is submitted to EPA for the
commercial processing of exempt substances.  Under the
proposed rule, the processor submits a notice virtually
identical to the domestic manufacturers' October 16, 1979
proposed section 5 notice.

-------
                                     -207-
     Regulatory     EPA would have a difficult time enforcing TSCA under option
     Authority;     3.  Without a processor rule, persons who process exempt
                    substances for nonexempt commercial purposes could not be
                    penalized for their activity.  Yet, the intent of TSCA is
                    to review such activity prior to commercial processing.
                    Option 1 would satisfy this requirement.

     Disclosure;    As described in A.I of this appendix, the proposed rule
                    raises the issue of confidentiality.  With no reporting,
                    the issue would not be raised.
A.3  PROPOSED PROCESSOR NOTICE RULE (1) VS. SECTION 8 PROCESSOR NOTICE (4)

     Coverage;      Under 1, the processors submit a section 5 notice as
                    proposed.  Under 4, "small processors," are exempt from
                    filing a section 8 notice.

     Reporting      Less information could be requested under option 4 than
     Requirements;  under option 1.  Also a less-than-90-day advance notice
                    could be permitted.

     Regulatory     Whereas under 1, the Administrator could take action under
     Authority;     sections 5(e) and  5(f) during the review period, under 4,
                    the Administrator would be forced, in limited cases, to
                    prohibit or limit  the manufacture, processing,
                    distribution in commerce, use, or disposal of the
                    substance under section 6 or 7.

     Other;         AS proposed, if a  section 8 notice is filed for a
                    processed substance, the substance would not be added to
                    the TSCA inventory.  This may or may not result in
                    multiple submissions for the same processed substance.
A.4  PROPOSED PROCESSOR REPORTING RULE  (1) VS. PROCESSOR RULE WITH EXEMPTION
     FOR ONE-TIME PROCESSING  (5)

     Coverage;      Coverage  is identical, except that option 5 provides an
                    exemption for one-time processors, i.e., persons
                    qualifying for the exemption need not submit a section 5
                    notice but could be required to submit a section 8 notice,
                    as proposed.

     Other;         An exemption for one-time processing would allow the
                    processor to recover all or part of his cost on the
                    substance as an alternative to disposal.  Without the
                    exemption, the cost of filing a section 5 notice may
                    induce the one-time processor to choose disposal.

-------
                                      -208-
A.5  PROPOSED PROCESSOR NOTICE RULE  (1) VS. COMBINED SECTION 5 AND SECTION  8
     NOTICE  ("eT
     Coverage;
     Reporting
     Requirements!
     Regulatory
     Authority;

     Other:
Under  6, only selected chemicals would be subject  to
section  5 notification as proposed.  The remainder of
exempt substances would be subject to section  8  review.
Under  1, no distinction is made and all exempt substances
processed for a TSCA commercial purpose would  undergo
section  5 review.

EPA expects that under 6, a majority of the substances
will undergo section 8 review, so the difference in
coverage between options 1 and 6 is similar to the
difference in coverage between options 1 and 4.  In
particular, "small processors" would be exempt from filing
a notice.

Again, because the majority of substances are likely to
fall under section 8 review, the difference in reporting
requirements between options 1 and 6 will be virtually
identical to the difference between options 1 and 4.
Under  6, however, two different types of notices would be
submitted to the Agency for review.
Refer to comparison between options 1 and 4 in Section A.3.

Option 6 would impose a significant burden on the Agency
to select which chemicals should be subject to section 5
review.  These additional burdens hold true in all
comparisons between option 6 and each of the other
alternatives, but will not be repeated in other sections.
A.6  PROPOSED PROCESSOR NOTICE RULE (1) VS. MINIMUM GUIDANCE NOTICE (7)

     Reporting;     A notice submitted under the minimum guidance option would
                    contain less information than a section 5 notice as
                    proposed.  This is based on a comparison between ADL cost
                    estimates for a section 5 notice and a notice submitted
                    under the minimum guidance option.  AOL's results suggest
                    that it would take less time to fill out a notice under
                    the minimum guidance option than under the proposed
                    section 5 rule.  (See discussion of cost estimates in
                    section 3.6 of main text.)
A-7  JANUARY 10 PROPOSAL (2)  VS. NO REPORTING FOR COMMERCIAL PROCESSING OF
     EXEMPT SUBSTANCES "(3)
     Coverage;
Although the January 10 proposal does not permit separate
processors to submit section 5 notices, the Agency could

-------
                                      -209-


                    resort to subsections 720.10(a)(3) and (b)(2) to regulate
                    processed substances,  in other v/ords, EPA could require
                    notices from manufacturers or importers who intend to use
                    or distribute in commerce an R&D or test marketing
                    substance in a manner inconsistent with the exemption.
                    Under option 3, there would be no reporting for commercial
                    processing of exempt substances by separate processors.

     Reporting      The January 10 proposal required information from manufac-
     RequirementS!  turers who either process their R&D or test-marketed
                    substances for nonexempt commercial purposes or know their
                    substances will be processed for nonexempt commercial
                    purposes by someone else.  Alternative 3 requires no
                    information from processors, regardless of whether the
                    processors are manufacturers or not.

     Regulatory     EPA can regulate some processed substances under alterna-
     Authority;     tive 2 using sections 720.10(a)(3) and (b)(2).  Under
                    alternative 3, this authority would be removed.


A.8  JANUARY 10 PROPOSAL (2) VS. SECTION 8 PROCESSOR NOTICE (4)

     Coverage;      Under 2, processed substances undergo section 5 review if
                    they are R&D or test-marketing substances and are to be
                    processed for a TSCA purpose by the manufacturer.
                    Similarly, if the manufacturer knows that someone else
                    will process the chemical for a TSCA purpose, it must
                    undergo section 5 review.  Under 4, substances
                    commercialized by "small processors" are not subject to
                    notification; other exempt substances processed for
                    nonexempt commercial purposes would be subject to section
                    8 notification.

     Reporting      Less information would be reported under a section 8 rule
     Requirement;   than under the January 10 proposal.

     Regulatory     Under the January 10 proposal, EPA can initiate section
     Authority;     5(e) or 5(f) action against R&D or test-marketed
                    substances that are used or distributed in a manner
                    inconsistent with the exemption if EPA determines that the
                    processed substances would present an unreasonable risk.

     Disclosure;    Because less information is reported under a section  8
                    rule, the risk of disclosing confidential  business
                    information or processing intentions is lower.


A.9  JANUARY 10 PROPOSAL (2) VS. PROCESSOR RULE WITH EXEMPTION FOR ONE-TIME
     PROCESSING  (5)"

     Coverage;      Under 2, certain processed substances do not undergo
                    section 5 review.  Under option 5, they do,  unless they
                    are processed on a one-time basis.

