SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
      This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed Of approved
                       by the chartered SAB, and does not represent EPA policy.

 1                Valuing the Protection of Ecological Systems and Services:
 2                 An Expanded and Integrated Approach (tentative title)
 3
 4   1  INTRODUCTION AND BACKGROUND	2
 5      1.1.    Scope of this Report and its Intended Audience	2
 6      1.2.    The Importance of Valuing Ecosystems and their Services	3
 7   2.  ECOSYSTEM VALUATION AT EPA	8
 8      2.1.    Policy Contexts at EPA Where Ecosystem Valuation Can be Important	8
 9      2.2.    Institutional and Other Issues Affecting Benefits Assessment at EPA	10
10      2.3.    An Illustrative Example of Ecosystem Benefit Assessment at EPA	12
11   3.  AN INTEGRATED AND EXPANDED APPROACH TO ECOSYSTEM
12   VALUATION	18
13      3.1.    Context and Scope	19
14      3.2.    Ecological Services to be Included	19
15      3.3.    Measuring Benefits	21
16   4.  CONCLUSIONS AND RECOMMENDATIONS	24
17

-------
           SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft i a work in progress, does m;l reflect consensus advice or recommendations, has not been reviewed or approved
                          by the chartered SAB, and does not represent EPA policy.
 2
 3

 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
                        1.  INTRODUCTION AND BACKGROUND
1.1.   Scope of this Report and its Intended Audience

       The Science Advisory Board (SAB) Committee on Valuing the Protection of
Ecological Systems and Sen ices (C-VPESS) began its work in 2003 on a project
developed by the SAB to strengthen the Agency's analysis for protecting ecological
resources. The SAB saw a need to complement the Agency's ongoing work in ecological
science, ecological risk assessment, and ecological benefit assessment by offering advice
on how EPA might better value the protection of ecological systems and services and
how that information might better support decision making to protect ecological
resources. In this project the SAB set the goals of assessing Agency  needs and the state
of the art  and science of valuing protection of ecological systems and services and
identifying key areas for improving knowledge, methodologies, practice, and research at
EPA. Senior EPA managers supported the concept of this SAB project and participated
in the initial background workshop that launched the work of the C-VPESS. The
committee is an interdisciplinary group of experts from the following areas:  decision
science, ecology, economics, engineering, philosophy, psychology, and social sciences
with emphasis  in ecosystem protection.'  The committee sees its work as a three-year
initiative.

       This report is intended to provide an overview of the committee's conclusions to
date.2 It is aimed at providing initial advice for strengthening the Agency's approaches
for  valuing the protection of ecological systems and services, facilitating their use by
decision makers, and identifying the key research areas needed to strengthen the science
base. The committee will prepare additional reports with more detailed advice at the
       The SAB Staff Office published a Federal Register Notice on March 7, 2003 (68 FR 11082-11084)
      announcing the project and called for the public to nominate experts in the following areas: decision
      science; ecology; economics; engineering; psychology; and social sciences with emphasis in ecosystem
      protection. The SAB Staff Office published a memorandum on August 11,2003 documenting the steps
      involved in forming the new committee and finalizing its membership.

       The committee developed the conclusions in this report after multiple public meetings and workshops: a)
      an Initial Background Workshop on October 27, 2003 to learn the range of EPA's needs for science-based
      information on valuing the protection of ecological systems and services from managers of EPA
      Headquarters and Regional Offices; b) a Workshop on Different Approaches and Methods for Valuing the
      Protection of Ecological Systems and Services, held on April  13-14, 2004; c) an advisory meeting focused
      on support documents for national rulemakings held on June 14-15, 2004; d) an advisory meeting focused
      on regional science needs, in E?A's Region 9 (San Francisco) Office on Sept. 13,  14, and 15,2004; and e)
      advisory meetings held on January 26-26,2005 and April 12-13, 2005 to review EPA's draft Ecological
      Benefits Assessment Strategic Plan and to discuss economic and other methods for valuing the protection of
      ecological systems and service;;. The committee discussed a draft version of this report at a public meeting
      on (INSERT DATE).

-------
          SAB Draft Report Dated September 27,2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and does not represent EPA policy.

 1    compietion of the project.3 However, given the importance of the committee's charge, it
 2    felt that it would be useful to the Agency to issue an initial report that would indicate the
 3    direction that the committee's work is taking and serve as a prelude to the subsequent
 4    committee report(s). These subsequent reports will further develop the concepts in this
 5    initial advisory report and provide more detailed discussion of issues, methods, and
 6    application.  In particular, they will describe in more detail how different methods could
 7    be used more effectively to understand the benefits of the protection of ecological
 8    systems and services and how results of analyses could be better integrated and
 9    communicated to decision-makers.
10
11           This initial report focuses on the need for an expanded and integrated approach
12    for valuing EPA's efforts to protect ecological systems and services. It provides advice to
13    the Administrator, EPA managers, EPA scientists and analysts, and EPA staff across the
14    Agency concerned with ecological protection. It adopts a broad view of EPA's work,
15    which it understands to encompass national rulemaking, regional decision making, and
16    programs in general that protect ecological systems and services.  It focuses directly on
17    EPA's contributions and impacts, however, and not on the general question of the value
18    of ecosystems or ecological services in themselves.  It outlines a call for EPA to expand
19    and integrate its approach in important ways.
20
21           This report appears at a time when there is lively interest internationally,
22    nationally, and at EPA itself in the issue of valuing the protection of ecological systems
23    and services. Since the establishment of the SAB C-VPESS major reports have been
24    developed focusing on how to improve the characterization of ecological resources
25    (Millennium Ecosystem Assessment 2005; Millennium Ecosystem Assessment Board
26    2003; National Research Council 2004;  Pagiola, von Hitter, and Bishop 2004; Silva and
27    Pagiola 2003) The committee's work has benefited from  and will build upon those recent
28    efforts. The C-VPESS distinguishes its work from those efforts, however, in the
29    following ways.  The C-VPESS focuses on EPA as an audience.  The committee focuses
30    specifically on how EPA can value its own contributions to the protection of ecological
31    systems and services, so that the agency can make better decisions in its eco-protection
32    programs. The C-VPESS is inter-disciplinary and does not focus solely on economic
33    methods or values.  The committee will  offer advice on several benefits assessment
34    approaches and in each case will emphasize issues relevant to EPA policy and decision-
35    making and address how the Agency could better  characterize the benefits of ecological
36    protection.

37    1.2,    The Importance of Valuing Ecosystems  and their Services
38
39    1.2.1   The Concept of Ecosystem Services
40
       The Committee has already issued a related advisory report on the Agency's draft Ecological Benefits
            Assessment Strategic Plan (EPA SAB, EPA-SAB-ADV-05-OOX). This report complements the
            EBASP Advisory, and provides a discussion of an integrated framework alluded to in that report.

-------
  1
  2
  3
  4
  5
  6
  7
  8
  9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
    SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Da not Cite or Quote
 This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                   by the chartered SAB, and does not represent EPA policy.

       The term "ecosystem" describes the organisms in a given area interacting with
their physical environment as a functional unit. Ecosystems can describe organism-
physical environment interactions in a woodlot, a watershed, or an extensive landscape.
Ecosystems encompass all organisms within the prescribed area, including humans, who
are often the dominant element. Processes that link organisms, with their physical
environment are considered ecosystem processes and include primary productivity and
the cycling of nutrients and. water. These processes in total describe the functioning of
ecosystems. Processes that link organisms with each other, indirectly influencing flows of
energy, water and nutrients, can also be considered ecosystem processes, such as
pollination, predation and parasitism.

       "Ecosystem services" is an anthropocentric concept denoting the benefits that
humans derive from the functioning of ecosystems. An operational categorization of
ecosystem services has recently been proposed.by the Millennium Ecosystem
Assessment:

       a) Provisioning services (products obtained from ecosystems). These include
food, fuelwood, fiber, biochemicals, genetic resources and fresh water. Generally these
services are traded in the open marketplace.

       b) Regulating services (benefits received from regulation of ecosystem
processes). This category includes a host of benefits that humans derive from the
presence and functioning of ecosystems. These include flood protection, human disease
regulation, water purification, air quality maintenance, pollination, pest control and
climate control. These services are generally not marketed but many have clear value to
society and this value will increase for many  of these services as the many dimensions of
global  change proceed.

       c) Cultural services (the nonmaterial benefits people obtain from ecosystems).
Ecosystems provide cultural, spiritual and aesthetic values, and a sense of place.

       d) Supporting sen ices. These are the processes that maintain ecosystem
functioning such as: soil formation, primary productivity, biogeochemistry, and
provisioning of habitat. They all affect human well-being, but generally indirectly
through their support of the provisioning, regulating  and cultural service functions.

