e^  •  -   r f    -  R-





-------

-------
                                                   EPA 230-6^95-005
             A FRAMEWORK FOR MEASURING THE ECONOMIC
                       BENEFITS OF GROUND WATER
                               OCTOBER 1995
This report is the product of the EPA Interoffice Ground Water Valuation Workgroup that
included two economists from outside the Agency. Workgroup members are identified on the
following page.

-------

-------
 EPA Interoffice Ground Water Valuation Workgroup Members






 Kevin J. Boyle, Department of Resosurce Economics and Policy, University of Maine




 John C. Bergstrom, Department of Agricultural and Applied Economics, University of Georgia




 Charles Job, Office of Ground Water and Drinking Water (Workgroup Chair)




 Mary Jo Kealy, Office of Policy, Planning and Evaluation




 Ron Bergman, Office of Ground Water and Drinking Water




 Rodges Ankrah, Office of Policy Analysis




 Ghulum A15, Office of Pesticide Programs




 Jihad Alsadek, Office of Pesticide Programs




 Gary Ballard, Office of Solid Waste




 Vivian Daub, Office of Water




Jacolyn Dziuban, Office of Radiation Programs




Richard Howes,  Office of Emergency and Remedial Response




 Susan Schulze, Water Management Division, Region II

-------

-------
                                Table of Contents
Introduction	  1

Framework for Valuing Ground Water	  4

      Defining Ground Water Values 	6
      Ground Water Functions	9
      Ground Water Services 	19
      Effects of Changes in Ground Water Services	20
      Measuring Economic Values  	21
      Aggregation Issues	24
      Uncertainty in Ground Water Valuation	25
      Intergenerational Issues	26

Previous Ground Water Valuation Studies  	29

Ground Water Valuation and Regulatory Impact Analyses  	39

      Draft Class V Injection Well Regulatory Impact Analysis  	39
      Draft RIA for Final  Rulemaking on Corrective Action for Solid Waste
          Management Units 	41
      Summing Up	43

A Structure for Considering the Value of Ground Water	45

      Protocol Components	46

Concluding Comments	55

References Cited	57

-------

-------
                                List of Tables

Table 1. FUNCTION: STORAGE OF WATER RESERVE (STOCK)	11

Table 2. FUNCTION: DISCHARGE TO STREAMS/LAKES/WETLANDS	14

Table 3. Ground Water Conditions in Study Areas 	31

Table 4. Information Presented on Ground Water Commodity
        (Change in Services)	35

Table 5. Changes in Ground Water Services - Stock Function	50

Table 6. Changes in Ground Water Services - Discharge Function	51

Table 7a. Available Data for Valuing Changes in Ground Water Services - Stock
         Function	53

Table 7b. Needed Data for Valuing Changes in Ground Water Services - Stock
         Function	54

Table 8. Other Valuation Considerations for Changes in Ground Water
         Services - Stock Function  	55


                                List of Figures

Figure 1	7

-------

-------
                                I. INTRODUCTION

       The primary goal of this project is to develop a framework for valuing ground

water that is applicable to all offices within U.S. EPA (EPA hereafter) that consider the

value of ground water resources when conducting Regulatory Impact Analyses (RIAs).1

The precedent for this effort was set with the development of "A Guide for Cost-

Effectiveness and Cost-Benefit Analysis of State and Local Ground Water Protection

Programs" by EPA's Office of Water (United States Environmental Protection Agency,

1993). The guide provides a concise discussion of the processes for accomplishing these

types of analyses, but does not provide specificity regarding the estimation of ground

water values.  It is the intent of this report  to begin to develop the framework for a

comparable guide for valuing ground water. We use the term value in a generic sense

such that  the values associated with reductions in ground water quantity or quality may

be considered losses and, conversely, increases are deemed benefits.

       The objectives of this research were to:

1.     Develop a conceptual framework that describes the steps in identifying and

       measuring the economic value of ground water.

2.     Identify responsibilities  within each office of EPA and link these responsibilities to

       the valuation framework;

3.     Consider the extent to which the benefits of ground water protection, as suggested

       by  the valuation framework, have been accounted for in previous RIAs; and
    Although we limit our application of the valuation framework in this report to RIA's, the framework
can also be applied to other ground water policies and programs.

-------
 4.     Provide guidelines for utilizing the valuation framework to consistently value



       ground water across EPA offices and policy issues within offices.




 The next section of this report describes the conceptual framework for identifying and




 measuring the economic value of ground water.  The valuation framework links changes




 in physical characteristics of ground water to uses (services) provided by ground water



 and the economic effects of changes in ground water services. It is the link between




 changes in the physical condition of ground water  and the resulting changes in (effects




 on) services that provides the changes in values that  are necessary for conducting RIAs.




 The basic valuation framework reported here was  developed in conjunction with the




 Ground Water Valuation Group in which we participated, chaired by Charles Job of the



 Office of Ground Water and Drinking Water.



       After meeting with representatives of the various offices within EPA, it became



 clear that each office did not have a clear mandate, at  least as perceived by the




 representatives to whom we talked, regarding the treatment of specific types of ground




 water values in RIAs conducted by their offices. Rather, they viewed their mandates as



 requiring them to address all potential resource effects when conducting RIAs.  If ground




 water was an affected  resource, then any changes in  the condition of ground water would




 be  taken  into  consideration.  Given this outcome, it was not possible for us to develop



Jinks  between responsibilities of various offices within EPA and the conceptual




 framework for valuing ground water.  In hindsight  this outcome is not entirely




 unexpected.  Each  office addresses policy issues relevant to their mission, but issues




 within any office can have dimensions that affect the condition of ground water

-------
resources.  For example, issues relating to pesticides and resulting decisions by the Office




of Pesticides can have implications for ground water quality in areas where pesticides are




produced or in areas where the pesticides are applied. Similarly, actions by the Office of




Solid Waste can have implications for ground water. Thus, there is not necessarily a




direct link between specific attributes of ground water valuation and the missions of the




various offices within EPA.  Having a consistent blue print for ground water valuation,




however, can help each office's assessments of ground water values, and can help to




insure  consistency in ground water value assessments within and  across offices.




       Before moving to a discussion of how the ground water valuation framework




applies to RIAs, we discuss studies that have  investigated the value of ground water.  We




do this because RIAs are often dependent on data available in the literature and




discussing existing studies helps to amplify issues raised in regard to the conceptual




foundation for valuing ground water. The  discussion of existing ground water valuation




studies is presented in Section III and we  focus on the commodity definitions as this is




the issue unique to ground water applications.




       It was decided as part of the development of this report that the link to RIAs




would  focus on two recent RIAs. This was done because the field  of environmental




valuation is evolving rapidly and RIAs conducted  five  to ten years ago had limited access




to much of the ground water valuation data currently  available.  It  was also believed by




the contract administrators that the selected RIAs present the most comprehensive




evaluations of ground water conducted by EPA for RIAs.  The RIAs examined are the




"Class  V Injection Well Regulatory  Impact Analysis and Regulatory Flexibility Analysis"

-------
                                                                                  4
by the Office of Water (U.S. EPA, 1993) and "Draft Regulatory Impact Analysis for the
Final Rulemaking on Corrective Action for Solid Waste Management Units" by the
Office of Solid Waste (Cadmus Group, 1993).
       In the final section, Section V, we suggest guidelines for EPA to use as a starting
point for developing a common approach for valuing ground water across offices.  These
guidelines should be equally appropriate for the design of original studies as well as
selecting available studies for transferring estimates to new applications in current  RIAs.