-------
                                      -210-
     Reporting
     Requirements;
     Disclosure;
                    Under 2,  no notice is filed for the processing of certain
                    exempt substances.  Under option 5, most processors file a
                    notice virtually identical to the premanufacturing notice,
                    in addition to information regarding sources of the
                    substance.   If the substance will be used on a limited
                    basis, limited information must be filed.

                    Like the  proposed processor notice rule, option 5 raises
                    the issue of confidentiality in the event that the
                    processor must contact the manufacturer  or importer for
                    additional information.
     Reporting
     Requirements!
A.10 JANUARY 10 PROPOSAL (2)  VS.  COMBINED SECTION 5 AND SECTION 8 NOTICE (6)

     Coverage;      Under 2,  certain processed substances do not undergo
                    section 5 review.   Under 6,  selected substances would be
                    subject to section 5 review and the remainder (perhaps a
                    majority of the substances)  would be subject to section 8
                    review.

                    The January 10 proposal requires section 5 notices,  as pro-
                    posed,  from manufacturers who process their R&D or  test
                    marketed substances for nonexempt commercial purposes,  or
                    who know their exempt substances will be processed  for
                    nonexempt commercial purposes by someone else.   Under
                    option  6, section  5 notices  would also be required  of
                    certain processors.   However,  under option 6,  some
                    processors would be subject  to less extensive section 8
                    reporting.

     Regulatory     Regulatory action  allowed under the January 10 proposal and
     Authority;     under option  6 is  the same except when section 8 rules
                    apply under option 6.  When section 8 notices are
                    submitted, EPA cannot take action under sections 5(e)  or
                    5(f).

     Disclosure;     The risks of  disclosure could be the same under both
                    options when  section 5 rules apply to option 6.   Neverthe-
                    less, when section 8 notices are submitted under option 6,
                    the risks are likely to be lower because less information
                    would be  reported  and because manufacturer contact may be
                    unnecessary.
A.11 JANUARY 10 PROPOSAL (2)  VS. MINIMUM GUIDANCE OPTION (7)

     Coverage;       Option 2  does not permit processor reporting.   Coverage
                    under option 7 is the same as under the proposed processor
                    rule, i.e.,  persons who process an exempt substance for a
                    nonexempt commercial purpose must submit section 5 notices.

-------
                                       -211-
     Reporting
     Requirements:
     Disclosure:
Under  2,  the  separate processor  is not required to report.
Under  7,  the  processor  submits a notice  identical to
notices  submitted with  minimum guidance.

Under  both options,  the processor risks  the disclosure of
information submitted to  EPA.  In addition, under option
7, the processor risks  the disclosure of processing
intentions that he may  have wanted to keep confidential.
A. 12 NO REPORTING FOR COMMERCIAL PROCESSING OF EXEMPT SUBSTANCES  (3) VS.
     SECTION 8 PROCESSOR NOTICE  (4)
     Coverage:
     Reporting
     Requirements;

     Disclosure:
Under  3, there would be no reporting for commercial
processing,  under  4, exempt substances undergo section 8
review prior to commercial procesing for a nonexempt use.

The processor would file a limited section 8 notice under
option 4 and no notice under option 3.

Under  a section 8 rule, processors risk the disclosure of
whatever information is submitted to EPA.  under the option
of no  reporting, there would be no risks of disclosure.
A.13 NO REPORTING FOR COMMERCIAL PROCESSING OF EXEMPT SUBSTANCES  (3) VS.
     PROCESSOR RULE WITH EXEMPTION FOR ONE-TIME PROCESSING  (5)
     Coverage:
     Other
     Differences;
Under 3, there would be no reporting for commercial pro-
cessing,  under option 5, a processor submits a section 5
notice as proposed in October unless he qualifies for a
one-time processing exemption.
Same as differences described in A.2.
A.14 NO REPORTING FOR COMMERCIAL PROCESSING OF EXEMPT SUBSTANCES  (3) VS.
     COMBINED SECTION 5 AND SECTION 8 NOTICE  (6)
     Coverage;
     Reporting
     Requi rements:
     Regulatory
     Authority;
Under 3, commercial processors of exempt substances are
not required to submit preprocessing notices,  under 6,
some processors submit section 5 notices as proposed in
October and the rest are subject to section 8 review.

There are no reporting requirements under 3.  under 6,
most processors will be required to submit limited section
8 notices.  A few will submit section 5 notices as
proposed in October.

Option 3 provides no mechanism for regulating processed
substances.  Option 6, on the other hand, permits enforce-
ment under TSCA sections 5(e) and 5(f)  in a few cases.

-------
                                     -212-
     pisclosure:    Option 3 poses no risk of disclosing confidential business
                    information.  Option 6, however, would pose such a risk in
                    the event that manufacturer or importer contact become
                    necessary to satisfy the information requirements of the
                    relevant notice form (i.e., section 5 or section 8 notice).
A.15 NO REPORTING FOR COMMERCIAL PROCESSING OF EXEMPT SUBSTANCES (3) VS.
     MINIMUM GUIDANCE OPTION (7)
     Coverage;


     Reporting
     Requirements;
     Regulatory
     Authority;
     Disclosure:
Under option 7, the processor submits a notice,  under 3,
no processors submit notices for processed substances.

Under 3,  no information is submitted to Agency.  However,
under option 7, a processor will submit a notice that is
consistent with the minimum guidance option.

EPA may regulate substances under sections 5(e) and 5(f)
of TSCA under 7.  under 3, these regulatory mechanisms
would not be allowed.

Under 7,  the processor risks disclosure of business
information reported under the minimal guidance policy.
in addition, he risks the disclosure of processing
intentions should manufacturer contact becomme necessary
in order to complete a notice.
A.16 SECTION 8 PROCESSOR NOTICE (4)  VS. PROCESSOR RULE WITH EXEMPTION FOR ONE-
     TIME PROCESSING (5)

     Coverage;      Option 4 exempts "small processors" from filing a section
                    8 notice while option 5 exempts one-time processors from
                    filing a section 5 notice as proposed.

     Reporting      under option 4,  the section 8 notice may be filed less than
     Requirements;  90 days before processing,  under option 5, the proposed
                    section 5 notice must be filed at least 90 days prior to
                    processing.  AS noted elsewhere, the section 8 notice
                    requires much less information than the proposed section 5
                    notice.

     Regulatory     EPA could not regulate the manufacture, processing, distri-
     Authority;     bution in commerce, use, or disposal of an exempt substance
                    under a section 8 type rule.  However, under a section 5
                    notice rule, the Agency could take action under sections
                    5(e) and 5(f)  during the review period.
A.17 SECTION 8 PROCESSOR NOTICE (4)  VS. COMBINED SECTION 5 AND SECTION 8
     NOTICE (6)

     Coverage;      A section 8 rule would exempt "small processors" from
                    reporting.  Option 6 would also exempt "small processors".