       Although there are different ways in which ecosystem services can be
categorized, the committee feels that the approach adopted in the Millennium Assessment
is a useful approach for conveying the concept of ecosystem services and the broad array
of functions and processes ecosystem services include. The ecosystem service concept is
useful  in many  ways.  First, it is a concept that is readily grasped by society, since it
relates directly  to human well-being. Secondly, it provides a tool for evaluating the
impacts of human actions in terms of the resulting change in the benefits provided by the
affected services. "Ecosystem health" can then be defined in terms of the output and
sustainability of services. When defined this way, the concept of ecosystem health
relates directly to the benefits provided to humans. However, life on earth can be revered

-------
          SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and does not represent EPA policy.

 1    and protected independent of human benefit. As discussed below, the committee
 2    recognizes that ecosystems can be valued not only because of the human-based services
 3    they provide but also for other non-anthropocentric reasons, including respect for nature
 4    based on ethical, religious,  or biocentric principles.
 5
 6    1.2.2  The Concept of Value

 7           Because people define and assign values, all values are anthropogenic. However,
 8    as noted above, not all values are anthropocentric. When people talk about environmental
 9    values, the values of nature, or the values of ecological systems and services, they may
10    have different things in mind.  People have moral, economic,  religious, aesthetic, and
11    other values, all of which can affect their thoughts, attitudes, and actions toward nature in
12    general or, more specifically, ecosystems and the  services they provide.
13
14           The most basic distinction in values is the distinction between means and ends.
15    To value something as a mean is to value it for its usefulness in helping to realize or bring
16    about some thing or state of affairs that is valued in its own right or as an end. Things
17    valued for their usefulness as means  in this sense  are said to have instrumental value.  Of
18    course, it would not make sense to value anything instrumentally or as a means unless
19    there was at least one thing or state that was valued for its own sake or as an end. Things
20    valued as ends are sometimes said to have intrinsic value.4 If intrinsic value applies to
21    things other than human beings or human experiences, then this conception of value is
22    non-anthropocentric.  Some people defend a non-anthropocentric conception of value or
23    goodness (Goodpaster 1978; Rolston III 1991; Taylor 1986).  However, others argue that
24    only human beings or human experiences have intrinsic value, thereby defending an
25    anthropocentric conception of value (Glover 1984; Sidgwick  1901; Williams 1994).
26
27           Ecological  systems  have instrumental value to the extent that they provide useful
28    sendees. Some people also claim, however, that an ecological system may have value
29    independently of the services it provides, i.e., its very existence has value. This claim can
30    mean several different things.  If it means that the existence of an ecological system is
31    valuable because people derive satisfaction from its existence, then it has what
32    economists call "existence value." This concept is anthropocentric. In addition, it is a
33    kind of instrumental value, since it is based on the premise that the existence of the
34    species or ecological system is one of many things that generate human satisfaction, and
35    that the various things that contribute to human satisfaction are potentially substitutable.
36    Some people, however,  claim that an ecological system may have intrinsic value of its
37    own. and that we should protect it for its own sake.  If the explanation of this claim refers
38    to reasons that are  independent of the contribution that the existence of an ecological
      "' There is controversy over the meaning of intrinsic value that we \vill not try to resolve here (Korsgaard
      1996). Many people take intrinsic value to mean that the value of something is inherent in that thing. Some
      philosophers have argued that value or goodness is a simple non-natural property of things (see Moore
      1903 for the classical statement of this position), and others have argued that value or goodness is not a
      simple property of things but one that supervenes on the natural properties to which we appeal to explain a
      thing's goodness (this view is defended by, among others, contemporary moral realists; see (Brink 1989;
      McDowell 1985; Sayre-McCord 1988; Sturgeon 1985).

-------
  1
  2
  3
  4
  5
  6
  7
  8
  9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
    SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
 This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                   by the c bartered SAB, and does not represent EPA policy.

system can make to human well-being, then this claim of intrinsic value should be
understood in a non-anthropocentric sense.

       This committee recognizes that there are many possible sources of value derived
from ecosystems and the services they provide. Thus, throughout this report, the term
"value"  is used broadly to include values predicated on their contributions to human
society (broadly defined), as well as those  based on an ethical, religious, or biocentric
notion of intrinsic value.

       Related to the cone apt of value are  the concepts of "benefits" and "valuation."
Both of these terms are relative to a specific change.  In this report, the change of interest
is the change in the state of an ecosystem or the flow of services it provides stemming
from an actual or proposed action by EPA. Thus, the term "ecosystem benefits" refers to
the increase in the value of the ecological system and/or its services.  This assumes a
positive change in value, Analogously, a reduction in value, for example from damages
to an ecosystem, can be viewed as a "negative benefit" or  cost.

       Similarly, the term "valuation" will refer to the process of characterizing or
measuring benefits or changes in value using various  methods and techniques.  For
example, economic valuation measures benefits in terms of the amount people are willing
to pay (WTP) to ensure an ecological improvement or the  amount people are willing to
accept (WTA) to forego the improvement.5 A social/psychological assessment method
might present the same ecological change and ask  people to rate the importance of
achieving (or preventing) that change relative to a  selection of changes in a number of
other (potentially competing) social goals.  An ecological  approach might assess the
value of the targeted change in terms of the magnitude of its effect on biodiversity or
some other indicator of ecological health based on the consensus that ecological health is
important to human/social well-being.  All of these assessments are based on an
anthropocentric view of values, where ecological values are assessed in terms of their
contribution to human well-being.  However, they  differ in terms of the means by which
values are expressed, and by the extent to which the value of the targeted ecological
change can then be explicitly compared (traded off) against other social values.
Economic assessments claim the broadest range and most explicit method for assessing
tradeoffs between, for example, ecological improvements  and changes in other goods or
services that also contribute to human well-being.  The social/psychological methods
generally settle for a relative measure of the value  of the targeted ecological change and
largely constrain tradeoff implications to options and  circumstances that are closely
related to the set of alternatives explicitly presented in the assessment. Ecological
assessments might restrict tradeoff implications to the biosphere.  In all cases, the
ultimate purpose of the valuation process is to characterize or measure the benefits (or
costs) associated with an ecological change in a way that provides useful information
about these benefits to policymakers and the public at large. The committee plans to
discuss these methods, what they may offer analysts and decision makers at EPA in
      5 A large literature exists on the use of economic valuation methods to estimate the value of changes in
      environmental quality. For a comprehensive description of these methods, see Freeman (1993).

-------
          SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and does not represent EPA policy.

 1    capturing different kinds of benefits, and their limitations and related issues in a future
 2    report.
 3
 4           1.2.3 The Importance of Assessing Ecosystem Benefits
 5
 6           Given the important role that ecosystems play in supporting life on earth and
 7    providing goods and services that people value, changes  in the state of these systems  or
 8    the flow of sendees they provide can have important implications. This importance has
 9    been increasingly  recognized by many, both within the U.S. and internationally. The
10    recent study by the National Research Council and the Millennium Ecosystem
11    Assessment are indicative of this growing recognition.
12
13           Many  EPA actions (e.g., regulations, rules, programs, policy decisions) affect the
14    slate of ecosystems and the flow of services derived from them.  EPA actions can either
15    lead to changes in the conditions of ecosystems (improvement or deterioration)  or prevent
16    changes that would otherwise have occurred. These impacts can occur both at a
17    relatively small, local scale as well as more broadly at a national scale. Yet, to date,
18    ecosystem impacts have received relatively limited consideration in EPA policy analyses.
19    Failure to consider these impacts as fully as possible can lead to distorted policy
20    decisions, particularly in regulatory contexts where benefits are being compared to costs.
21    In many cases, the result will be an under-valuation of (or failure to fully recognize) the
22    benefits of EPA actions aimed at protecting the environment. This can occur, for
23    example, when actions are evaluated based primarily on their impacts on human health,
24    without a recognition of potentially important ecosystem impacts.
25
26           Valuing the changes in ecological systems and services and assessing the
27    ecosystem benefits that result from EPA  policies or programs is  challenging for a number
28    of reasons.  Major challenges include:  a) understanding the many sources of value that
29    ecosystems generate, b) predicting the ecological impacts of alternative EPA actions, and
30    expressing those predictions in the temporal and spatial scale most appropriate for
31    decision-making, c) linking those impacts to changes in the dimensions of ecosystems or
32    the service flows that people value,  d) developing methods and techniques that can be
33    used to characterize and/or measure the value of protecting ecological systems and
34    services so that they may be incorporated or properly reflected in environmental
35    decisions and policies, e) aggregating to a national level using local or regional  studies
36    from regions with different ecological and/or economic characteristics, and f) finding
37    measures or means of representing ecological values or benefits  that are commensurable
38    with values of non-ecological changes caused by EPA actions, such as human health.
39    Despite these challenges, it is imperative that EPA improve its ability to assess ecosystem
40    benefits to ensure that ecological impacts are adequately considered in the evaluation of
41    EPA actions.