                II.  FRAMEWORK FOR VALUING GROUND WATER
       Any research project or empirical analysis begins with the investigator making a
number of decisions that define the conceptual and empirical domain of the investigation.
These decisions are the  direct consequence of explicit and implicit questions posed by the
investigator(s) and to a large extent determine the outcome of the investigation. In an
original study these decisions are important in terms of deciding what attributes of a
ground water resource are to be valued and ultimately defines the uses to which the
resulting value estimates can be applied  in the policy process. In the case of benefits
transfer, where values estimates are taken from existing studies, analysts' decisions are
crucial in terms of defining the valuation issue at the "policy site" where value estimates
are to be transferred and interpreting value estimates from available "study sites" where
empirical estimates  of value already exist.
       A fundamental issue is the definition of the change in the condition of a resource
and the ensuing changes in services generated by the resource,  i.e., commodity definition.

-------
                                                                                   5
This begins with an understanding of whether the change has occurred or is proposed.
Given ex post or ex ante standing, the next step is to develop a technical definition of the
reference condition of the resource and identify whether the increment of change is an
enhancement or diminishment of the quantity and quality of the resource.  For either
enhancement of,  or preventing harm to, the expected condition of the resource must be
defined.  Differences between the reference condition and expected condition define the
change in the quantity and quality of a resource to be evaluated.  Consideration should
also be given as to whether the mechanism(s) employed to accomplish the change can
achieve the proposed resource condition with certainty.  It is also necessary to know the
geographical extent of the changes to address the issue of whose values should count in
the computation of aggregate benefits or costs.  This information collectively  constitutes
the formal commodity definition for a resource being valued. These questions must be
asked for original investigations of value as well for transfers of value estimates to
unstudied sites.
      After the commodity definition is established, it is necessary to map changes in the
resource  condition into changes in the provision of services from which humans derive
value.  Accomplishing this step can be difficult regardless of whether an original
investigation or transfer exercise is being performed. Benefit transfer practitioners have
an added  complication in that they must interpret value estimates at study sites and
assess their transferabiJity, conceptually and  statistically,  to the policy site. In turn, the
increment of change being evaluated at the policy site must be carefully defined, not only

-------
                                                                                   6
 for relevance to the current policy issue, but also to accomplish the transfer exercise
 itself.

 Defining Ground Water Values
       Valuing ground water requires a clear definition of the ground water "commodity"
 to be valued. Figure 1  summarizes the technical data required to define a ground water
 commodity.  The first step is monitoring (Box 1) to assess the current or baseline aquifer
 condition in quantity and quality dimensions (Box 2).  The  next step is to assess how the
 current quantity and quality of ground water will change  "with" and "without" the
 proposed regulation (Boxes 3 and 4).  These factors include extraction rates, natural
 recharge and discharge, natural contamination (e.g., salt infiltration) and human-induced
 contamination (e.g., pesticide  contamination, industrial chemical contamination), and
 public policies regarding the use and protection  of ground water. The results of the
 assessments provide estimates of the reference (without policy) water quantity (X°) and
 quality (Q°),  and the subsequent (with policy)  water quantity (X1) and quality (Q1)
 (Boxes 5 and 6).  Given estimates of the reference and subsequent ground water
 conditions, we define the change in water quantity and quality (X° - X1, Q° - Ql) (Box 7).
The steps and linkages illustrated by Boxes 1-7 primarily  involve the work of hydrologists,
geologists, engineers, ecologists, soil scientists,  and other physical  and biological scientists.
Investigations of ground water conditions by these specialists must be sufficient to identify
changes in ground water services linked to the prescribed policy in a manner that

-------
                              FIGURE 1
                         AQUIFER MONITORING
                                  i
                    CURRENT QUANTITY AND QUALITY

                      /                  N
    ASSESSMENT OF FACTORS
   AFFECTING QUANTITY AND
   QUALITY "WITHOUT POIJCY*
    ASSESSMENT OF FACTORS
    AFFECTING QUANTITY AND
     OUA1JTY "WITH POLICY"
                                                       I
REFERENCE WATER QUANTITY (X°)
       AND QUALITY (Q°)
SUBSEQUENT WATER QUANTITY (X1)
       AND QUALITY (Q1)
                CHANGE IN WATER QUANTITY AND QUALITY
                            (X* - X1, Q° - Q1)
                                  I
                                  8
                   CHANGE IN GROUNDWATER SERVICES
                    S° = f(X°, Q°|Sg) TO Sl = f(XS
                                  i
                      ECONOMIC VALUE (BENEFITS)
                       V=g(AS|S£), WIIKRE AS=S°-S1

-------
                                                                                   8

facilitates the estimation of economic values. Formally modeling the steps illustrated by

Boxes 1-7 represents one of the greatest challenges that needs to be addressed to

estimate economic values of ground water protection.

       Reference services (S°) supported by ground water are determined by the without

policy ground water quantity  (X°) and quality (Q°) and subsequent services (S1)  are

determined by the with policy ground water quantity (X1) and quality (Q1).  Reference

and subsequent ground water services are conditional upon given levels of substitute and

complementary service flows  (S°) (Box 8).  The interactions of scientists and policy

analysts facilitate the mapping of changes in the condition of ground water to changes in

service flows  which affect economic activities.  We can then estimate economic value

(e.g., willingness-to-pay) as  a  function of the change in the ground water service  flows,

given the specified reference  and subsequent ground water conditions, and service flows

from substitutes and complements to  the ground water resource (Box 9).

       The steps and linkages illustrated by Boxes 8 and 9 involve the work of

economists, building on the biophysical analyses developed for Boxes  1-7.2 It is difficult

to overemphasize this important point.  When it comes to estimating economic values

associated with natural resource service flows, the most complex and limiting step is often

establishing clear linkages between changes in the biophysical condition of a natural

resource and changes in natural resource policies or programs.  Economic valuation of

ground water therefore requires that progress be made on two fronts: establishing formal

linkages between ground water protection policies and changes  in the biophysical
   2We use the term "biophysical" to indicate biological, ecological, hydrologic, chemical, and other physical
factors.

-------
                                                                                  9
condition of ground water (Boxes 1-7), and developing these linkages in a manner that
allows for the estimation of policy-relevant economic values (Boxes 8-9).  Ideally, steps 1
through 9 involve interactions and cooperation between economists and other scientists
to ensure a smooth and productive flow of data and models to develop policy-relevant
ground water value estimates.

Ground Water Functions
      The linkages between biophysical changes in ground water quantity or quality (Box
7), changes in ground water services (Box 8) and changes in economic values (Box 9) can
be better understood by considering aquifer functions.  The  biophysical dimensions of
ground water quantity and quality determine two broad functions of any aquifer.  The.
first function is storage of a water reserve or stock (Table 1). Ground water stored  in an
aquifer provides a reserve of water with given quantity and quality dimensions.  The
quantity dimension includes the amount of ground water available within a specific
geographic region  in a given time period, and the change  in this  quantity over time from
recharge and extraction.  Rates of natural recharge,  natural  discharge, and human-
induced extraction must be considered. Quality includes both natural and human
induced contaminants that may affect the services to which ground water can be applied
in a given time period, and the change in quality over time due to natural filtration and
the leaching of contaminants.   The rates of human-induced contamination and natural
sources of contamination must also be considered.