-------
                                      -213-
     Reporting
     Requirementsi
     Regulatory
     Authority;
                    in addition, selected processors would file section 5
                    notices.

                    Except when the proposed section 5 must be submitted,
                    option 6 would involve reporting the same information as
                    alternative 4.

                    If option  6 were selected, EPA would be authorized to
                    regulate substances under sections 5(e) and 5(f) in a few
                    cases.
A. 18 SECTION 8 PROCESSOR NOTICE  (4) VS. MINIMUM GUIDANCE OPTION  (7)
     Coverage;
     Reporting
     Requirements:
     Regulatory
     Authority;
                    Section  8 rules may apply to all processors except those
                    with  "small business" status.  The minimal guidance notice
                    would cover all processors regardless of size.

                    Based on ICF estimates of submitting a section 8 notice and
                    ADL estimates of submitting a notice with minimum
                    quidance, it will cost more to provide information under
                    option 4.   (See specific estimates in Section 5.1.4 of
                    text.)

                    TSCA sections 5(e) and 5(f) apply to the minimim guidance
                    option but not to section 8 notices.
     Reporting
     Requirements:
A.19 PROCESSOR RULE WITH EXEMPTION FOR ONE-TIME PROCESSING  (5) VS. COMBINED
     SECTION 5 AND SECTION 8 NOTICE  (6)

     Coverage;      under option 5, only the one-time processor is exempt from
                    reporting.  All other processors would submit section 5
                    notices,  under option 6, a few processors would submit
                    section 5 notices while the remainder would be subject to
                    a section 8 rule which exempts "small processors".

                    Except when a section 8 notice is submitted under option 6,
                    reporting requirements would be the same for both
                    alternatives.  When a section 8 notice is submitted, the
                    differences are similar to those between options 1 and 4.
                    That is, less information would be required under a
                    section 8 rule, and the processor may submit a notice less
                    than 90 days before processing.

     Regulatory     under option 6, EPA could take regulatory action against a
     Authority;     substance in only a few instances, whereas under option  5
                    EPA would be able to regulate most substances during the
                    review period.

     Disclosure;    Where the section 8 rule applied to option 5, the
                    processor would risk disclosing less information than
                    under a section 5 rule.

-------
                                      -214-
A.20 PROCESSOR RULE WITH EXEMPTION FOR ONE-TIME PROCESSING (5)  VS. MINIMUM
     GUIDANCE OPTION (7)

     Coverage;      Coverage under both options is almost identical except
                    that under 5 an exemption is provided for persons who
                    intend to process substances on a one-time basis as an
                    alternative to disposal.  Persons qualifying for an
                    exemption will probably be required to submit section 8
                    notices.

                    A minimal guidance notice requires less time to complete
                    than a section 5 notice as proposed in October.  (See
                    section 3.6 of text for relative cost estimates.)   When
                    the one-time exemption applies and a processor submits a
                    section 8 notice, the minimum guidance option takes longer
                    to complete than a section 8 notice.  (See section 5.1.4
                    of text.)

A.21 COMBINED SECTION 5 AND SECTION 8 NOTICE (6)  VS. MINIMUM GUIDANCE
     OPTION  (7)
Reporting
Requirements;
     Coverage:
     Reporting
     Requirements;

     Regulatory
     Authority;
               Coverage under both options is the same when section 5
               notices are submitted under option 6.  However, when
               section 8 notices are submitted, coverage differs because
               "small processors" are exempt from reporting.
               Same as differences described in A.20.
               Same as differences described in A.20.

-------
                                      -215-
                                    CHAPTER 9

                              POSSESSION OR CONTROL
     Under section  5 of the Toxic Substances Control Act  (TSCA), when firms
submit a premanufacture notice to the Environmental Protection Agency (EPA),
they must include with it any test data.53/ in their possession or control
which is related to the health or environmental effect of any manufacture,
processing, distribution in commerce, use, or disposal of the substance in
question.54/

     in order to determine the scope of  information that  firms are responsible
for submitting, EPA has proposed three alternative definitions for "possession
or control."  in this paper, the costs to industry of these alternatives will
be assessed.  The assessments will be based on information in the public
record, on literature about the state of the art of chemical information
science,j>5/ an(j on telephone interviews with chemical information scientists
and experts in transnational data flow problems.

     in the following sections we will:

     •    specify the options under consideration;

     •    specify the type of costs and magnitude of costs incurred
          by industry under each option; and

     •    estimate the relative burden of each option to  industry.
    ^3/Test data is defined as "data including chemical identity from a
formal or informal study, test, experiment, recorded observation, monitoring
or measurement; and information concerning the objectives, experimental
methods and materials, protocols, results, data, analyses  (including risk
assessments), and conclusions from a study, test, experiment, recorded
observation, monitoring or measurement."  Federal Register, Vol. 44, No. 7,
January 10, 1979.

    li/TSCA, section  (5)(d)(1)(B).

    JiVchemical information science is the study of how to classify and
organize chemical information.

-------
                                      -216-


POSSESSION OR CONTROL OPTIONS

     EPA has provided three alternative definitions for "possession and
control."  All of them define the sources of data as:

     •    company files;

     •    commercially available data bases to which the company has
          purchased access (called commercial data bases);  and

     •    files maintained in the course of employment by employees
          or agents associated with the commercial development of
          the new chemical substance (called employee files).

The definitions also include the concept of a parent company.   A parent
company is considered to have control over another company  if  it owns or
controls 50 percent or more of the other company's stock.56/

     The options differ on two issues.   The first is whether submitters who
are not parent companies have control of the files of their parent companies
and the files of other subsidiaries of their parent companies.  We call this
the "one-way vs. two-way" issue.  If. a company is considered to have no
control over the files of its parent or of its parent's subsidiaries, then
"possession and control" only creates a reporting responsibility in one
direction:  the parent is responsible for its subsidiaries' files.  If a
company is considered to have control over these files, then "possession or
control" creates a reporting responsibility in two directions:  the parent and
subsidiaries are defined to have access to each other's files.  Option 1 and
option 3 are two-way options and option 2 is a one-way option.

     The other issue is whether companies are defined to have  control over
files of associated companies whether or not they were involved in the
commercial development of the chemical.  This is called the present venture
issue.  Many companies in the chemical industry have numerous  subsidiaries
and/or sister subsidiaries.  Rarely, if ever, will all of them participate in
the development of a chemical.  The issup becomes, should all  companies be
required to search through their files whether or not they  participated in the
present venture?  Option 1 and option 2 are present venture options and option
3 is not.
    JL§/For financial purposes,  APB Opinion 118 (March 1971)  defined control
as the ability of the investing company to determine the operating and
financing policies of another company in which they own shares of the voting
stock.  Control is assumed when the investing company owns over 50 percent of
the stock.  The Export Administration Regulations dealing with restrictive
trade practices or boycotts (43 FR 3508)  use a much broader definition, but
they use "control in fact" rather than just "control".