-------
          SAB Draft Report Dated September 27, 2005 tu Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does o( reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and does not represent EPA policy.
 1                            2.  ECOSYSTEM VALUATION AT EPA
 2
 3          There are several different contexts in which EPA policy decisions have
 4    ecological impacts and her.ce in which the need for ecosystem benefits assessment will
 5    arise.  In addition, when assessing benefits, EPA must operate within a set of
 6    institutional, legal, organizational and practical constraints that affect this process at the
 7    Agency. Thus. EPA has specific needs in this regard that must be recognized and
 8    addressed. These needs arise in different parts of the Agency for different purposes and
 9    for different audiences.  Seme of the needs present structured requirements for valuing
10    protection of ecological systems and services, while needs in other contexts are less
11    prescriptive.
            The most prescriptive requirements are for national rule making.  Benefit
     assessments are required for national rulemaking by two of EPA's governing statutes (the
     Toxic Substances Control Act and the Federal Insecticide, Fungicide and Rodenticide
     Act) and by Executive Order 12866 for "significant regulatory actions". The circular on
     "Regulatory Analysis" issued by the Office of Management and Budget (OMB) in
     September 2003, OMB Circular A-4, identified key elements of a regulatory analysis for
     such "economically significant rules." One of these elements is an evaluation of the
     benefits and costs of a proposed regulatory action and the main alternatives identified.
     The circular provided general guidance on how to provide monetized, quantitative, and
     qualitative information to fully characterize benefits and EPA itself has developed initial
     guidance for ecological benefit assessment (U.S. Environmental Protection Agency
     2000). In developing its draft Ecological Benefits Assessment Strategic Plan and in
     discussions with the committee (U.S. Environmental Protection Agency Science
     Advisory Board 2003), EPA identified the need for improved models and methods to
     help implement the requirements of the circular.  The Agency identified needs both to
     expand methods and data far economic valuation through benefit-cost or cost-
     effectiveness analysis and 1o explore other assessment methods to provide information on
     ecological effects that are currently  un-monetized and assigned an implicit value of $0.
     Managers seek approaches that are "sound, credible, and scientifically supportable" as
     well as flexible, affordable, and able to be implemented within the time constraints
     required by rulemaking (U.S. Environmental Protection Agency Science Advisory Board
     2004).

            EPA's  regional offices, although generally not responsible for national rule-
     making, are responsible for several kinds of decisions and activities where the benefits of
     ecological protection come into question:

                  Priority setting for regional action, such as targeting projects for wetland
                   restoration and enhancement or identifying critical ecosystems or
                   ecological resources  for regional attention

-------
          SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed Or approved
                        by the chartered SAB, and does not represent EPA policy.

 1                 Setting Supplemental Environmental Protection (SEPs) penalties for
 2                 enforcement cases where those penalties involve protection of ecological
 3                 systems and services
 4                 Choice of options for Superfund and Resource Conservation and Recovery
 5                 Act (RCRA) cleanups that could take ecological benefits into account
 6                 Review of Environmental Impact Statements prepared by other federal
 7                 agencies to comply with the National Environmental Protection Act
 8                 Assisting state and local governments and other federal Agencies with
 9                 protecting lands and land uses, where assessment of the value of
10                 protection options could help decision-makers make  better-informed
11                 decisions.
12
13           Regions seek low-cost methods that can be implemented quickly to inform
14    "place-based" decisions. They seek methods that provide information on the value of
15    ecological services; ecological diversity; conservation opportunities and threats;
16    sustainability; and historical and cultural values associated with ecological systems or
17    parts of ecosystems at the watershed or landscape scale.  Regions experience the need to
18    communicate the value of ecological protection as they collaborate with other federal
19    agencies and with government partners at the local, state, and regional levels.
20
21           EPA's need to communicate the value of its ecological protection programs has
22    two dimensions: 1) a retrospective dimension, because assessments  focus on the value of
23    EPA's current and past protection efforts and 2) a prospective dimension, because such
24    assessments are meant to inform decisions about future EPA programs and priorities.
25
26           The need to assess the ecological benefits of policy options is woven into most of
27    the Agency's decisions, including the assessment of ecological protection programs .
28    Program assessments are mandated for EPA, as they are for all agencies of the executive
29    branch, by the Government Performance and Results Act of 1993. As part of that
30    assessment, OMB requires EPA to periodically identify its strategic goals and describe
31    both the social costs and budget costs associated with them. EPA's Strategic Plan for
32    2003-2008 described the current social costs and benefits of EPA's programs and policies
33    under each strategic goal area for the year 2002 (U.S. Environmental Protection Agency
34    2003).  This analysis repeatedly points out that EPA lacks data and methods to quantify1
35    the ecological benefits associated with the goals in its strategic plan.
36
37           In addition, the Government Performance Results Act of 1993 established
38    requirements for assessing the effectiveness of federal programs.  Part of that assessment
39    involves assessing the outcomes of programs intended to protect ecological resources.
40    EPA must report annually on its progress in meeting program objectives linked to
41    strategic plan goals and must engage periodically in an in-depth review [through the
42    Program Assessment Rating Tool (PART)] of selected programs to identify their net
43    benefits and to evaluate their effectiveness in meeting meaningful, ambitious program
44    outcomes. Characterizing ecological benefits associated with EPA programs is a
45    necessary part of the program assessment process.

-------
          SAB Draft Report Dated S.-ptember 27,2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, tines not reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and docs not represent EPA policy.

  1    2.2.    Institutional and Other Issues Affecting Benefits Assessment at EPA
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
        The committee recognizes that ecological benefits assessment at EPA must be
conducted within a set of institutional, legal, organizational, and practical constraints that
affect what is and can be done to incorporate ecosystem values into policy evaluations.
In an effort to better understand these issues and their implications for the committee's
charge, the committee conducted a series of interviews with Agency staff.6  The
interviews were focused on the process of developing benefit analyses for Regulatory
Impact Assessment (RIA) for rulemaking and the relationship between EPA and the
Office of Management and Budget. However, many of the questions raised are  equally
applicable to strategic planning, performance reviews, regional analysis, and other
situations in which the agency is called upon to assess the value of ecosystems. Below are
some key observations made by the committee based on those interviews.

       EPA Program Offices responsible for new rules initiate, finance, and administer
the process for developing ecological benefit assessments. The development of a new
rule - including definition of the rule itself, options to be weighed, and the assessment of
impacts arising from the rule - involves much more than  scientific assessment.  Political
negotiations and legal analysis arguably dominate the process.  EPA has a formal rule-
development process with several stages, each which impose demands on the Agency and
the Agency also develops rules to meet court-imposed deadlines.

       Several aspects of these imposed constraints deserve emphasis.  First, despite the
commonality of the underlying rule-development process, it is clear that there is no single
way in which ecological valuation is conducted within the Agency. Practices vary
considerably across program offices, reflecting differences in mission, in-house  expertise,
etc.  Program offices have different statutory and strategic missions. The organization,
financing, and skills of the program offices differ enormously.  The National Center for
Environmental  Economics (NCEE) is the Agency's centralized reviewer of economic
analysis within the agency.7 However, the primary expertise and development of the
rules resides within the program offices.