-------
                                                                                                               10
Table 1. FUNCTION: STORAGE OF WATER RESERVE (STOCK). Ground water stored In an aqulf.r provides • reserve
(stock) of water which can be directly used to generate services. Potential service (lows and effects of these services are listed
below.*
           SERVICES
            EFFECTS
              VALUATION
             TECHNIQUES
      Provision of Drinking Water
Change in Welfare from Increase or
Decrease in Availability of Drinking
Water
                                    Change in Human Health or Health
                                    Risks
Market Price/Demand Function
Supply or Cost Function
Producer/Consumer Cost Savings
Contingent Valuation
Hedonic Price/Property Value
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer
      Provision of Water for Crop
      Irrigation
Change in Value of Crops or
Production Costs
                                    Change in Human Health or Health
                                    Risks
Market Price/Demand Function
Supply or Cost Function ',<
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Property Value
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer
      Provision of Water for
      Livestock
Change in Value of Livestock
Products or Production Costs
                                    Change in Human Health or Health
                                    Risks
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Property Value
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer

-------
11
Table 1. Continued
4












5




6




7




8




9




10

















Provision of Water for Food
Product Processing











Provision of Water for Other
Manufacturing Processes



Provision of Healed Water
for Geoihermal Power Plants



Provision of Cooling Water
for Oiher Power Plants



Provision Water/Soil Support
System for Preventing Land
Subsidence


Provision of Erosion and
Flood Control through
Absorption of
Surface Water Run-Off

Provision of Medium for
Wastes and Other By-
Products of Human
Economic Activity














Change in Value of Food Products
or Production Costs




Change in Human Health or Health
Risks





Change in Value of Manufactured
Goods or Production Costs



Change in Cost of Electricity
Generation



Change in Cosi of Electricily
Generation



Change in Cost of Maintaining
Public or Private Property



Change in Cost of Maintaining
Public or Private Property



Change in Human Heallh or Health
Risks Attributable to Change in
Ground water Quality




Change in Animal Health or Health
Risks Attributable to Change in
Ground water Quality



Change in Economic Outpul
Attributable to Use of Ground water
Resource ax "Sink" for Wastes


Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Property Value
Benefits Transfer
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer
Market Price/Demand Function
Supply or Cos! Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Averting Behavior
Benefits Transfer
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer

-------
                                                                                                                    12
Table 1. Continued
11

















12


Provision of Clean Waier
through Support of Living
Organisms















Provision of Passive or Non-
Use Services (e.g.. Existence
or Bequest Motivations)
Change in Human Health or Health
Risks Attributable to Change in
Water Quality




Change in Animal Health or Health
Risks Attributable to Change in
Water Quality



Change in Value of Economic
Output or Productions Costs
Attributable lo Change in
Water Quality

Change in Personal Utility


Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Averting Behavior
Benefits Transfer
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer
Contingent Valuation
Benefits Transfer

'This (able was developed with the input and assistance of the U.S. EPA Ground Water Valuation Group directed by Charles Job and
Mary Jo Kealy.

-------
                                                                                  13
       The second function is discharge to surface water (streams, lakes, and wetlands)
(Table 2).  In the Eastern U.S., for example, the base flow of many streams and rivers is
supported by ground water discharge.  Through discharge  to surface water, ground water
indirectly contributes to the services generated by surface waters and wetland ecosystems.
Once again there are quantity and quality dimensions in terms of rates of discharge to
surface waters and the quality of the discharge supply. It should also be noted that
surface water may recharge ground water.  In this  case, a portion of the services provided
under the water reserve or stock function should be attributed to surface water.  To
simplify exposition we focus on the flow of water from ground  water to surface water.
Similar logic can be applied to develop values for the effects of surface water flows to
ground water.
       The'share of surface water services that can be  legitimately credited to ground
water is very difficult to quantify. The primary challenge is to  model the physical
interactions between ground water and surface water services such that the incremental
(marginal)  contributions of ground water discharge to surface water can be identified and
measured.  This task is necessary to avoid double-counting of service flows and, in  turn,
economic values (e.g., attributing the same service  and associated value to both ground
water and surface water). For example, assume an aquifer provides a major source of
recharge water for a stream which is heavily used for recreational fishing.  Assume also
that normal land run-off also contributes substantially to the flow of the stream.  Suppose
two water quality protection policies are implemented during the same time period. One

-------
                                                                                                                14
Table 2.  FUNCTION: DISCHARGE TO STREAMS/LAKES/WETLANDS. Ground wafer contributes lo the Row or stock of
water in streams, lakes, and wetlands.  A portion of surface water and wetlands services are therefore attributable to the
ground water resource.  Potential service (lows and effects of these services are listed below.*
             SERVICES
                                                  EFFECTS
                                                VALUATION
                                                TECHNIQUES
        Provision of. Drinking Water
        through Surface Water Supplies
Change in Welfare from Increase or
Decrease in the Availability of
Drinking Water (Access Value)
                                       Change in Human Health or
                                       Health Risks
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Property Value
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer
        Provision of Water for Crop
        Irrigation through Surface
        Water Supplies
Change in Value of Crops or
Production Costs
                                       Change in Human Health or Health
                                       Risks
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Property Value
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer
        Provision of Water for
        Livestock through Surface
        Water Supplies
Change in Value of Livestock
Products or Production Costs
                                       Change in Human Health or Health
                                       Risks
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Property Value
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer
        Provision of Water for Food
        Product Processing through
        Surface Water Supplies
Change in Value of Food Products
or Production Costs
                                       Change in Human Health or Health
                                       Risks
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Property Value
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer

-------
                                                                                                                   15
Table 2. Continued
        Provision of Walcr for Other
        Manufacturing Processes
        through Surface Water Supplies
Change in Value of Manufactured
Goods or Production Costs
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer
        Provision of Cooling Water for
        Power Plants through Surface
        Water Supplies
Change in Cost of Electricity
Generation
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent  Valuation
Benefits Transfer
        Provision of Erosion, Flood,
        and Storm Protection
Change in Cost of Maintaining
Public or Private Property
                                        Change in Human Health or Health
                                        Risks through Personal Injury
                                        Protection
                                        Change in Economic Output
                                        Attributable to Use of Surface
                                        Water Supplies for Disposing
                                        Wastes
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer
        Transport and Treatment of
        Wastes and Other By-Producis
        of Human  Economic Activity
        through Surface Water Supplies
Change in Human Health or Health
Risks Attributable to Change in
Surface Water Quality
                                        Change in Animal Health or Health
                                        Risks Attributable to Change in
                                        Surface Water Quality
                                        Change in Economic Output
                                        Attributable to Use of Surface
                                        Water Supplies for Disposing
                                        Wastes
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Averting Behavior
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer

-------
                                                                                                                   16
Table 2. Continued
        Support of Recreational
        Swimming, Boating, Fishing,
        Hunting, Trapping and Plant
        Gathering
                                Change in Quantity or Quality
                                Recreational Activities
                                       Change in Human Health or
                                       Health Risks
                                   Market Price/Demand Function
                                   Supply or Cost Function
                                   Consumer/Producer Cost Savings
                                   Contingent Valuation
                                   Travel Cost Method
                                   Benefits Transfer

                                   Market Price/Demand Function
                                   Supply or Cost Function
                                   Consumer/Producer Cost Savings
                                   Contingent Valuation
                                   Hedonic  Price/Wage
                                   Averting  Behavior
                                   Benefits Transfer
 10
Support of Commercial
Fishing,
Hunting,
Trapping, Plant Gathering
Change in Value of Commercial
Harvest or Costs
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer
 11
Support of On She
Observation or Study of Fish,
Wildlife, and Plants for
Leisure, Educational, or
Scientific Purposes
Change in Quantity or Quality of
On-Site Observation or
Study Activities
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Travel Cost Method
Benefits Transfer
 12
Support of Indirect, Off-Site
Fish, Wildlife, and Plant  Uses
(e.g. viewing wildlife photos)
Change in Quantity or Quality of
Indirect, Off-Site Activities
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Travel Cost Method
Benefits Transfer
 13
Provision of Clean Air through
Support of Living Organisms
Change in Human Health or
Health Risks Attributable to
Change in Air Quality
                                       Change in Animal Health or
                                       Health Risks Attributable to
                                       Change in Air Quality
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic  Price/Wage
Averting  Behavior
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Averting  Behavior
Benefits Transfer