-------
                                      -217-


Option 1;  The Two-Way  Present Venture Option

     Test data 'in the  "possession or control" of the submitter would include
test data in the possession or control of any of the following as long as they
were associated with the submitter in the research and development, test
marketing, or commercialization of the substance:

     •    the submitter's subsidiary;
     •    the submitter's parent company; and
     •    a subsidiary  of the submitter's parent company.

Included among sources  of data are commercial data bases, company files, and
employee files for each of the above parties.

     This is the option proposed by EPA in the January 10 proposal.  It is a
two-way option in that  reporting requirements are the same whether or not a
parent or its subsidiary submits a notice, since both are defined to have
access to each other's  files.  it is a present venture option in that
"possession or control" is limited to companies involved in the present
venture of commercializing the chemical in question.

Option 2;  The One-Way  present Venture Option

     Under this option, information in the "possession or control" of the
submitter would only include information in the possession or control of the
submitter and its subsidiaries.  The submitter would be presumed not to have
access to its parent's  files or its parent's other subsidiaries' files.  AS in
option 1, the definition is limited to those companies associated with the
submitter in the research and development, test marketing, or
commercialization of the substance.  AS in option 1, the sources of
information may be commercial data bases, company files, and employee files.

Option 3;  The Two-Way  Non-Present Venture Option

     AS in option 1, test data in the "possession or control" of the submitter
would include test data in the possession or control of the submitter's
subsidiaries, its parent, and any other subsidiaries of its parent,  unlike
option 1, the definition is not limited to the present venture.  All these
companies must check through their files to see if they have ever done any
work related to the chemical in question whether they are involved in the
current venture or not.  As in options I and 2, sources of data include
commercial data bases, company files, and employee files.

     under all three options, EPA may elect to provide some mechanism to allow
the submitter to demonstrate good faith in attempting to obtain the
information required from companies over which the submitter has no real
control.  This especially applies to foreign subsidiaries and parents.  We
will look at the economic impact of each option with and without this
mechanism.

-------
                                     -218-



                                   EXHIBIT  9-1

                           DIFFERENCES BETWEEN OPTIONS


                                    One-way                    Two-way

Present Venture                    Option 2                    Option 1

Non-Present Venture                                            Option 3



INDUSTRY COSTS

     This section discusses the types of costs faced by industry.   It includes
a discussion of the factors affecting the magnitude of cost and then analyzes
the five kinds of costs:

     (1)  costs of gathering data from company files;

     (2)  costs of gathering data from commercial data;

     (3)  costs of gathering data from employee  files;

     (4)  costs associated with international data flow; and

     (5)  intangible costs.

     The first four kinds of costs are all direct out-of-pocket costs.
intangible costs encompass costs resulting from  delays in the introduction of
new chemicals, trade secret disclosure,  and uncertainty.  All costs are pre-
submission costs, but they each have their counterpart post-submission cost.
For example, EPA may require supplemental data the scope of which is
determined by "possession or control".   Potential confidentiality costs are
mostly post-submission costs but some are pre-submission costs.

Factors Affecting the Magnitude of Costs

      For any given company, there are three factors that can affect the
magnitude of costs.  These are:

     •    the size of the company;

     •    the sophistication of its chemical information system; and

     •    the definition of possession or control that EPA chooses.

     The size of the company affects the quantity of information that must be
gathered.  A large company requires more time to gather information than a
small company with the same information system.   This occurs because large
companies have more employees and more projects, and they may perform more
thorough tests on new chemicals than small companies.

-------
                                      -219-


     The sophistication of  a company's  chemical  information system can
dramatically  affect  the costs  of  gathering  information  in its "possession or
control",  with  a modern, computer-based  information system, a large
multinational chemical company can gather all of the information in this
possession or control in  less  than a day.   At the other extreme, a company
with a crude  manual  system  may require  many days to gather the same
information.

     A company with  a modern system is  able to store a  large number of
document titles  and  classify them by chemical characteristics, uses, or any
other category.  This allows the  company  to find the documents pertaining to a
particular subject quickly.  For  a survey of the state of the art of chemical
information science, see  the appendix to  this chapter.

     It should be expected  that large companies will have better information
systems than  small companies.  They have more capital to invest in a large
information system,  and they are  able to  hire the technical people necessary
to  install and maintain such a system.  Furthermore, because of the size of
their data bases, they have more  incentive  to install a sophisticated system.

     Since the level of sophistication  of a company's information system will
affect the costs of  gathering  information in "possession or control,"
companies will have  an incentive  to improve their information systems.  This
does not mean that without  TSCA no company would have improved its information
system.  According to chemical information  science experts,  companies were
radically improving  their information systems long before TSCA; improvement
was necessary to keep up with  the growth of information.  But TSCA increased
the incentive to improve  information systems.  Therefore, when analyzing the
costs of "possession or control," we must consider the state of the industry
after being affected by these  costs.  We must also consider the costs of
providing for these  changes.

     For the  purposes of  this  analysis, we will consider three classes of
companies in  the chemical industry.  Class 1 companies will include those
companies with crude information  systems that decide not to improve them
despite the demands  of TSCA.   The only  costs they bear will be those connected
with individual  notices.  AS time goes  on, these companies will improve their
systems, but  not because of TSCA.  Class  2 companies will include those
companies that decide to  improve  their  information systems.   They will incur
lower notice  costs,  but they will also  spend money to install and maintain
their new systems.   it should  be  noted  that only those costs that will not be
incurred without TSCA will  be  considered a cost of TSCA.  Since most companies
will eventually  improve their  information system with or without TSCA, only
part of the installation and maintenance costs will be attributed to TSCA.
For example,  if  a company decides to install a computer five years earlier
than it might have because of  TSCA, then the costs attributed to TSCA should
be the cost of installing the  system minus the cost of  installing it five
years later (properly discounted), plus the cost of maintaining the system for
five years.   If  the  company installs a  system solely meant to deal with TSCA,
then it should be totally costed  to TSCA.  An example of this might be a
special program  used to gather and organize TSCA-related material.  Class 3
companies will include those companies  that already have modern systems and

-------
                                      -220-


would have continued updating them without TSCA.  Like Class 1 companies, they
will only incur costs connected with each notice.  NO installation or
maintenance costs should be attributed to TSCA because TSCA is not the
motivating force in installing the system.  But, such costs will be much lower
than those of class 1 companies.

     All companies are to some extent level 2 companies and will have to
adjust their information systems to some degree to deal with TSCA.  They will
differ in the degree of system improvement they make.