       Secondly, the timing of the process largely determines the kinds of analytical
techniques that are employsd.  This is related to court-imposed deadlines on the rule
process, as well as intervening requirements related to the collection and analysis of new
data.  The scientific community is used to much longer time horizons for their analyses.
They are also used to the idea that a new rule should call  for the collection of new kinds
      6 These interviews were conducted by one Committee member, Dr. James Boyd, in conjunction with the
      Designated Federal Officer, Dr. Angela Nugent, over the period September 22, 2004 through November 23,
      2005. In seven sets of interviews, Dr. Boyd spoke with staff from the Office of Policy, Economics and
      Innovation, Office of Water, Office of Air and Radiation, Office of Solid Waste and Emergency Response.

       NCEE is typically brought in ly the program offices to both help design and review RIAs. NCEE can be
      thought to provide a centralized '"screening" function for rules and analysi s before they go to OMB. NCEE
      is actively involved in discussions with OMB as rules and supporting analysis are developed and advanced.
                                               10

-------
          SAB Draft Report Dated September 27,2005 to Assist Meeting Deliberations - Do not Cite nr Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and does not represent EPA policy.

 1    of data. Unfortunately, collecting new data poses a significant bureaucratic problem for
 2    the Agency. To collect original data, the Agency must submit an Information Collection
 3    Request, which is reviewed within the Agency and by OMB. This hurdle alone can add
 4    significant drag to the assessment process.  With perhaps a year or two at most to conduct
 5    a study, this kind of review significantly limits the kind of analysis the Agency can
 6    conduct.
 7
 8           A third issue is the role of the Office of Management and Budget (OMB) in
 9    defining or directing ecosystem valuation exercises  at EPA. It was difficult for the
10    committee to ascertain the EPA-OMB relationship precisely.8  EPA has been given
11    explicit guidance by OMB in the  Circular A-4, which the committee views as a
12    reasonable document on its own because of its call for a full characterization  of the
13    impacts of different policy options and inclusion of language calling for characteri/ation
14    of benefits that cannot be monetized or cannot be quantified (Office of Management and
15    Budget 2003)9, However, the implications of some sections of the Circular, particularly
16    relating to the treatment of benefits that cannot be readily  monetized, remain  somewhat
17    ambiguous.  For a benefit or cost  that cannot be expressed in monetary terms, the
18    Circular instructs Agency staff to "try to measure it in terms of its physical units," or, if
19    this is not possible either, to "describe the benefit or cost qualitatively." However, little
20    guidance is provided on how this  should be done. Instead, the Circular urges regulators
21    to "exercise professional judgment in identifying the importance of non-quantified factors
22    and assess as best you can how they might change the ranking of alternatives based on
23    estimated net benefits."
24
25           It is clear that the Agency views the OMB as a kind of "court" that reviews its
26    analysis. In front of this  "court,"  methods that have been accepted in the past create
27    incentive for the use of the same or similar methods in the future. The thinking seems to
28    be "if it made it through OMB once, it will make it through again." There appears to be a
29    pronounced tendency to use "off-the-shelf methods to avoid problems with OMB. This
30    creates a bias toward the status quo and a reluctance to explore new or innovative
31    approaches. To this end, the committee sees the need to strike an appropriate balance
32    between the use of established methods and the possible need  to innovate in an effort to
33    conduct more comprehensive and defensible benefit assessments for use in decision
34    making and evaluation.
35
      8 OMB responded lo written questions, but declined to be interviewed by Dr. Boyd. EPA staff were
      informed that their formal responses to all questions, including the OMB-EPA interview were to be
      documented as part of the Committee report and this is likely to have had a chilling effect on the
      discussions.

      * eg., see pp.27 "If monctization is impossible, explain why and present all available quantitative
      information" and. pp "If you are not able to quantify the effects, you should present any relevant
      quantitative information along with a description of the unquantified effects, such as ecological gains,
      improvements in quality of life, and aesthetic beauty."
                                               11

-------
  1
  2
  3
  4
  5
  6
  7
  8
  9
10
11
12
13
14
15
16
17
18
19
20
     SAB Draft Report Dated September 27,2005 to Assist Meeting Deliberations - D> not Cite or Quote
  This draft is a work in progress, docs not reflect consensus advice or recommendations, has not been reviewed or approved
                    by the Chartered SAB, and does not represent EPA polity.

        A related issue involves RIA review by external parties. The Agency does not
take a standardized approach to RIA review.1  EPA staff and managers reported that peer
review was focussed only on "novel" elements of an analysis  This raises the question of
how the Agency (and perhaps OMB) defines "novel" Moreover, the novelty standard
actually creates a clear incentive to avoid conducting novel analyses (however defined).
It is clearly cheaper and quicker to avoid review altogether.  The committee advises the
Agency to consider whether there is a role for a standing expert body  that can bring
consistency to the review of analysis, avoid duplication of review, and be sensitive to
timing and resource constraints.

        Finally, the committee notes the importance of organi^ation of assessment science
within the Agency.  Currently, the Agency relies upon a variety of offices to develop
assessments, with varying degrees of reliance on other offices  (e.g.. NCEE) or outside
assistance.11 It is not clear which work better than others.  In addition, it is not clear how
different programs integrate social science and biophysical science.12

Do we want to advocate ai "ecosystem services valuation paradigm" and/or
development of a set of guidelines for doing ecosystem valuation? Relationship to
risk assessment paradigm/guidelines??? See  more on this in footnote below.
21
22
23
24
25
26
27
23.   An Illustrative Example of Ecosystem Benefit Assessment at EPA

       In an effort to better understand the current state of ecosystem valuation at EPA,
the committee examined ir. detail one specific case where benefit assessment was
undertaken, namely, the Environmental and Economic Benefits Analysis that EPA
prepared in support of nevs regulations for Concentrated Animal Feeding Operations
(CAFOs) (U.S. Environmental Protection Agency 2002).13'14 The Agency indicated that
        In some cases, review panels ,ire appointed, in others not. In some cases, contractors are called upon to
      manage the review.  In other cases, Program Offices themselves manage the review process.

      1' Another issue that relates to the organization of science within the Agency is the availability and location
      of data to support ecosystem valuation.  The choice of methods is clearly related to the practical availability
      of data across the Agency. It is important that data that are housed within individual program offices are
      made public and readily shared with other offices.

      12 One anecdote is that Dr Boyd was able to speak with only one ecolegist during the interviews designed
      (in part) to interview a set of ecologists. Economists in the agency were not able to identity ecologists to
      interview, for example. It also became clear that simple "counts" of professional background can be
      deceptive. What the agency terms an "ecologist" is not necessarily what the scientific community would
      call an ecologist.

        The Committee reviewed and critically evaluated the Environmental and Economic Benefits Analysis at
      its June 15, 2004 meeting.  As stated in the Background Document for SAB Committee on Valuing the
      Protection of Ecological Systems and Services for its Session on June 15. 2004, the purpose of this exercise
      was "to provide a vehicle to help the Committee identify approaches, methods, and data for characterizing
      the full suite of ecological 'values' affected by key types of Agency actions and appropriate assumptions
      regarding those approaches, meihods, and data for these types of decisions." The Committee based its
                                                12

-------
           SAB Draft Report Dated September 27, 2005 to .Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                           by the chartered SAB, and does not represent EPA policy.

 1    this analysis was typical of other EPA regulatory analyses of ecological benefits in form
 2    and general content.
 3
 4            Because the proposed new CAFO rule constituted a "significant regulatory
 5    action" under Executive Order 12866, EPA was required to assess the costs and  benefits
 6    of the rule.15  EPA identified a wide variety of potential "use" and "non-use" benefits as
 7    part of its analysis.16 Using various economic  valuation methods, EPA provided
 8    monetary  quantifications in its CAFO report for seven environmental benefits.17
 9    Approximately eighty-five  percent of the monetary benefits quantified by EPA were
10    attributed to recreational use and non-use of affected waterways.  According to Agency
11    staff, EPA's analysis was driven by what it could monetize.  EPA focused on those
12    benefits for which data were known as available for quantification of both the baseline
13    condition  and the likely changes from the proposed rule, and translation of those changes
14    into monetary equivalents.  EPA's  final benefits assessment provides only a brief
15    discussion of the benefits that it could not monetize.  The benefits table in the Executive
      review on EPA's final benefits report (EPA 2002) and a briefing provided by the EPA Office of Water
      staff.  During the June meeting, members of the Committee divided into two workgroups.  The workgroups
      each worked independently and reported their findings to the combined Committee.  The leaders of the two
      working groups then prepared a consolidated summary of comments from the two workgroups.