-------
                                                                                                                     17
  Table 2. Continued
   14
Provision of Clean Water
through Support of Living
Organisms
Change in Human Health or
Health Risks Attributable to
Change in Water Quality
                                          Change in Animal Health or
                                          Health Risks Attributable to
                                          Change in Water Quality
                                          Change in Value of Economic
                                          Output or Productions Costs
                                          Attributable to Change in Water
                                          Quality
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent  Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent  Valuation
Averting Behavior
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent  Valuation
Benefits Transfer
   IS
Regulation of Climate through
Support of Plants
Change in Human Health or
Health Risks Attributable to
Change in Climate
                                          Change in Animal Health or
                                          Health Risks Attributable
                                          to Change in Climate
                                          Change in Value of Economic
                                          Output or Production Costs
                                          Attributable to Change in Climate
Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Hedonic Price/Wage
Averting Behavior
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Conlingenl Valuation
Averting Behavior
Benefits Transfer

Market Price/Demand Function
Supply or Cost Function
Consumer/Producer Cost Savings
Contingent Valuation
Benefits Transfer
   16
Provision of Non-Use Services
(e.g., Existence Services)
Associated with Surface Water
Body or Wetlands
Environments
or Ecosystems Supported by
Ground water
Change in Personal Utility
or Satisfaction
                                                                             Conlingenl Valuation
                                                                             Benefits Transfer
•This table was developed with the input and assistance of the U.S. EPA Ground Water Valuation Group directed by Charles Job and
Mary Jo Kealy.

-------
                                                                                  18
 policy is targeted towards the recharge aquifer and the other is targeted towards land
 run-off.  Assume the policies will collectively increase recreational fish catch by 50%.
 The economic value of this increase in fish catch cannot be attributed to both policies.
 In order to avoid double-counting, the total economic value of this increase is fish catch
 should be divided between the two policies based on the relative contribution of each
 policy to the 50% increase in fish catch.
       Because of the interrelationships between ground water and surface water,  surface
 water recharge to ground water and from ground water discharge to surface water, the
 aquifer functions listed in Tables 1 and 2 are not independent. Ground water recharge
 and discharge are both part of the water reserve or stock function because each affects
 the quantity and quality of water which exists in an aquifer in  a given time period.
 Ground water recharge and discharge also are  both part of the surface discharge function
 because both affect the quantity and quality of surface water.  Because ground water
 discharge affects a different set of economic services supported by surface water quantity
 and quality, we include ground water discharge to surface water as a separate function
 (primarily for economic benefit accounting purposes).  From a biophysical or ecologic
 perspective, however, it should be kept in mind that our two broad functions are highly
 interrelated.  Interrelationships between these two functions need to be accounted for
when modeling the  linkages between policy changes, changes in ground water quantity or
quality, and changes in economic values, as illustrated in Figure  1.

-------
                                                                                  19
Ground Water Services
       As with value, we use the term "service" in a neutral sense to imply that a service

is neither inherently good nor bad.  Services may have both  positive and negative effects.

depending upon the affected party's preferences or perspective.  Services associated with


the water reserve or stock function  are listed in Table  1. A  major service with this

function is the provision of drinking water. In the United States, ground water accounts

for about 35 percent of public water supplies and 80% of rural domestic supplies

(American Institute of Professional  Geologists, 1985).  Overall, ground water supplies


dnnking water to 53 percent of the  U.S. population (this figure includes private wells).

Ground water is also extracted for use in irrigated agriculture, many industrial purposes,
                                                                  r
heated water for geothermal power plants, and cooling water for other power plants.

       In some regions of the United States, ground water provides the service of

supporting underground water/soil structure which acts to prevent land subsidence

(sinkholes).  The water storage function also helps to control flooding and erosion by

providing a medium for absorbing surface water run-off. The underground water/soil


structure of an aquifer also provides a  medium for the absorption, transport, and dilution

of wastes (e.g., sewage) and other by-products of human economic activity.  Note that

each of these services are jointly provided by soil structure and ground water in a given

area.  As with the services of the surface water discharge function, the incremental

(marginal) contributions of ground water to these services must be quantified.


       An aquifer  may also generate non-use or passive use  services (Bishop and Welsh,

1992;  Freeman, Chap.  5, 1993).  For example, these services may be attributable to the

-------
mere existence of an aquifer, independent of any current or future use.  Alternatively,

passive use services of providing potable drinking water to future generations may arise

from bequest motivations on the part of the current generation.

       Most major services provided by ground water under the water reserve or stock

function are also included as indirect services associated  with the surface water discharge

function (Table 2).  To the extent that ground water supports healthy and abundant

surface waters, it also contributes to a variety of services generated by these

environments.  These services include recreational swimming,  boating, fishing,

hunting/trapping and plant gathering, and commercial fishing, hunting/trapping and plant
                                                           *
gathering.  Unless biophysical data are available to identify ground water's marginal

contributions to these services, there is a high probability of double counting such that

surface water values may be  assigned to ground water or vice versa.

Effects of Changes in Ground Water Services

       Moving towards  the goal of estimating changes in economic values (Box 9, Figure

1), we need to identify the effects  on (changes in) economic activities resulting from

changes in ground water services.  Examples of potential effects on economic activities

are listed in the second columns of Tables 1 and 2.  Under the "stock" function, for

example, the potential effects of a change in the provision  of drinking water include a

change in utility from an increase or decrease in the availability of drinking water

(access/quantity) and a  change  in human health or health risks (quality).

       Defining changes in human health or health risks  requires careful consideration of

such issues as changes in mortality and morbidity, and cancerous and noncancerous

-------
                                                                                 21

health threats.  Identification of the various types of health effects which can result from

changes in ground water quality requires input from health professions.  What is

ultimately needed are dose-response models that link contaminant sources to  changes in

contaminants in ground water and then changes in human health. These dose-response

models will facilitate defining the  baseline and alternative service flows (S° and S1) and

the estimation of policy-relevant values.  Such linkages are essential for identifying

changes in all service flows, not just human health effects.



Measuring Economic Values

      Complete  valuation of a change in the condition of ground water involves

measuring the economic values for all relevant changes in ground water services

associated with changes in the X and Q vectors.  Economic values for ground water

protection or remediation should  capture the value for the total change in the ground

water condition (X1 - X°,  Q1 - Q°).3  Thus, as suggested in the previous section,

extensive  knowledge of the ground water resource itself and its functions are crucial to

defining the change in  service flows, and the effects on economic activities of these

changes in service flows.

      Once changes in ground water services are identified and quantified (Box 8,

Figure 1), the final step in the benefit estimation process is to assign monetary values to

these service changes (Box 9, Figure 1).  When measuring the economic value of
    3 See Boyle and Bishop (1987) for an application of total valuation to valuing
endangered species.