     The last factor affecting costs is the definition of "possession or
control" that EPA chooses.  Because the only difference between the options is
the number of corporate entities that must supply information, we can consider
the cost of gathering data for each corporate entity and add them together.
To this sum, we should add the cost of transmitting data between entities to
derive the total cost of each option.  For example, if a company has five
subsidiaries, the cost of gathering information for each subsidiary and the
parent can be determined separately, then added together.  The cost of sending
the data to the parent is then added, to get the total cost of gathering data.

     Thus, the approach here is to determine the costs of gathering data for
each "level" company.  The cost of each option will become a function of this
cost and of the number of corporate entities subject to data-gathering
requirements.

     Because small firms with no subsdiaries or parents are not be affected by
the definition of possession or control, they will not be considered here.

Cost of Gathering Data from Company Files

     The first cost of gathering data is the cost of gathering data from the
company files maintained by the section 5 notice submitter.  All companies,
whatever the sophistication of their information retrieval system, incur some
variable costs connected with each submission.  "Level 2" companies also incur
some fixed costs connected with installing and maintaining a new information
system.

     Class 1—Crude information Systems.  All companies are assumed to have at
least an indexed manual filing system for their documents,  without an indexed
filing system, an employee would have to search through half the existing
documents, on average, each time he needed one.  If documents are only stored
by subject, time requirements are significantly lower than they would be
without any indexed system.

     For firms with manual systems, costs of gathering data can be broken into
the cost of searching for documents and the cost of gathering and preparing
documents.  if the index does not list documents, then a worker with a
technical background would have to search for and gather the necessary
documents, and a clerk would prepare them.  if a document index does exist,
then a technical worker may have to search through the index, but a clerical
worker could do the rest.

-------
                                      -221-


     Companies may  also  have  computerized  information  systems that are not
well suited for  TSCA.   in  the worst  case,  the cost of  searching for and
gathering data would  be  the same  as  that for manual  systems.  However, it
would probably be much less because  the time required  to search through an
index would be significantly  less.   The cost of  running the computer for the
time needed to do a search would  also be included, but this would be very low.

     It should be remembered  that, even without  TSCA,  most companies are
improving their  information systems,  whatever the cost of gathering
information in company files,  it  will decrease with  time.

     Class 3—Sophisticated information Systems,  companies with sophisticated
information systems will bear the same type of costs as Class 1 companies, but
the costs will be lower.   Some companies now have systems that allow a
technical worker to search through a company's document titles by computer in
fifteen minutes; and  they  have clerical workers  who  can gather the documents
in less than a day.   Thus, the cost  becomes insignificant.

     As computers become able to  store data more efficiently, firms may become
able to store the actual documents,  as well as the titles, on computer.  This
would reduce the costs of  gathering  data from company  files even further.

     Class 2—Improved information Systems.  Class 2 companies are those that
install sophisticated systems in  order to  deal effectively with the reporting
requirements of  TSCA.  The cost of gathering data for  a notice will be the
same as that for a  Class 3 company.  But there will  also be some fixed costs
connected with installing  and maintaining  the new system.  They are fixed in
that they do not depend  upon  the  number of notices submitted.

     The total value of  these fixed  costs  is equal to  the discounted value of
each year's expenditure  on fixed  installation and maintenance, minus each
year's non-TSCA  benefits derived  from the  new system.  It must be remembered
that eventually  the company would have installed a new system (not necessarily
the same one)  anyway.  Therefore  the expenditures for  installing and
maintaining the  system later  must be deducted from costs, while the savings
derived from the later system must be deducted from  benefits.  This means that
the only significant costs and benefits are those accrued between the time the
TSCA-motivated system was  installed  and the time another similar system would
have been installed in the absence of TSCA.

     The costs associated  with installation include  costs of buying hardware
and/or software,  installation  costs, programming costs, and testing costs.
The cost of the  system that would have been used in  the absence of TSCA should
be subtracted from  this  after  being  properly discounted.

     The costs of maintaining  the new system are only  equal to the marginal
costs.  If no new employees are needed to maintain the system, then there are
no costs.  For those companies that  had no computer  system, maintenance costs
will be sizable.  For those that  had computer systems  and only had to add
additional programs, the maintenance costs should only be equal to the costs
of maintaining the  new programs and data files.

-------
                                      -222-


     The net present value of the fixed cost of systems that are installed
because of TSCA is not currently known, but it is negative.  If it were
positive, companies would have developed the systems without TSCA.  It also
seems that in most cases the negative number will be a relatively small one
because in a few years the company probably would have instituted the changes
without TSCA.  This would seem to indicate that the improved system has a
positive rate of return, but that other projects have higher rates of return.
The rate of return rises as the information demands of the company increase
and as other companies solve some of the "start-up" problems.

Cost of Gathering pata from Commercial Data Bases

     The second source of information that the submitter must search is the
commercial data bases to which it has purchased access.  The fact that the
submitter has access to the files implies it has access to the programs
necessary to use the data bases efficiently; the two are sold together.   The
cost of searching through those files depends upon the size of the data banks,
the efficiency of the commercial programs used to access them, and the cost of
computer time.  Carol Haberman from the Toxicology Data Bank estimates that,
at most, 15 minutes would be required to get information from all data bases
on a specific chemical. $2/  During peak hours, computer time costs $15.00
per hour.^8/  it costs 12 cents to print a paqe of output.  Toxicology Data
Bank provides access to a number of different chemical data banks (a
reasonable upper limit on the number of data banks to which any one firm would
have access).  The cost of gathering data is equal to the computer costs plus
the labor costs and should be minimal, considering the time involved.   If the
company had already generated a list from the commercial data bank in the
course of researching the chemical, then it does not have to generate the list
again and the cost is zero.

     What will happen to these costs over time is not certain,  whether
computer costs will increase or decrease is not clear, nor is it clear how
user firms will react to the increase in the number of available data banks.
    JLZ/Telephone conversation with Carol Haberman, Technical Information
Specialist, Toxicology Data Bank, April 30, 1980.

    .P_§/The cost of computer time is expected to increase soon.

-------
                                      -223-


Cost of Gathering  information  from Employee Files

     The last source of data is files maintained by employees in the normal
course of employment.  One method of collecting data would be for the company
to publish in prominent places of a newsletter or post a request for pertinent
information.  The  costs associated with this method would be the cost of
preparing a news item and posting it if necessary, the cost of each employee
searching through  his files, and the cost of preparing documents for the
public record.  These costs will depend upon the number of technical workers
and the condition  of their files.

     Because the definition of files maintained by employees is broad in the
proposed regulation—it includes such information as conversations between
employees—the employee files  would be the most expensive source to search.
It also involves the greatest  chance of neglect in collecting relevant data,
and it will be the most difficult part of the requirement for EPA to enforce.

Post Associated with international Data plow

     The notice information requirements may cause significant problems for
those companies with foreign subsidiaries or parents.  in the last few years,
a number of laws have been passed to restrict the flow of information across
national borders.  Most of these have been written as privacy laws, but they
have been used to  restrict the flow of information harmful to a nation's
security.  Fourteen industrial nations, including all of our major trading
partners have such laws.j>V  Thus, test data could be restricted for
national security  reasons.