      14 In December 2000, EPA proposed a new CAJ''O rule under the federal Clean Water Act to replace 25-
      year-old technology requirements and permit regulations (66FR 2959). EPA published its final rule in
      December 2003 (68 FR 7176). The new CAFO regulations, which cover over 15,000 large CAFO
      operations, reduce manure and wastewater pollutants from feedlots and land applications of manure and
      remove exemptions for stormwater-only discharges.

        Prior to publishing the draft CAKO rule in December 2000, EPA spent two years preparing an initial
      assessment of the costs and benefits of the major options. After releasing the draft rule, EPA spent another
      year collecting data, taking public comments, and preparing assessments of new options.  EPA published
      its final assessment in 2003. An intra-agency team at EPA, including economists and environmental
      scientists in the Office of Water, Office of Air and Radiation, Office of Policy Economics and Innovation,
      and Office of Research and Development, worked on the benefit assessment. EPA also worked with the
      U.S. Department of Agriculture in developing the assessment.  Dr. Christopher Miller of EPA's Office of
      Water estimated  that EPA spent approximately $1 million in overall contract support to develop the benefit
      assessment. EPA spent approximately $250.000-5300,000 on water quality modeling as part of the
      assessment.

      10 The potential "use" benefits included in-stream uses (commercial fisheries, navigation, recreation,
      subsistence, and  human health risk), near-stream uses (non-contact recreation, such as camping, and
      nonconsumptive, such as wildlife viewing), off-stream consumptive uses (drinking water,
      agricultural/irrigation uses, and industrial/commercial uses), aesthetic value (for people residing, working,
      or traveling near water), and the option value of future services. The potential "non-use" values included
      ecological values (reduced mortality/morbidity of certain species, improved reproductive success, increased
      diversity, and improved habitat/sustainability), bequest values, and existence values.

      17 These benefits were recreational use and non-use of affected waterways,  protection of drinking water
      wells, protection of animal water supplies, avoidance of public water treatment, improved shellfish harvest,
      improved recreational fishing in estuaries, and reduced fish kills.
                                                  13

-------
  1
  2
  3
  4
  5
  6
  7
  8
  9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
    SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
 This draft is a work in progress, does r ot reflect consensus advice or recommendations, has not been reviewed Of approved
                    by the chartered SAB, and does not represent EPA policy.

Summary listed a variety of non-monetized benefits18 but designated them only as "not
monetized." EPA represented the aggregate effect of these "substantial additional
environmental benefits" simply by attaching a "+B" place-holder to the estimated range
of total monetized benefits. Although the Executive Summary gave a brief description of
these "non-monetized" benefits, the remainder of the report devotes little attention to
them.

       Although much effort  was invested in the CAFO benefits assessment, the
assessment illustrates a number of limitations in the current state of ecosystem valuation
at EPA.

       First, EPA's analysis and report focused nearly exclusi\ely on meeting the
requirements as described in Executive Order 12866.  This may not be surprising since
the Executive Order provided the reason for preparing the analysis and report.  However,
when EPA prepares a benefit  assessment specifically to comply with Executive Order
12866, the Agency need not limit itself to the goals and requirements of the Executive
Order.  The Executive Order does not preclude EPA from adopting broader goals. The
Executive Order provides merely that EPA shall conduct an "analysis" and "assessment"
of the "benefits anticipated from the regulatory action" and, "to the extent feasible, a
quantification of those benefits." By adopting a narrow focus, the report failed to
consider or reflect the broader purposes that a benefit assessment can serve.
Environmental benefit assessments, such as the CAFO study, can serve a variety of
important purposes, including helping to educate policy-makers and the public more
generally about the benefits that stem from EPA regulations.

       Second, as noted above, in implementing the Executive Order, the CAFO analysis
did not provide the full characterization of ecological benefits using quantitative and
qualitative information, as required by the OMB Circular A-4  Instead, the report
focused on a limited set of environmental benefits,  driven primarily by the ability to
monetize these benefits using generally accepted models and existing  value measures
(benefit transfer).19 These benefits did not include all of the major environmental
benefits that the new CAFO rule would likely generated, nor all of the benefits that
generated public support for the new rule.20 The Circular requires that a benefit
      18 These include eutrophication of estuaries; reduced pathogen contamination of drinking water supplies;
      reduced human and ecological risks from hormones, antibiotics, metals, and salts, improved soil properties
      from reduced over-application of manure; and "other benefits".

      19 EPA apparently conducted no :iew economic valuation studies (although a limited amount of new
      ecological research was conducted) and did not consider the possible benefits of developing new
      information where important benefits could not be valued in monetary- terms based on existing data. The
      CAFO report emphasizes EPA's predisposition toward conservative benefits estimates and identifies the
      lack of adequate data and/or models meeting EPA standards of quality as a basis for truncating the CAFO
      analysis.

      20 For example, while the report notes the potential effects of discharging hormones and other
      Pharmaceuticals commonly used in CAFOs into drinking water sources and aquatic ecosystems, the nature
      and possible ecological significance of these effects is not adequately developed or presented. Similarly,
                                                14

-------
           SAB Draft Report Dated September 27,2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                           by the chartered SAB, and does not represent EPA policy.

 1    assessment identify and characterize all the important benefits of the proposed rule, not
 2    simply those that can be monetized.  By focusing only on a narrow set of benefits, the
 3    CAFO analysis and report understates the benefits of the rule change and distorts  the
 4    rationale supporting the final rule.21  An unfortunate effect of this presentation is to
 5    suggest to readers that the monetized benefits constitute the principal justification for the
 6    CAFO rule.22 Although in this case the focus on monetized benefits did not affect the
 7    outcome of the  regulatory review, it is  certainly possible that in a different context, this
 8    conservative approach to benefits assessment (based only on easily monetized benefits)
 9    could inadvertently undermine support for a rule that would be justified based on  a more
10    inclusive characterization of benefits.
11
12           Third, the monetary values  for many of the emphasized benefits were  estimated
13    through highly leveraged benefit transfers that were generally  based on dated studies
14    conducted in contexts quite different from the CAFO rule application.23  This was
15    undoubtedly driven to a large extent by time, data, and resource constraints, which make
16    it very difficult for the Agency to conduct new surveys or studies and virtually force the
17    Agency to monetize benefits using existing value estimates. However, reliance on dated
18    studies in quite different contexts raises questions about the credibility or validity of the
19    monetary benefit estimates. This is particularly true when values are presented as point
20    estimates, without adequate recognition of the underlying limitations, due to uncertainty
21    and data quality.
22
23           Fourth, EPA apparently did not engage in a detailed and systematic effort  at the
24    outset to model the rule's ecological  impacts. The report presents only a simple
25    conceptual model that traces outputs (a list of pollutants in manure - Exhibit 2-2 in the
26    CAFO report) through pathways (Exhibit 2-1) to environmental and human health
      the report does not adequately address the well-known consequences of discharging TMC precursors into
      drinking-water sources.

      n One of the benefits of monetary benefit estimates obviously is the ease of aggregating them by simple
      arithmetic. However, the Committee does not believe that reporting that a rule produced a total of "218.9
      million dollars in annual benefits" is necessarily more useful, meaningful, or defensible for environmental
      policy than reporting, for example, the achievement of a "10% reduction in the pollution of over 129,000
      miles of streams and rivers, 3.2 million acres of lakes and ponds, and 2,800 square miles of estuaries."

      22 In the case of this CAFO rule, 97% of the monetized benefits arise from recreation (boating, swimming
      and fishing) and from private well owners' willingness to pay for water quality, estimated using contingent
      valuation or travel cost methods.