-------
                                                                                22
environmental changes, theoretically appropriate measures of changes in consumer and
producer welfare (or well-being) must be used.  There is a consensus among economists
that Hicksian compensating or equivalent welfare measures should be used (Freeman,
Chaps. 3 and 4,  1993; Just, Hueth, and Schmitz, 1982; Varian, 1978).  Because of
problems with estimating willingness to accept, the most commonly applied measure of
natural resource economic values is an individual's maximum willingness to pay (WTP).
Hicksian WTP measures (compensating or equivalent) should reflect both the quantity
and quality dimensions of the ground water resource being valued.
      A number of empirical techniques are available for estimating changes in
economic value associated with  changes in ground water services. We do not attempt to
define and explain each potential valuation technique in detail in this report.  An
overview of valuation  techniques relevant to ground water quantity and quality is
provided in Appendix A of the "Guidelines for Performing Regulatory Impact Analysis"
(U.S. EPA, 1983). More detailed descriptions of valuation techniques for environmental
policies, including advantages and disadvantages of the various techniques, can be found
in a number of references (e.g., Braden and Kolstad, 1991; Freeman,1993). We list
potential valuation techniques for changes in ground water services in the last column of
Tables 1  and 2.  Although we advocated estimates of Hicksian welfare in the preceding
paragraph, each  of the techniques listed in the tables that utilize market, or choice, based
data yield estimates of Marshallian surplus, i.e., income is held constant rather than
utility. We do not intend to  imply that estimates of Marshallian surplus are not
appropriate for valuing ground water.  Rather, these are  not the conceptually desired measures.

-------
                                                                                  23
      Selection of a valuation technique for a particular policy application (e.g., RIA)
involves many considerations.  All else constant, techniques that measure maximum
Hicksian WTP with minimal bias are preferred.  Consumer/producer cost savings
estimates, for example, may only provide minimum estimates of value because they do
not reflect maximum WTP based on consumer preferences or producer production
functions. Another major consideration is data availability.  In many environmental
valuation situations, revealed preference data (e.g., water market data) are not available.
In contrast, contingent valuation relies on stated preference data (e.g., data on
preferences obtained directly from people in a survey setting), measure Hicksian WTP
directly, and can be applied to value a wide variety of the services listed in Tables  1 and
2. The largest distinctions between contingent valuation and revealed preference
techniques is that contingent valuation measures Hicksian surplus and is the only
methodology capable of measuring nonuse  values. The application of contingent
valuation to measuring nonuse values, however, is currently a subject of much debate
(e.g., see Arrow et a I, 1993).
      Other important factors an analyst must consider when selecting a valuation
technique include the time and expense involved in implementing the technique as
compared to the timing of policy decisions  for which the value estimates are needed and
the available budget for the data collection and value estimation process.  Related to the
time and expense of implementing a valuation technique is the decision-makers desired
levels of accuracy and reliability associated  with value estimates. In general, increased
accuracy and reliability (in a statistical  sense) requires greater  allocations of both time

-------
                                                                                   24
 and money.  For certain policy decisions, extremely high levels of accuracy and reliability
 may be required.  For other policy decisions, decision-makers may be able to tolerate
 ("make do with") lower levels accuracy and reliability.
       In a number of cases, the selection of a valuation technique, or techniques, is a
 fairly clear-cut decision (e.g., data availability may dictate the decision).  In other cases,
 the decision may not be so clear.  The final selection is likely to involve a "balancing" of
 all relevant considerations (e.g., theoretical consistency, data availability, estimation
 robustness, time constraints, budget constraints, acceptable accuracy and reliability).

 Aggregation Issues
       Once the economic value of ground water to an individual is determined,
 aggregate economic value is estimated by summing individual economic values (e.g.,
 mean willingness-to-pays) over the total number of people in the "market area" of a
 particular aquifer who utilize water from the aquifer, and summing these values over
 time (Freeman, Chap. 7, 1993).  For a given aquifer, there are likely  to be different
 market areas associated with each of the services listed in Table  1.  Determining the
 scope of these market areas is a complex process, involving careful study of the spatial
 distribution of consumers and producers who benefit from the services of ground water
 from a specific aquifer.
       There is not, however,  a clear consensus in the  literature as to how to determine
market size.  Nearly all environmental economists agree that the market should include
all individuals who are affected by a change in  the condition of ground water resource,

-------
                                                                                  25
but this agreement breaks down when discussions move to who specifically is affected.
This problem is exacerbated for nonuse values.  In addition, physical data is often missing
to develop direct links between changes in ground water and potentially affected
populations, as we will note in the review of existing ground water valuation studies (the
does-response function called for above).
       Ground water policies also result in changes in the flow of ground water services
over some time horizon (e.g., 50 years).  The economic value of the policy in each time
period (t) is the difference in the value of ground water quantity and quality with the
policy  in that time period (X],QJ) and the value of what ground water quantity and
quality would have been without the policy (X?,Q°).  That is,
       AS, = S|(X|,Ql) - S«(X?,Q!).
The total value of the ground water resource over the  planning horizon (T) is the
discounted sum of the values attributable to all individuals affected by the change in
ground water services in each time  period (AS,).

Uncertainty in  Ground Water Valuation
       Because we have to deal with imperfect data regarding the quantity and quality of
ground water, the actual changes in ground water services may be uncertain with
associated probabilities of occurrence. This uncertainty may exist with respect to both
the current level  of services (S°) projected into the future and the alternative  level of
services (S1).  Thus,  we are dealing with  expected, rather than deterministic, changes in
services.

-------
                                                                                   26

       The expected changes in ground water service flows is a function of possible

                                                           »
 alternative changes in  the baseline and future ground water conditions, and the


 probabilities of each one of these alternatives occurring.  In some situations, there may


 be a number of possible alternative service flow changes, each having a different


 probability of occurring. In other situations, there may be only one service flow of


 interest with several competing policies for accomplishing the goal and each  policy has a


 different probability of success.


       When demand  and (or) supply uncertainty are present, measures of economic


 value (e.g., willingness-to-pay) should reflect  this  uncertainty. The appropriate welfare


 measure is option price (Bishop, 1982; Smith, 1983; Freeman, Chap. 8,  1993).  Option


 price is defined as a representative individual's maximum willingness-to-pay to obtain a


 specific ground water condition with certainty.  Measurement of option prices is primarily


 accomplished using contingent valuation  (Mitchell and Carson, 1989).




 Intergenerational Issues


       In many cases,  the effects of ground water depletion and contamination may be


 long-term in nature, raising concerns related  to intergenerational equity and


 irreversibility. The process of discounting benefits to calculate present values


 automatically downweights future benefits. Assuming the same monetized value of


 aggregate benefits in each time period, discounting results in an ever decreasing present


value of benefits in each successive time period.  After a certain point in the future (e.g.


 50 years), the discounting process renders the present value of future benefits trivial.

-------
                                                                                  27
Thus, it is sometimes argued that the process of discounting or downweighting future




benefits to calculate present values is "unfair" to future generations. Moreover, the




benefits, costs and discount rate used in any analysis are solely representative of the




preferences of the current generation.




       Intergenerational equity or fairness concerns have resulted in debates over how




best to (or not to) discount future benefits.  These concerns have often focused




discussion on the choice of a discount rate to use  in calculations  of net present values.




Individuals and groups who desire to see more weight placed on  future benefits, for




example because of concern over the well-being of unborn generations, argue for  lower




discount rates.  Individuals and groups who are more worried about the negative effects




on the current economy of reducing current private consumption argue for higher




discount rates (Sassone and  Schaffer, Chap. 6, 1978).




       The discount rate used  in ground water policy analysis, or the analysis of any




public program,  is based on  societies' marginal time preference for consumption.  Since




this concept is difficult to quantify, we believe the choice of a discount rate is




fundamentally a normative decision.  In  the case of environmental policy analyses, this




decision has been made by some branch or office  of the federal government  (Office of




Management and Budget, 1992).  That is, the discount rate which should be used  to




discount future ground water benefits (which reflects some subjective assessment of the




preferences of future generations and weighting of their well-being) is  "handed down" to

-------
                                                                                     28

 policy analysts.4  Although ground water policy analysts may be required to use a certain

 discount rate, the present value of future ground water benefits can be calculated using a

 variety of discount rates to assess the sensitivity of present-value calculations to the

 choice of a discount rate.  Sensitivity analyses should not be used to identify a desired

 outcome, but to examine the effects of a number of plausible discount rates.