     The scope of  these laws is unclear.  Some experts feel there are no laws
that directly restrict the flow of scientific information.^/  Chris Vizas
of the House Subcommittee on Government Information and individual Rights
cites a statute recently passed in Great Britain, the Protection of Trading
interest Act of 1980, specifically dealing with commercial information which
includes scientific information.j>!/  Vizas also points out that the
international Traffic in Arms  Regulations of this country have been used in
the past to restrict the flow  of scientific information out of the country.
The ban on technology for the  Soviet union is a case in point.
     IE/Telephone conversation with  Alden  Heintz,  Vice  President,  industrial
Corporate operations,  Tymshare Corporation, May  2,  1980.

     l£/Telephone conversations with Alden Heintz,  Tymshare  Corporation, May
2, 1980; Geza  Feketekuty, U.S. Trade Representative's office,  Assistant U.S.
Trade Representative  for  Policy Development,  May 1,  1980;  Tim  Donovan,
Executive Director, Transnational  Data  Reporting Service,  May  8,  1980.

     H/Telephone conversation with  Chris  Vizas, May 2,  1980 and  May 6,  1980.

-------
                                     -224-


     All the experts we consulted agree that a great potential exists for
restrictions of scientific data flow.   Vizas mentions stronger enforcement of
existing laws, and all the experts note the potential for new stronger laws.
One expert feels that restrictions will become worse before they get better.
He says no one has thought through the  real costs of restricting information
in order to protect domestic industry and preserve secrets, and few have
looked at the increased costs of doing  business created by this practice.

     The costs of transmitting foreign  data are also being kept artificially
high by restrictions on the telecommunications and computer industries in many
countries.  Japan, prance, Germany, and Brazil (as well as others)  all
restrict the use of foreign telecommunications and computer systems.62/

     Gathering information from foreign companies could involve significant
costs, depending on how foreign nations decide to handle the whole issue of
transnational data flow.  The costs will depend greatly upon what standards
EPA sets.

intangible Costs

     Three intangible costs are connected with submitting information for
section 5 notices.  The first is the cost of a time delay caused by the
information gathering process.  This should be negliglible, except when
foreign regulatory agencies are involved.  Foreign governments may restrict
the flow of information for months.  The "good faith" standard that EPA
establishes will determine the cost of  this time delay.63/

     The second cost is the cost of the uncertainty of whether EPA will
consider the company's effort a good faith effort.  This cost will be
determined by the policies of EPA.  AS  time passes and companies determine
what constitutes a good faith effort,  these costs should decrease.

     The final cost is the cost of disclosing confidential data.  Many
industry members feel this will be the  most significant cost.  The cost of
trade secret disclosure may increase when data requested by EPA is in a
foreign country.  The foreign government may require the company to submit the
information to its agencies for review to determine that disclosure will not
be in violation of its laws.  Thus, the chemical company will provide the data
to two governments instead of one, increasing the possibility of disclosure.
     ^/Telephone conversation with Alden Heintz, Tymshare, May 2, 1980.

     ^I/This should be considered in some of the issues, there is a cost in
a time delay caused by EPA extending the review period under section 5(c).
But since the definition of "possession or control" will not affect the
magnitude of this cost, it need not be considered in this issue.

-------
                                      -225-


Summary of Posts

     With the exception of the costs of transnational data flow,  all the above
costs are incurred by single corporate entities.  TO determine the cost of
each option, we need only decide how many corporate entities are  covered by
the option.  The cost of transnational data flow is the cost of sending data
from foreign companies to an associated domestic submitter.   The  cost of
sending data from one domestic company to another will not exceed postage
costs which are negligible.

     Costs of collecting data for a corporate entity can be divided into
tangible  (direct out-of-pocket) and intangible costs.  The tangible costs are
modest with the possible exception of gathering data from employee files.   The
intangible costs may be very high.  They will depend on the policies of EPA
and of foreign regulatory agencies.
POST OF OPTIONS

Option 1;  The Two-way Present Venture Option

     Under option 1, the submitter must provide information from the files of
any of the following companies if they were associated with the submitter  in
the research and development, test marketing, or commercialization of the
substance for which a notice is being submitted:

     •    the submitting company;

     •    its subsidiaries;

     •    its parent company; and

     •    any subsidiaries of its parent company.

     The cost of option 1 is the cost of gathering data from each company's
corporate files, each company's commercial data files, and each company's
employee files.  The cost does not depend on who submits the notice.

     The cost of gathering data from each company's corporate files will
depend upon the level of sophistication of each company's information system,
each company's role in the development of the chemical, and the degree of
integration in the information systems between corporate entities,  without
integration between systems, the total cost is equal to the cost of gathering
data from corporate files plus the cost of sending the data to the submitter.
(The cost of gathering data has been discussed in detail.)  The cost of
sending it may include telecommunications costs, mail costs, and the cost  of
time delays.   if any corporate entities sending data are in foreign nations,
it may also include the cost of filing for permission to send data across
national borders.  This includes the actual cost of filing, the cost of lost
confidentiality  (in many cases, to send data across national borders, the
contents of those data must be revealed), and the cost of uncertainty.  If a
foreign government does not allow information to be sent, then the cost may
range from the cost of filing for permission to send the data to the cost of

-------
                                      -226-


having a notice rejected, depending on how EPA defines good faith efforts in
such situations.  if the companies' information systems are totally
integrated, costs may be dramatically reduced.  Some multinationals store
copies of all research data in a central library,   with a modern information
system, the cost of gathering data for all corporate entities approaches the
cost of collecting it for only one entity.

     If companies are working together on a project, it is reasonable to
expect communication between them and individual company familiarity with the
other companies' progress.  If all companies involved in a venture are given
copies of all relevant work, then for notice purposes, there is total
integration.

     The cost of gathering data from commercial data files is similar to the
cost of gathering data from each corporate entity's commercial data files plus
the cost of sending the data to the submitter.  If two entities have access to
the same commercial data file, then only the one with the lower transmittal
costs needs to gather and send the data.  If all entities have access to the
same data files, then the cost for all entities is reduced to the cost for one
entity.

     If all entities are domestic, then total cost should be minimal.  If some
entities are foreign, then they may have access to other commercial data
bases.  This could increase costs in the short-run because of transnational
data flow problems.  Again, the "good faith" standard which EPA establishes
becomes important, but there has been a trend towards the development of
international data bases.  Germany and some other  EEC countries are in the
process of joining the major American data bank, the Chemical Abstracts
Service.  As this trend continues, all companies will gain access to the same
data banks and will reduce costs associated with transnational data flow in
the long-run.