       "  EPA used estimates based on a variety of public surveys in its benefit transfer efforts, including: a
      national survey (1983) that determined individuals' willingness to pay for changes in surface water quality
      relating to water-based recreational activities (Section 4 of the CAFO Report); a series of surveys (1992,
      1995, 1997) of willingness to pay for reduced/avoided nitrate (or unspecified) contamination of drinking
      water supplies (Section 7); and several studies (1988. 1995) of recreational fishers' values (travel cost,
      random utility model) for improved/protected fishing success related to nitrate pollution levels in a North
      Carolina estuary (Section 9).
                                                 15

-------
           SAB Draft Report Dated September 27,2005 to Assist Meeting Deliberations - Do nut Cite or Quote
       This draft is a work in progress, does rot reflect consensus advice or recommendations, has not been reviewed or approved
                          by the chartered SAB, and does not represent EPA policy.

  1    effects.24 This model provided useful guidance, bul was not sufficiently detailed to assure
  2    an adequately comprehensive and balanced analysis of the rule's ecological impacts.  As
  3    a consequence the analysis was unduly directed by Agency presumptions (or discoveries)
  4    about the availability of relevant data and the likely opportunities to quantify effects
  5    precisely and to link and monetize associated benefits. This was undoubtedly driven in
  6    part by the time pressures of putting together the regulatory impact analysis. However,
  7    without a detailed and comprehensive modeling effort at the outset, EPA had insufficient
  8    insight into the potential benefits that needed to be analyzed and valued. Developing
  9    integrated models of relevant ecosystems at the outset of a valuation project would also
10    help in identifying important secondary  effects, which frequently may be of even greater
11    consequence or value than the primary effects.25
12
13            Fifth, the CAFO analysis clearly demonstrates the challenges of conducting
14    ecological benefit assessments at the national level.26  National rule-makings inevitably
15    require EPA to generalize away from geographic specifics, both in  terms of ecological
16    impacts and associated values. However, it is possible (and desirable) to make use of
17    intensive case studies (e.g.. individual watersheds, lakes, streams, estuaries) in support of
18    the national-scale analyses  Existing and ongoing research at local  and regional scales
19    offers more detailed data and models that could be better exploited, both to fill in gaps
20    and to systematically validate the national-scale analyses. Systematically performing and
21    documenting comparisons to intensive study sites could indicate the extent to which the
22    national model needs to  be adjusted for local/regional conditions and could provide data
23    for estimating the range of error and uncertainty in the projected national-scale effects.
24
25            Sixth, although EPA invited public  comment on the draft CAFO analysis as
26    required by Executive Order 12866, there is no indication in the draft CAFO report that
27    EPA consulted with the public during its analysis to help it identify, assess, and prioritize
28    the effects and values addressed in its analysis, nor is  there discussion in the final CAFO
29    analysis of any comments received on the draft CAFO analysis. Early public
30    involvement could play a valuable role in helping the Agency both a) identify all of the
31    systems and services impacted by the proposed regulations and b) determine the
      24 Although EPA later prepared more detailed conceptual models of the CAFO rule's impact on various
      ecological systems and services. EPA did not prepare these models until after the Agency finished its
      analysis.

      23 Contamination of estuaries, for example, might negatively affect fisheries in the estuary (a primary
      effect) but might have an even greater impact on offshore fisheries that have their nurseries in the estuary (a
      secondary effect).

      26 The goal of EPA's analysis was a national level assessment of the effects of the CAFO rule. This
      involved the effects of approximately 15,000 individual facilities, each contributing pollutants across local
      watersheds into local and regioral aquatic ecosystems. A few intensive cas-e studies were mentioned in the
      report and used to calibrate the national scale models (e.g., NWPCAM, GLEAMS), but there was no
      indication that these more intensive data sets were strategically selected or used systematically for formal
      sensitivitv tests or validations of the national-scale model results.
                                                16

-------
          SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and does not represent EPA policy.

 1    regulatory effects that are likely to be of greatest value. This would ensure that the
 2    benefits assessment includes the most important impacts.
 3
 4           Finally, while EPA in its analysis and report appropriately emphasized the
 5    importance of using outside peer-reviewed data, methods, and models, EPA did not seek
 6    to peer review its application of them or its integration of these components in deriving
 7    benefit values for the CAFO rule. Once again, this is undoubtedly due in part to time and
 8    resource constraints.  However, peer review, especially early in the process, would help
 9    EPA staff identify relevant and available data, models, and methods to support its
10    analysis, and provide encouragement, direction, and sanction for more vigorous and
11    effective pursuit of ecological and human wellbeing effects associated with the proposed
12    rule. The general idea is to have individual components of the analysis (e.g., watershed
13    modeling, air dispersal, human health, recreation, aesthetics) each reviewed, as well as a
14    more general review of the overall analytic scheme.
15
                                               17

-------
  1
  2
  3
  4
  5
  6
  7
  8
  9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
          SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and does not represent EPA policy.
        3.  AN INTEGRATED AND EXPANDED APPROACH TO ECOSYSTEM
                                       VALUATION

       The CAFO example discussed above highlights a number of limitations to the
current state of ecosystem valuation at EPA. The committee's analysis points to the need
for a comprehensive, integrated approach to valuing the ecological impacts of EPA
actions, one that focuses on the impacts of most concern to people and integrates
ecological analysis with valuation.  This section describes a proposed framework, based
on the committee's deliberations to date. A more detailed discussion of the methods that
could be used to implemem this framework and the issues that arise in doing so will be
provided in a subsequent committee report. The goal in this report is simply to provide
an organizing framework to guide the more detailed discussion regarding
implementation.

       A key feature of the framework outlined here is that it integrates ecological
analysis with valuation.  Tiis integration needs to occur both at an early stage (in the
identification of the impacts that matter) and at a later stage (when estimating the value of
impacts). Thus, instead of having ecologists work independently initially to estimate
ecological impacts and then "pass the baton" on to economists or other social scientists to
value those impacts, it envisions collaborative work across disciplines to ensure that the
analysis focuses on the impacts that are of greatest concern and that the ways in which
these impacts are defined and measured are informative during the selection and, if
necessary, design of the valuation techniques/methods.  Such a framework requires a
committed dialog among the relevant bio-physical, ecological, and social/economic
scientists and analysts. The various disciplines must reach out to establish useful and
credible links to each other This interaction should commence at the beginning of the
process and continue until the completion of the analysis. Ecological models need to be
developed, modified, or extended to provide usable inputs for value assessments.
Likewise, valuation methods and models need to be developed, modified, or extended to
address important ecological/bio-physical effects that are currently underrepresented in
value assessments.

       In addition, the framework envisions the use of a variety of methods to
characterize and measure benefits or values, including economic methods,
social/psychological assessments, and ecological approaches.  The suite of methods to be
used will vary  with the specific policy or valuation context, due to differences  across
contexts in: a) information needs, b) the underlying sources of value being captured; c)
data availability; and d) methodological limitations. The framework should serve as a
guide to EPA staff as they conduct RIAs and seek to implement the provisions of
Circular A-4 (including the provisions relating to benefits that are not readily quantified
or monetized), as well as in regional decision-making and program assessment.

       The proposed framework has three main components:  a) identify the context and
scope of the benefit assessment, b) identify the ecological services that will be considered
in the assessment, and c) characterize, represent or measure those impacts in bio-
                                              18

-------
          SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not ate or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and does not represent EPA policy.

 1    physical, human, and/or monetary terms. This proposed framework would parallel the
 2    Agency's Framework for Ecological Risk Assessment (U.S. Environmental Protection
 3    Agency Risk Assessment Forum 1992) and ultimately be merged with it as part of a
 4    broader framework for ecological assessment.

 5    3.1.    Context and Scope
 6
 7           As noted above, ecological benefit assessment can play a key role in a number of
 8    different decision contexts, including national rule-making, local/regional decision-
 9    making, and program evaluation. There is  a need to formulate the benefit assessment
10    problem within the specific EPA context.  These contexts differ not only in the required
11    scale for the analysis (e.g., national vs. local) but possibly also in the type of valuation
12    information that is needed, i.e., whether it requires that benefits be characterized or
13    measured in terms of bio-physical impacts, or the resulting impacts on humans, or both.
14
15           The information needed for a given policy decision will  in turn depend on the
16    decision approach to be used in evaluating  alternatives. The rule to be used could be
17    dictated by statute, regulation, or executive order, or could be determined by the EPA
18    staff.  Possible approaches include maximization of (expected) net present value (based
19    on cost-benefit analysis), minimization of the (expected) cost of meeting a given goal
20    (cost-effectiveness), use of a safe minimum standard, use of the precautionary principle,
21    or use of a moral or rights-based rule  based on intrinsic value. For example, the
22    Endangered Species Act is based on an underlying presumption that species should be
23    preserved (either because of high existence value or high intrinsic value), and hence the
24    value information necessary to support decisions in this context can be expressed solely
25    in bio-physical terms.  In contrast, if a strict cost-benefit rule is to be used in a rule-
26    making context, aggregate dollar values of benefits (and costs) are needed. Under a
27    broader interpretation (e.g.. OMB Circular A-4), use of cost-benefit analysis would
28    require that ecological benefits be a) measured in dollar terms when possible, b)
29    measured using other metrics for impacts on humans (e.g., population affected) when
30    monetary valuation is not possible, and c) fully described in qualitative terms, when
31    quantitative information is not available.