       Concerns over the  effects of current policy decisions on future generations

 intensify when suspected irreversibilities are present. For example, suppose a particular

 aquifer is threatened by contamination, purification of  the aquifer would be extremely

 costly and  natural filtration may take decades or longer.  Also, suppose that the aquifer is

 not currently a significant  source of water for human use. However, there is a chance,

 because of population growth, that the aquifer may become a major source of water for

 humans in the future. The uncertainty of future population growth combined with the

 discounting process may result in very  low weights  being  placed on  the possible future

 benefits of protecting the aquifer from contamination.  Consequently, a policy'to protect

 the aquifer from contamination  may not pass a standard  benefit-cost test.

       Whether or not these costs should be borne by future generations is largely a

 normative  issue.  The flip-side of the issue is that protecting the aquifer from

 contamination may impose major costs on the current generation.  Paying these costs

 may reduce the well-being of the present generation, and could end up having little or no
   4 Benefit estimates are based on the preferences of the current generation and the choice of a discount
rate is based on the preferences of the current generation. Benefit-cost analyses, therefore, contain the
implicit assumption that preferences do not change over time. Special concern for future generations only
enter if nonuse values, based on bequest motivations perhaps, arc included in (he benefit assessment.
                                                                                        J

-------
                                                                                29
effect on future generations if future demand for water from the protected aquifer never
materializes.
      When uncertainty and irreversibility are  major issues and benefits to future
generations are of concern, the costs to the present generation of protecting  ground
water should be considered but may not comprise the definitive decision criteria.
Although the economics of a safe minimum standard (Bishop,  1993) for resource
protection are not clear (Ready and Bishop, 1991),  decision makers may still want to
consider protecting selected ground resources if the costs to the present generation are
not unreasonably  high. In such cases, ground water managers may want to develop
several policy scenarios for protecting ground water resources and then investigate the
cost effectiveness  of accomplishing the protection programs.  The question remains
whether the protection costs are  unreasonably high  since benefits no longer play a central
role?  This again  is a normative decision which must eventually be made at some
administrative level.

              HI. PREVIOUS GROUND WATER VALUATION STUDIES
      Although we acknowledge service flows of ground water received by both private
individuals and commercial interests, our exposition in this section focuses on ground
water values held by individuals.  The parameters of ground water valuations differ
between applications to consumers and commercial interests, but  we do not loose
generality regarding the complexity of commodity specification by considering one group
of users. Previous ground water  valuation studies have used contingent valuation (Boyie,

-------
                                                                                30
 1994; Boyle et al., 1994) avoided costs (Raucher, 1986) or avoidance expenditures
 (Abdalla et al, 1992; Abdalla, 1994) to estimate ground water benefits held by individuals.
       To illustrate the scope of work involved with defining and measuring ground water
 values, we  include a review here of the previous ground water studies which used
 contingent  valuation.  When applying contingent valuation to measure ground water
 values, it is necessary to explain the change in the condition of ground water (commodity
 definition)  to survey respondents. The full complexity of the ground water valuation
 problem is  encountered head-on when attempting to explain the ground water
 commodity to survey respondents in contingent valuation studies.
      To our knowledge there have been nine contingent-valuation studies of ground
water conducted to date;  eight in the United States and one in the United  Kingdom. We
discuss the  studies conducted in the U.S. (Table 3).5 The first study was conducted by
Edwards (1988) and estimated the benefits of reducing the probability of ground water
contamination in the community of Falmouth, Massachusetts.  Shultz (1989) also
estimated the benefits of reducing the probability of ground water contamination, but in
Dover, New Hampshire (see also Schultz and Lindsay, 1990). Sun (1990) estimated the
benefits of  protecting ground water in Dougherty County, Georgia such that
contamination levels would be below U.S. EPA health advisory standards (see also Sun et
al., 1992).  Powell (1991)  evaluated the protection of ground water in 15 communities
located in Massachusetts  (four), New York (four) and  Pennsylvania (seven) (see also
Powell and Alice, 1991).  Caudill (1992) estimated the benefits of protecting ground
     See Boyle (1994) for a more complete discussion of these studies.

-------



























(A
a
t
a
e
e
•5
o
U
i.
1
*H
w
a
Q^
rS

S
(2











o
00
c
Q
*o
1
o
co


c
_o

<0
.5
^•i4 *C
o 5
H.|
f^O





C
.2
'•£*
^
a
1
,^

(A

P
c
o
p> CO
*"• U
O -jj
•^ v^

o
u'^-^
3 "H.
W5 p..,
•E S
~< .y

II
^j Q^
&J C
3 0
O M
100% private wells (
screen out individual

•





1/5
O
to
IM
Z
»


c
8
2l
c c
^ s
"El4
s <«-l
*° «
CO ><
—i -
{J o
is!
•5-g
Si
^3* OA

•


•o
»
In
o -o
« .a
.•s «j
Z c.
•


U
1
1
*o
*
s.g
si
^
^1
r- «
CO t?

•
c

^o
ON "O
O CO

3 "§ &
ca co 2
UO C
E
8
«Hl
03

W3
^
O
(u
3
>•> o
>
3 0
<"o |
c t: *
•S g.^
|s|

•


•o
-— '
'o

o

•


ej 5 u-
u * *O ti CJ
^3 'E 1 ^
PC 3 «
< oo '- E c .S
«« t^^ " C O ff, C
J ° -2 g ° = .2
S « .S 'E ^ 8 u •-
>iE|S^oS
2\OOOm«Wo
[J_1inou-->5u

•


u
13
C
JO
"53 ^— s
n ft)
\j ^
O &\

M
>,
E &
O 3

>> -^
Q. JO
C. 3
3 Q.
tfl
^ u o
o, 5 .^
18% private wells
82% public water su
-communities draw v
ground water
-percent of commun
from 0% to 100%

• •
c
c < ^ >
"^ f^r ^
""* (5 '*—*' W
^J .wm ^^ £J
a c> ±
•5 3 Z "3
•*-« O * S?

[S vo ^^i5 o
• «
52
eo g^
•n >^ O
w *^> ^.^
||1
S .5 1 §
-§ I •§ 1
'5 § 'E 'E
3 -E 3 n
E E E S
E 2 E o
o c o u
u o u <*•*
r- u oo °

• •


o
^j
o\ "O
s-x CO

CL, tX C'









W5
^
*

ts
Q.
U
Z

•
?
15
I**
*« ^"O
w Q
3 yi *.-•
HI
^ *j "O
»
Q.
Q.
3
(A
O
«
89% public systems
-communities draw v
ground water
11% private wells

• •





S
«
z
z
•






^
jj
ce
CO

•


I

W5
"O
U.
co

-------
                                                                                  32

water in Michigan (see also Caudill and Hoehn, 1992).  McClelland et al. (1992)

estimated the national benefits of cleaning ground water contaminated by landfills.

Jordan and Elnagheeb (1993), like Sun, estimated the benefits of protecting ground water

so that contamination levels would be below health advisory levels, but for the entire

state of Georgia. Finally, Poe (1993) estimated the benefits of protecting ground water
                                            •
so contamination levels would not exceed health advisory levels in Portage County,

Wisconsin.  In Table 3 we cite the most recent study first and then work backwards

listing studies in reverse chronological order.