     The cost of gathering information from employee files will again be equal
to the sum of gathering information for each entity plus the cost of
transmitting the data.  Little can be done to integrate employee files.
Transmittal costs will still be minimal for domestic entities but may become
very high for foreign entities.  Most restrictive  foreign laws are directly
aimed at "personal data".  Because information from employee files is somewhat
personal, it may be difficult to transfer.

     Each piece of information that is submitted has a potential cost of lost
confidentiality.  The larger the number of companies that must submit data,
the higher the total costs of lost confidentiality.  Furthermore, if foreign,
associated companies are required to expose their  data to foreign regulatory
agencies before they can send it to their domestic submitter, the cost of lost
confidentiality would be high no matter what rules EPA establishes for
confidentiality.

     The only estimate left to be made is the number of companies involved in
the development of a new chemical.  in many cases, only one entity is involved
in the process.   in other situations, a few entities will be involved.  For

-------
                                      -227-


example, Eastman Chemical does a great deal of joint research with Eastman
Kodak research labs, and ICI Americas receives technical help from its British
parent.  The final possible joint  research relationship involves companies
such as Exxon Chemical and Chevron Chemical.  Most research done for these
companies is done by another subsidiary of their parents.  in most cases, the
total number of companies exposed  to reporting requirements should be limited
and should only include companies  with a  reasonable probability of providing
relevant data.

Option 2;  The Qne-Way Present Venture Option

     In option 2, the submitter must only provide information in its possession
or control or in its subsidiaries'  possession or control.  As in option 1, only
those companies involved in the present venture must provide information.  The
cost of gathering information now  depends upon who the submitter is.  This
section will refer to companies by the letters in Exhibit 9-2.
                                   EXHIBIT 9-2
     Total costs may again  be  broken  down  into costs of gathering data in
company files,  in commercial data  files, and  in  employee  files.  AS in option
1, the total cost is equal  to  the  sum of the  individual costs of gathering
data and of sending it to the  submitter for each entity.  The significant
difference between options  1 and 2 is the  number of entities that are subject
to reporting requirements.

     If company P is the submitter, then the  cost  is the  same for option 1 and
option 2.  But  if company A is  the submitter, only companies A, C, and D must
provide information.  The reduction in cost is the cost of gathering and
transferring data from companies P, B, E,  and F.  Costs may be  further reduced
if company P takes control  of  all  commercial  data  banks.  Company P can let
its subsidiaries use the data  banks,  but it does not have to search through
the banks.

     The  submitter of a notice  is  the manufacturer of  the new chemical.  Many
chemical  companies are organized by product lines, each product line being
manufactured  by a different subsidiary or  division.   If companies have
divisions rather  than subsidiaries, then option  1  and  option  2  have the same

-------
                                      -228-
costs.  But if companies have subsidiaries, then option 2 could be much less
costly.  There may be some incentive for companies to change their divisions
into wholly owned subsidiaries if this option is adopted.
Option 3:  The Two-Way Non-Present Venture Option
     According to option 3, the submitter is presumed to be in possession or
control of its own files,  its subsidiaries' files, its parent's files, and its
parent's other subsidiaries'  files,  even if some of them had nothing to do
with the development of the new chemical.  Again total costs are equal to the
sum of costs of gathering  data and transmitting it for each entity.  But the
number of entities subject to reporting requirements significantly increases.
using option 1 the number  of entities that must report data appears to be
between 1 and 4.  Exhibit  9-3 gives the number of entities that must report
for a number of companies  under option 3.
                                  EXHIBIT  9-3
                 NUMBER OF ASSOCIATED  COMPANIES UNDER OPTION  3
             Company
Amax
Akzona
Amoco Chemical
Ashland Chemical
BASF Wyandotte
Celanese Corporation
DuPont
Engelhard Mineral & Chemical
   All         Number of Entities That
Companies   Must Report Chemical Companies
    42                    2
    17                   12
    55                    1
    25                    0
    21                   21
    22                   22
    41                   41
    32                    4
Note:  These are conservative estimates taken from Standard & Poors and
       the National Register Publishing Company's Corporate Affiliations.
Summary
    Possession or control is an indicator of the depth to which submitters
must reach in filing a section 5 notice.  The more deeply the submitter must
reach,  the greater are the pre-submission costs in both time and money.  Thus,
pre-submission possession or control costs increase as the procedure
progresses from a one-way present venture option to a two-way non-present
venture option.

-------
                                      -229-
     The rationale for post-submission possession or control costs parallels
the logic presented previously.   Potential post-submission, out-of-pocket
costs decrease with the increasing depth of possession or control because the
more information  is required at  time of submission, the less post-submission
information will  be required.

     Potential, post-submission  delay decreases as the depth of possession or
control increases because delaying actions are less likely to be invoked when
there is additional information  provided initially.

     The potential for trade secret disclosure increases with the scope of
possession or control.  The relationship between potential restrictive actions
and possession or control is similar.  The more deeply submitters are required
to reach, the more likely it is  that some submitters will choose not to enter
the market because the perceived notice requirements are too great.

-------
                                      -230-
                                    EXHIBIT 9-4

             COMPARISON OF COSTS  ATTRIBUTABLE TO POSSESSION OR CONTROL
                                    OPTION 1
                                    Two-way
                                Present Venture
   OPTION 2
   One-Way
Present Venture
   OPTION 3
 Two-Way Non-
Present Venture
Pre Submission Costs

Out-of-Pocket

  Company files                        2
  Commercial data files                2
  Employee files                       2
  Transnational data flow              2

Intangible Costs

  Delay                                2
  Potential trade secret disclosure    2
  Uncertainty                          2
       1
       1
       1
       1
       1
       1
       1
Post Submission Costs

  Potential out-of-pocket              2
  Potential delay                      2
  Potential trade secret disclosure    2
  Uncertainty                          2
  Potential restrictions               2
       3
       3
       1
       3
       1
       1
       1
       3
       1
       3
A "1" indicates the least costly alternative.

-------
                                      -231-
                             APPENDIX TO CHAPTER  9;

                 STATE OF THE ART—CHEMICAL  INFORMATION SCIENCE
     During the last four decades, much work has been done in the chemical
industry on chemical information systems.  Some of the major milestones
include the introduction of a number of notation systems, most notably the
Wiswesser System, and the implementation of a number of  indexing systems, most
notably key word  (KWIC and KWOC) indices.  Computers have become a necessary
tool in the storage, indexing, and retrieval of all information.  Today,
companies can store internal report citations and index  them by key word,
chemical characteristics, or uses.  With the possible exception of access to
the files maintained by employees, all of the demands of the PMN information
requirements can easily be satisfied using modern technology.

     Most large chemical companies have already developed sophisticated
information systems.  For example, one company has centralized all documents
at its corporate headquarters.  it has more than 50 subsidiaries, most of
which are in foreign countries.  But because the files are centralized, it
would take a maximum of one day, largely spent pulling listed documents, to
gather information for a PMN.