32    3.2.    Ecological Services to be Included
33
34           Decisions  about the ecological services to be included in the analysis should be
35    based on an assessment of the impacts that are likely to be most important, depending on
36    both the magnitude and bio-physical importance of the effect and the resulting impact on
37    humans.
38
39           3.2.1 Identifying Potentially Important Bio-Physical  Impacts. The bio-
40    physical impacts of a given EPA action can be identified at different levels.  These
41    include the individual level, the population level, the community level, the ecosystem
42    level (union of biological populations with  their surrounding physical environment), and
43    the level of the global biosphere.  Ecological  science is organized according to these
44    scales.  For the purposes of ecological benefits assessment, ecological impacts
                                              19

-------
  1
  2
  3
  4
  5
  6
  7
  8
  9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
    SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
 This draft is a work in progress, does n >t reflect consensus advice or recommendations, has not been reviewed or approved
                   by the chartered SAB, and does not represent EPA policy.

correspond to changes in functions or services provided by the ecosystem, as described
above.  Living organisms supply goods and services that differ across all levels of
organization, from the individual to the ecosystem or global biosphere. For example, the
service provided by an individual animal unit is different from the service provided by a
given animal population.

       Many types of ecological models exist at various levels (e.g., population,
community, ecosystem, biosphere) to predict impacts of perturbations on ecosystems.
Some have been developed for specific contexts (species, geographic locations) while
others are more general.  In some cases "off-the-shelf models may be available, while in
others existing models may need to be modified or new models developed,  (what should
we say about the current state of ecological modeling??? We need the committee to
discuss and come to some agreement about this.)

       In identifying possible impacts, it is important to consider their full range,
including both primary and secondary effects, adequately accounting for uncertainty
(incomplete information), (instability of the system (including the effect of random
shocks or management errors and the system's resilience), heterogeneity within a
population or ecosystem, heterogeneity across populations or ecosystems, and dynamic
changes in the ecosystem over time. Ecosystems are complex, highly variable systems
with many interacting parts  They are subject to both natural and anthropogenic
disturbances that can propagate through the system in ways that are difficult to predict.
The complexity1, variability, and potential instability of the systems need to be considered
when identifying impacts with the greatest ecological significance.

       3.2.2 Identifying What Matters to People. For benefit assessments based on
anthropocentric values, it is important to identify early in the process what people care
about, i.e., which ecological services or functions are important to them.  For example,
are individuals likely to value the re-introduction of native grasses into a marshland, or
would they be just as happy with non-native grasses that perform similar ecological
functions and aesthetic appeal?  Is animal waste disposal a concern to people primarily
because of the recreational opportunities lost due to the resulting deterioration in water
quality, or are they primarily concerned about other impacts?  The range of services that
are the focus of the benefits assessment needs to include the services people care most
about. Previous benefit assessments have often focused on what can be measured
relatively easily rather than what is most important to people.  This diminishes the
relevance, usefulness  and impact of the assessment.

       Information about what matters to people can be obtained in a variety of ways.
Examples include survey information (from past surveys or surveys conducted
specifically  for the benefit assessment) or the results of previous valuation studies.  In
addition, early public  involvement27 or use of focus groups or workshops comprised of
      27 This could include either a robust public involvement process following AdministrativeProcedures Act
      requirements (e.g., FR publication), or some other public involvement process [see EPA's public
      involvement policy, (U.S. Environmental Protection Agency Office of Policy 2003) and the SAB report on
                                              20

-------
          SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, docs not reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and does not represent EPA policy.

 1    representative individuals from the affected population and relevant scientific experts can
 2    help to identify relevant or potentially important ecological services for the specific
 3    context of interest. (Add something about group processes here?  )
 4
 5           In eliciting information about what matters to people, it is important to bear in
 6    mind that what people say they care about depends on both their preferences and their
 7    information, i.e., the extent to which they  are informed about an ecological system and
 8    the services it provides. Survey respondents or even members of a focus group may have
 9    preferences that are representative of the general population but may not be fully
10    informed. Expressions of what is important (e.g.. in surveys) can change with the amount
11    of information provided. Collaborative interaction between analysts and public
12    representatives can ensure that respondents have sufficient information when expressing
13    preferences. (Add something about constructed preferences here?)
14
15           The information about those ecosystem functions and services that are important
16    to people and potentially impacted significantly should then be integrated to select the
17    services to include in the assessment,  As  noted above, this requires a collaborative effort
18    and dialogue among analysts from a variety of disciplines early  on in the valuation
19    process.

20    3.3.    Measuring Benefits
21
22           Given the services to be included in the assessment, the impact of the EPA action
23    on those services needs to characterized and,  when possible, measured or quantified. To
24    measure impact on humans, the bio-physical measures of ecological impact need to be
25    translated into their effects on the goods and services provided by those ecosystems to
26    humans.  These impacts can be measured  in non-monetary terms (e.g., population
27    affected, duration of effect, etc.) or, in  contexts where benefits are to be compared
28    directly to costs, in monetary terms if possible.
29
30           Estimating bio-physical impacts requires information about the ecological
31    production function for the services being considered. This allows an estimation of the
32    change in the level of services that  could result from a given EPA action or policy.
33    (e.g., percent reductions/avoidances of pollution in streams and lakes, reduced/avoided
34    eutrophication of estuaries, reduced risk from the introduction of hormones and
35    antibiotics into aquatic systems, improved/protected quality of community drinking water
36    sources). As when selecting the services to be considered , in estimating the effect of a
37    given action on those services, it is crucial to account for the complexity of ecosystems.
38    In particular, predicted changes need to account for the interconnectedness of
39    ecosystems, uncertainty about how the systems operate, possible instability of the system
40    (including the effect of random shocks or  management errors and the system's
41    resilience), heterogeneity within a population or ecosystem, heterogeneity across


      science and stakeholder involvement (U.S. Environmental Protection Agency Science Advisory Board
      2001)J.
                                              21

-------
  1
  2
  3
  4
  5
  6
  7
  8
  9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
    SAB Draft Report Dated September 27, 20(15 to Assist Meeting Deliberations - Do not Cite or Quote
 This draft is a work in progress, docs n:>l reflect consensus advice or recommendations, has not been reviewed or approved
                   by the chartered SAB, and does not represent EPA policy.

populations or ecosystems, and dynamic changes in the ecosystem over time. This
complexity and the associated uncertainty underscore the importance of presenting ranges
rather than point estimates of values when possible.

       In some contexts (e.g., endangered species) where bio-physical impacts are the
primary concern, the benefit assessment can end with quantification of the impact of the
EPA action on these bio-physical indicators. However, when EPA policies are to be
evaluated in terms  of impact on humans, the bio-physical effects must be translated into
the corresponding impacts on the flow of goods and services that humans value.  First
and foremost, this requires that the output from the ecological impact assessment be in a
form that can be used as an input in estimating the value of the change in ecosystem
services. Again, this requires that ecologists work closely with other disciplines to ensure
that the ecological  assessment is designed from the start with this requirement in mind.

       To translate bio-physical impacts into human benefits, it is necessary to project
how ecosystem changes will affect humans through changes in the flow of the  goods and
services they provide. The extent of the impact on humans can be measured in non-
monetary terms using a variety of metrics, such as the number and characteristics of the
people/communities affected, the number significantly affected, the likely symptoms
avoided or reduced, and the duration of the impact.