      Despite their small number, these ground water valuation studies  present a wide

variety of applications.  In the geographical dimension, for  example, the applications
             i
range from individual communities (Powell, 1991; Shultz, 1989; and Edwards, 1988) to

counties  (Poe, 1992 and Sun, 1990) to states (Jordan and Elnagheeb,  1993; Caudill, 1992)


to national estimates (McClelland et al., 1992). This diversity presents both advantages

and disadvantages.  The advantage is available value estimates potentially reflect a


variety of ground water conditions at the study sites that enhance the  potential for these

studies to collectively provide the value data necessary for accomplishing a RIA.  The


disadvantage is there is very little depth to the value data pertaining to specific attributes

of ground water conditions.


      All eight studies focus on quality dimensions of the "stock" function of ground

water. This focus is an artifact of the studies being primarily designed to value ground

water as  a source of drinking water.  Although changes  in the quality  of ground water

can affect the quality of surface waters, we suspect hydrologic data were  not available to

-------
                                                                                 33
make these connections.  All of the studies, except McClelland et ah (1992), employ the
implicit assumption that the stock of ground water is currently sufficient to meet demand,
but the quality of supply is  threatened by contamination. McClelland et al. ask their
survey respondents to assume that contamination will result in a shortfall of potable
water.
       In Table 3 we consider the condition of ground water in each of the study areas
before presenting the studies' baseline and reference ground water commodity
specifications. Four of the  studies have information that indicates ground water in the
study areas is contaminated (Poe, 1992; Caudill, 1992;  McClelland, 1992; and  Powell,
1991), and the other four implicitly assume the current condition is uncontaminated, or at
least is below health advisory standards.  The question marks beside  the entries for these
latter four indicate that we  are unsure what survey respondents assumed regarding the
current groundwater conditions when answering the valuation questions.  Poe (1993)
established contamination levels by mailing respondents water testing kits with which
water samples were submitted for analysis.  McClelland et al. (1992) asked respondents
about their knowledge of ground water contamination  in their community and selected
one subsample in a location with a history of contamination.  Powell  (1991) selected
communities  for study based on whether they had a history of ground water
contamination.
       Three studies considered nitrate contamination  (Poe, 1992; Jordan and Elnagheeb,
1993; and Edwards, 1988), while two studies considered concurrent nitrate and pesticide
contamination (Caudilt, 1992; Sun, 1990), one  study considered chemical and  diesel fuel

-------
                                                                                 34
contamination (Powell, J991), and the type of contaminates were not specified in the
McClelland et aj. (1992) study.  Respondents to the McClelland et al. (1992) survey were
told contamination was from a landfill. We presume respondents employed subjective
perceptions as to what contaminants were leaching into ground water.
      Table 4 outlines the commodity descriptions used in each of the studies, and it is
these descriptions that form the link between the physical data on ground water
conditions at study sites, as discussed in Figure 1, and service flows provided by ground
water within each study area (Tables 1 and 2). It is important to note that most of the
studies asked respondents to evaluate more than one scenario of ground water
contamination.  In the discussion here we focus on selected scenarios that give the flavor
of the commodity descriptions employed in the studies.
      Having previously discussed the current ground water condition in the study areas,
as presented in study  publications, it is interesting to note the  reference ("without policy")
condition respondents were asked to assume when answering the contingent-valuation
questions.  Poe (1993), Jordan and Elnagheeb (1993) and McClelland (1992) provided
information in the survey questionnaire which objectively defined the reference condition

-------
                                                                              35
Table 4.  Information Presented on Ground Water Commodity (Change in Services)
Author(s)
(Publication dates)
Reference Condition
Subsequent Condition
Poe (1993)
    Stage I - respondents
    old 18% of wells above
    health standard
    Stage II - well-specific
    test results provide
    Below health standards
Jordan and
Elnagheeb (1993)
  • Asked to assume level
    of nitrates exceed safety
    standard
  • Reduce levels to below
    safety standard
McClelland et al.
(1992)
  • Asked to assume 40% of
    supply from
    ground water  and in
    contaminated
    Complete cleanup
Caudill (1992)
    Subjective perceptions
    measured
    Well water - eliminate
    health threat
Powell (1991)
    Respondents subjective
    rating of ground water
    condition (unsafe,
    somewhat safe, safe, or
    very safe)
    Very Safe - "I fee!
    absolutely secure.  I have
    no worries about the
    safety of the community
    water supply at present. I
    am certain the level of
    protection is excellent and
    I cannot foresee any
    contamination occurring in
    the future."
Sun (1990)
    Subjective perceptions
    measured
    Protect so below EPA
    health advisory levels for
    pesticides and fertilizers
Shultz (1989)
    Not specified objectively
    or subjectively
    Reduce potential of
    contamination (increment
    not specified)
Edwards (1988)
    Subjective perceptions
    measured
  • Future contamination -in
    5, 10, 20, or 40 years
    -0%, 25%, 50%, 75% or
    100% probability of
    contamination

-------
                                                                                  36
of drinking water services. Powell (1991), Sun (1990) and Edwards (1988) measured
respondents' subjective perceptions of the reference condition of drinking water services
in the study areas.  In  these studies, the investigators appear to have made a conscious
decision to conduct the valuations based on respondents subjective perceptions of the
reference condition. Caudili (1992) and Shultz (1989) did not establish either an
objective or subjective reference conditions for their valuation exercises.
       All eight studies specified the subsequent ("with policy") condition of services.
Each study took a different approach to describing the change in services to be valued as
defined by equation 1, some providing more complete definitions than others.  Poe
offered the most complete commodity definition.  Poe conducted his study in two stages.
In the first stage respondents water was tested for contaminants.  In the second stage the
well-specific test results were used to  set the reference condition and the subsequent
condition was below health standards. In contrast, Powell's respondents rated current
conditions on a four point scale, ranging from  "unsafe" to "very safe".  In the valuation
exercise, respondents' subjective rating of the current condition of drinking water services
became the  reference condition, and then stated a value for an increase in water quality
to a rating of "very safe".  This approach allowed  respondents to translate the
information  presented  and frame their own commodity definitions when responding to
the contingent-valuation questions.
       Studies attempting to completely (Poe,  1992; Jordan and Elnagheeb,  1993; and
McClelland, 1992) or partially (Sun, 1990; Schultz, 1989; and  Edwards, 1988) frame the
change in ground water services have both strengths and weakness.  The strength of

-------
                                                                                    37

completely specifying commodity descriptions is experimentally induced bias and variation

in valuation responses may be reduced. The disadvantage is respondents may reject the

objective information resulting in valuation responses that are based on subjective

perceptions (Kask and Maani, 1992; and Lichtenstein, et  al., 1978).6  The Powell study

meets this hurdle head on, but also raises questions.  For instance, how can value

estimates, based on subjective perceptions, be linked to actual changes in ground water

conditions?  These are fundamental issues in any environmental commodity valuation

study. These questions must be  addressed if ground water value estimates  are to be

useful for public-policy analyses.

       A basic insight from this overview is that the library of ground water valuation

studies measuring individual values is very  thin in terms of the number of studies, and

consequently, in terms of values  for specific dimensions of ground water.  For example,

all eight studies account for only the direct provision of drinking water service (service

row 1, Table 1). This implies a need for more primary data on values for other ground

water services if the library of ground water value studies is going to be sufficient for

RIAs (and other policy needs).  Original valuation studies are needed for all of the

potential service/effect flows of ground water identified in Tables 1 and 2.