     Another company, a subsidiary of a major foreign chemical company, has
tried to integrate the present TSCA regulations into its information systems
and has anticipated what EPA (as well as other government agencies such as
OSHA) will require.  This company set up an internal computer system to deal
with the TSCA Inventory requirements.  It identified the relevant chemicals
and kept records on the reporting status of each one.  Presently, physical
properties are not stored on the system but will be added in the future.  The
company has easy access to the information of its parent that is relevant to
its own areas of business.  It could easily comply with option 1 and 2.

     Although there are many large firms not as well equipped as the two
profiled above, they all should have sophisticated information systems.  Most
of their information systems are constantly being updated and adapted to TSCA
regulations.

     Many companies have written articles about their information systems.
These include:

 (1)  Benson, Frederic R. (ICI Americas), "Research information at ICI united
     States, "Journal of chemical Documentation, 1974.

 (2)  Skolnik,  Herman (Hercules, inc.), "The Hercules Technical Information
     Division:  Services Special Systems, and R&D," Journal of Chemical
     Documentation, 1974.

-------
                                      -232-


 (3)  Schultz, John L. (E.I. du Pont de Nemours & Co., inc.), "Handling
     Chemical Information in the DuPont Central Report Index," Journal of
     Chemical Documentation, 1974.

 (4)  Sleng, A., £^£1. (Hoffman-LaRoche, Inc.), "HOffman-LaRoche1s
     On-Line/Batch Interactive Chemical Information System," Journal of
     Chemical Documentation, 1974.

 (5)  Gordon, Irving, et^ a_l. (Hooker Chemical Corp.), "Design and
     implementation of a Chemical Information Center," AS IS Journal, 1972.

 (6)  Edge, Eleanor B., e_t a^., (E.I. du Pont de Nemours & Co., Inc.), "System
     for indexing Research~Reports using a Punched Card Machine," American
     Documentation, 1957.

 (7)  Costello, J. c., Jr.  (E.I. du Pont de Nemours & Co., Inc.), "Storage and
     Retrieval of Chemical Research and Patent information by Linds and Roles
     in DuPont," American Documentation, 1961.

 (8)  Mendenhall, Donna M.  (uniroyal Chemical), "Cost Comparison of Four Data
     input Methods," Journal of Chemical Documentation, 1974.

 (9)  Brown, Horace D., e_t £l.   (Merck & Co., Inc.), "The Computer-Based
     Chemical Structure  information System of Merck Sharp and oohme Research
     Laboratories," Journal of Chemical information and Computer Sciences,
     1976.

(10)  Merritt, R. L., (Shell),   "The Shell Chemical File System," Journal of
     Chemical Documentation, 1974.

     Most of these articles were written before 1976, and none mention TSCA—
a fact which indicates that chemical information science would progress with
or without TSCA.  Firms must worry about information requirements of other
government agencies (OSHA, FDA, other offices of EPA).  Most importantly, they
must satisfy their own internal information requirements.  One expert thought
that the pharmaceutical and pesticide firms had the best information systems.
The stringent regulations concerning these industries have probably motivated
the industry to improve their   information systems, but only indirectly.  There
are no general clauses in FIFRA or the 1962 FDA amendments requiring firms to
submit "... any test data in their possession or control. . . ."64/
instead, the laws are set up to require specific tests.  Firms have set up
good information systems to aid them in doing research, not to pull reports
indiscriminately from their files.

     As time goes on and chemical information systems become more
sophisticated, the costs associated with any of the  "possession or control"
options will decrease.  The volume of information will significantly increase,
but information science should be able to provide systems capable of dealing
with this constant growth.
    li/TSCA Sec.  (5)(d)(1)(A).

-------
                                    TECHNICAL REPORT DATA
                             [Phase read f/izmic'iuus on ike reverse before completing/
 . REPORT NO.
   EPA-560/  12-80--005
,2.
                               3. RECIPIENT'S ACCESSIOf+NO.
4. TITLE AND SUBTITLE
Economic  Impact Analysis of  Proposed
Section 5 Notice Requirements
                                                            5. REPORT DATE
                                                              September 1980
                               6. PERFORMING ORGANIZATION CODE
7. AUTHORIS)
Robert Dresser,  James Edwards,  Joseph Kirk,
Stuart Fribush
                               8. PERFORMING ORGANIZATION REPORT NO.
 3. PERFORMING ORGANIZATION NAME AND ADDRESS

 ICF  Incorporated
 1850 K Street,  N.W., Suite 950
 Washington,  D.C.  20006
                               10. PROGRAM ELEMENT NO.
                                 2LS811
                               11. CONTRACT/GRANT NO.

                                 68-01-5878
 12. SPONSORING AGENCY NAME AND ADDRESS
 Office  of Pesticides and Toxic Substances
 U.S. Environmental Protection Agency
 401 M Street, S.W.
 Washington, D.C.  20460
                               13. TYPE OF REPORT AND PERIOD COVERED
                                 Proposed Report
                               14. SPONSORING AGENCY CODE
 •3. oUrPLt viENTARY NOT 53
 EPA Project Officer;  Sammy  K.  Nq
 16. ABSTRACT

    This  report presents the  analysis of the economic  impact of TSCA section 5 rules  on
 the chemical industry.  The  industry will be impacted when it introduces new chemicals.
 Of  the six distinguishable consequences for the chemical  industry, the most important
 are the  nonquantifiable uncertainty consequences.  The more unclear EPA's rationale  in
 making section 5 notice decisions,  the greater are the uncertainties.

    There will likely be a short-run drop in the number of new chemicals introduced  into
 commerce as chemical companies  shift their innovation activities into "safe" chemicals.
 Current  data do not allow a  quantitative estimate to  be made of the rate of chemical
 introductions, or the extent of the reduction caused  by the section 5 notice require-
 ments; and, even if the data were available, it is doubtful that accurate quantitative
 predictions could be made.

    Smaller  companies will face  greater uncertainties  and  the direct costs will more  often
 be  a factor in company decisions.  In the long run, this  regulation may cause the chemi-
 cal  industry to be composed  of  a fewer number of larger competitors better able  to  absorb
 the  direct  costs and regulatory uncertainty associated with the requirements.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                               b.IDENTIFIERS/OPEN ENDED TERMS
                                               COSATl Field, Croup
TSCA Section  5  Notice Requirements
Economic Impact Analysis
 3. DISTRIBUTION STATEMENT

 Release Unlimited
                 19. SECURITY CLASS (This Report)
                    Unclassified
                                                                           21. NO. OF PAGES
                                               20. SECURITY CLASS (This page)
                                                 Unclassified
                                             22. PRICE
EPA Form 2220-1 (9-73)
                                                               •U.S. GOVERNMENT PRINTING OFFICE:

-------