       Estimation  of impact on humans in terms of the extent of exposure or similar
measures is crucial in three possible ways.  First, in some contexts, decisions based on
moral or religious principles (e.g., protection of children's health) may look directly to
these measures as indicators of the appropriate policy choice.  Second, even in contexts
where monetary measures of value are sought, the human benefits captured by
information on exposure or symptoms need to be translated into their monetary
equivalents. This requires an understanding of those impacts on humans before this
translation can occur. Third, in some cases where monetary values are sought, it may not
be possible to monetize all benefits  due to data or methodological constraints. In these
cases, there may be a tendency simply to "ignore" the benefits that cannot be monetized.
Using methods that defensibly report the magnitude and human significance of such
effects,  rather than ignoring them, would allow the policymakers to draw their own
conclusions regarding the associated potential value or benefit. Thus, in all of these
cases, estimates of the impact of the ecosystem change on human populations are needed.

       In contexts where monetary  metrics are sought and the necessary data and
methods exist, the impact of the ecosystem change on the provision of services to human
populations can be translated into a monetary equivalent of that change using standard
economic valuation techniques to determine the tradeoffs that people are willing to make.
Economic or monetary methods for valuing changes are relatively well-developed. They
are designed to estimate the benefit or cost of a given change in ecological services using
a willingness-to-pay or will ingness-to-accept measure of the utility equivalent  of that
change. These methods have been applied to the valuation of ecosystem services in a
number of studies that have produced results that are useful for policy evaluation.
However, as in the CAFO .study, monetary valuation methods have generally been
                                             22

-------
          SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                         by the chartered SAB, and does not represent EPA policy.

 1    applied to a relatively narrow set of services. In some cases, these might not have been
 2    the services that people are most concerned about protecting.  There is a need to expand
 3    the range of services to which economic valuation is applied.
 4
 5           As with ecological impacts, in estimating the values of impact on humans  in
 6    either monetary or non-monetary terms, it is necessary to address cross-cutting issues
 7    such as uncertainty (randomness, level of information), dynamics, scale (temporal,
 8    geographic), and heterogeneity (spatial variability, heterogeneity across people). In
 9    subsequent reports, the committee will assess the challenges of uncertainty arising out of
10    data limitations, theory limitations, and randomness, and will  recommend  approaches for
11    reducing uncertainty and conveying the magnitude and nature of uncertainty to
12    policymakers.
13
                                               23

-------
          SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do not Cite or Quote
      This draft is a work in progress, docs nol reflect consensus advice or recommendations, has not been reviewed Of approved
                         by the chartered SAB, and does not represent EPA policy.
1
2
3
4
                    4.  CONCLUSIONS AND RECOMMENDATIONS
[to be added]
                                                24

-------
          SAB Draft Report Dated September 27,2005 to Assist Meeting Deliberations - Do not Cite or Quote
       This draft is a work in progress, does not reflect consensus advice or recommendations, has not been reviewed or approved
                        by the chartered SAB, and does not represent EPA policy.

 1
 2                                    REFERENCES
 3
 4
 5
 6
 7    Brink. David O. 1989. Moral Realism and the Foundation of Ethics. Cambridge:
 8           Cambridge University Press.
 9    Federal Insecticide, Fungicide and Rodenticide Act (FIFRA). 1972.
10    Freeman,  A. My rick III, 2003, The Measurement of Environmental and Resource Values:
11           Theory and Methods, 2nd edition, Washington, DC., Resources for the Future.
12    Glover, Jonathan. 1984. What Sort of People Should There Be? Penguin,
13    Goodpaster, Kenneth. 1978. On Being Morally Considerable. Journal of Philosophy
14           75:308-25.
15    Jamieson, Dale. 2002. Morality's Progress. Oxford: Oxford University Press.
16    Korsgaard, Christine. 1996. Two Distinctions in Goodness. In Creating the Kingdom of
17           Ends, edited by C. Korsgaard. Cambridge: Cambridge University Press.
18    McDowell, John. 1985. Values and Secondary Qualities. In Morality and Objectivity,
19           edited by T. Honderich: Routledge and Kegan Paul.
20    Millennium Ecosystem Assessment. 2005. Ecosystems and Human  Well-being:
21           Synthesis. Washington, D.C.: Island Press.
22    Millennium Ecosystem Assessment Board. 2003. Ecosystems and Human Well-being; A
23           Report of the Conceptual Framework Working Group of the Millennium
24           Ecosystem Assessment. Washington, DC: Island Press.
25    National Research Council. 2004. Valuing Ecosystem  Services; Toward Better
26           Environmental Decision-Making. Washington, D.C.: The National Academies
27           Press.
28    Office of  Management and Budget. 2003. Regulatory Analysis. Circular A-4.
29    Pagiola, Stefano,  Konrad von Ritter, and Joshua Bishop. 2004. Assessing the Economic
30           Value of Ecosystem Conservation. In World Bank Environment Department
31           Papers.
32    Rolston HI, Holmes.  1991. Environmental Ethics: Values in and Duties to the Natural
33           World. In The Broken Circle: Ecology, Economics, and Ethics, edited by F. H. B.
34           a.  S. Kellert. New Haven: Yale University Press.
35    Sayre-McCord, Geoffrey. 1988. The Many Moral Realisms. In Essays on Moral Realism,
36           edited by G.  Sayre-McCord. Ithaca:  Cornell University Press.
37    Scanlon, T.M. 1998.  What We Owe to Each Other. Cambridge: Harvard University Press.
38    Sidgwick, Henry. 1901. Methods of Ethics. Vol. 1. New York: Macmillan.
39    Silva, Patricia, and Stefano  Pagiola. 2003. A Review  of the Valuation of Environmental
40           Costs and Benefits in World Bank Projects. In World Bank Environmental
41           Economics Series. The World Bank.
42    Sturgeon, Nicholas.  1985. Moral Explanations. In Morality, Reason, and Truth,  edited by
43           D. C. a. D. Zimmerman: Rowman and Allenheld.
44    Taylor, Paul. 1986. Respect for Nature: A Theory of Environmental Ethics. Princeton:
45           Princeton  University Press.
                                            25

-------
          SAB Draft Report Dated September 27, 2005 to Assist Meeting Deliberations - Do nut Cite or Quote
       This draft is a work in progress, does nut reflect consensus advice or recommendations, has not been reviewed or appro%'ed
                        by the chartered SAB, and does not represent EPA policy.

 1    U.S. Environmental Protedion Agency. 2004 Ecological Benefits Assessment Strategic
 2          Plan November 3: 2004 SAB REVIEW DRAFT
 3    U.S. Environmental Protection Agency. 2000. Guidelines for Preparing Economic
 4          Analyses: EPA240-R-00-003.
 5    U.S. Environmental Protection Agency. 2003. Appendix 1: Social Costs and Benefits. In
 6          Direction for the Future:  2003-2008 EPA Strategic Plan: Appendix 1: Social
 7          Costs and Benefits.
 8    U.S. Environmental Protection Agency Office of Policy, Economics and Innovation.
 9          2003. Public Involvement Policy of the U.S. Environmental Protection Agency.
10          EPA 233-B-03-002.
11    U.S. Environmental Protection Agency Risk Assessment Forum. 1992. Framework for
12          Ecological Risk Assessment. EPA/630/R-92/001.
13    U.S. Environmental Protection Agency Science Advisory Board. 2001. Improved
14          Science-Based Environmental Stakeholder Processes: An EPA Science Advisory
15          Board Commentary, EP A-S AB-EC-COM-001-006.
]6	. 2003. Summary Minutes of the U.S. Environmental Protection Agency (EPA)
17          Science Advisory Board (SAB) Committee on Valuing the Protection of
18          Ecological Systems and Services Initial Background Workshop October 27, 2003,
19          J. W. Marriott Hotel, Washington, DC.
20    	. 2004. Summary Minutes of the U.S. Environmental Protection Agency (EPA)
21          Science Advisor}' Board (SAB) Committee on Valuing the Protection of
22          Ecological Systems and Services Workshop on Different Approaches and
23          Methods for Valuing the Protection of Ecological Systems and Services April 13-
24          14, 2004, Four Points Sheraton, 1201 K Street, NW Washington, DC, 20005.
25    Williams, Bernard. 1994. Must a Concern for the Environment Be Centred on Human
26          Beings? In Reflecting on Nature, edited by L. Gruen and D. Jamieson. Oxford:
27          Oxford University Press.
28
                                            26

-------