       Another basic insight from  this overview is that to be useful for policy assessment,

valuation studies must be very detailed and complete. For example, following the

valuation framework summarized by Figure 1 and Tables 1 and 2, the basic ground water
   6 The possibility of subjective editing of information points out the desirability of eliciting information
about respondents' subjective assessments of the ground water valuation scenario so that these subjective
assessments (e.g., subjective risk assessments) can be incorporated into value estimation, interpretation and
application.

-------
                                                                                  38
information required for policy analysis is changes in ground water service flows.
Assessment of this change requires knowledge of the current (baseline), reference, and
subsequent ground water conditions.  In general, the descriptions of the current,
reference, and subsequent ground water conditions are quite vague in the eight studies.
This vagueness makes it difficult to establish the linkages between changes in ground
water policies, ground water conditions, services provided, and estimated values.  Of
particular concern is the difficulty of ascertaining how the value estimates correspond to
actual biophysical changes in ground water resources and the resulting change in  service
flows.
       If valuation studies do not provide sufficient information for establishing the
technical linkages illustrated in Figure 1 and Tables 1 and 2, the usefulness of valuation
estimates for policy assessment is greatly reduced. Valuation studies need to measure
values for changes in service flows that have clear linkages to biophysical changes in
ground water  resources.  To complete the policy assessment process, clear linkages must
also be established between changes in ground water policies and biophysical changes in
ground water  resources.  Improvements are needed in the assessments conducted by
physical scientists and economists, and these investigations need to work to enhance the
interfaces between these analyses. The difficulties encountered when assessing changes
in ground water policies are further illustrated by considering two policy assessment case
studies in the  next section.

-------
                                                                                39

    IV. GROUND WATER VALUATION AND REGULATORY IMPACT ANALYSES

       U.S. Presidential Executive Order 12866 issued in 1994 instructs government

agencies to conduct regulatory impact analyses (RIAs) on all major regulations.  RIAs

are to include assessments of the benefits and costs of the full range of effects associated

with a proposed regulation (USEPA, 1991). The full range of effects includes benefits

and costs which can be quantified monetarily, and those which cannot be quantified

monetarily.7 The RIA  guidelines were developed for evaluating any type of

environmental regulation, e.g., air, surface water or ground water. Our focus is

specifically on ground water.  In the remainder of this section we will review two RIAs

that dealt  with ground water resources while complying with the overall RIA guidelines.

Our general process for evaluating these RIAs is  to consider how the benefit assessment

components correspond to the framework we have proposed in this report.



Draft Class V Injection Well Regulatory Impact Analysis

      The purpose of  this RIA was to  consider the benefits of regulating Class  V

industrial wells. Four types of industrial facilities operating Class V injection wells were

considered as case studies: automotive repair, dry cleaning, metal fabrication and

electroplating.  Within each industry actual pollution incidents or events were considered.
   7 Although our focus is on potential benefits, the discussion also provides insight on
potential costs of a proposed regulation since social costs are often foregone benefits (or
opportunity costs).

-------
                                                                                40
       Class V injection wells represent a case where groundwater is used as a medium
to dispose of wastes (Row 10 in Table 1).  In  the current RIA, disposal is presumed to
pose a human health threat so injections of waste are being proposed for regulation.
       Within the empirical section of this RIA, Chapter 4, a very specific perspective is
taken regarding benefits. Detailed breakdowns of contaminants and contaminant
concentrations were developed for the RIA.  No discussion of the dispersion of the
contaminants were provided and explicit consideration was not given to the multifaceted
ground water services documented in Tables 1 and 2.  Only human  health risks were
quantified, the quality component of Row 1 in Table  1. Although a larger domain of
benefits may have been considered for inclusion in the analysis, we  could not discern this
from the available documentation.
       Benefits were computed using breakeven analyses (contaminant concentration
resulting in zero  net benefits) and an averting behavior approach  (avoidance cost).
Health benefits for the breakeven analysis were  computed using the number of statistical
lives saved and a range of values from the literature were employed.  Uncertainty was
considered in the analyses by considering the expected efficiency of proposed regulations.
Avoidance costs  were based on the most cost  efficient response to the contamination
events and uncertainty was factored in by considering the probability that contamination
would  go undetected. Both of these approaches are likely to give minimum estimates of
value because they do not reveal the  public's maximum willingness to pay to avoid
contaminated ground water.
                                                                                    J

-------
                                                                                41

       The primary questions that arise when the analyses for this RIA are compared to

the ground water valuation framework in Section II are:

       •      Were important benefit categories omitted?

       •      Were benefits of reducing health risks underestimated?

These  issues may not be relevant for the RIA, but the available documentation does not

allow us to answer these questions.

Draft RIA for Final Rulemaking on Corrective Action for Solid Waste Management
Units

       The purpose of this RIA was to  present methodology to be used to estimate costs

and benefits of site cleanup at hazardous waste facilities regulated under the Resource

Conservation and Recovery Act. An application is included to provide an illustration of

the methodology. We concentrate our discussion on the benefits  component of the

application.

       Referring back to Figure 1, the RIA clearly defined the reference and subsequent

ground water conditions and projected these conditions through time as we recommend

in Section II. This work was done through interactions of environmental scientists,

economists and engineers,  an approach we also advocate to provide policy relevant value

estimates.  Notably, the RIA focus on the quality of ground water, thus ground water

services listed in rows  five  through nine of Table 1 can be reasonably excluded because

the physical stock of ground water would not appear to be affected by the action being

evaluated.

       The types of values estimated include human health benefits, ecological benefits,

and nonuse values.  Health benefits  from protecting ground water arose from reducing

-------
                                                                                 42
three paths of exposure: ingesting contaminated drinking water, inhaling volatile
compounds during household use of ground water, dermal uptake while showering. The
pathways of contamination arise from drinking ground water and household uses of
ground water (first row of Table 1).  From the information provided in the RIA we can
not discern whether other indirect pathways of human health effects, (rows two through
four in Table 1) were not considered or were deemed to be  minor or were not relevant.
       The averted water use applied the cost of water treatment as a proxy for benefits,
likely yielding an underestimate of benefits in this category.  This component measures
the access value of potable drinking water in Row  1 of Table 1.  No mention is made of
averting costs for commercial users of ground water, (rows two through seven of Table
1).  If commercial users derive their water from municipal sources, then the benefits
accruing  to these users may have been counted.  If commercial users derive ground water
from private wells and invest in purification, benefits in this category are underestimated.
       National  nonuse benefits were estimated, addressing Row 12 of Table 1.  Nonuse
benefits were not estimated for the function of ground water discharging to surface
water. For each of the benefit categories listed in Tables 1 and 2, data were developed
for specific contamination sites.  The RIA does not discuss how national averages of
nonuse values should, or can, be adjusted for  application to corrective  actions at specific
sites.
       The property value analysis considered the effects on residential property values
located near solid waste facilities. Although this is a valid  method for  estimating ground
water values, property value effects  may result is potential double counting with the use

-------
                                                                                43
value measures.  Precisely, is there double counting with the averted-cost and hedonic-
price measures of benefits? Since an integrative framework for the various benefit
components is not given and the component value estimates are implicitly assumed to be
additive, it is difficult to ascertain if double-counting of benefits occurred.
      The solid waste corrective action RIA appears to be consistent with the ground
water valuation framework we proposed in Section II.  Despite the general consistency of
the approaches, issues that arise when comparing the RIA with our proposed ground
water valuation framework are:
      •     Were some of the indirect effects of contaminated ground water
             inadvertently overlooked, e.g., health effects other than household
             consumption?
      •     Were use benefits underestimated due  to use of averting expenditures and
             not considering  commercial  users of ground water?
      •     Does  the lack of a conceptual framework for integrating the various
             benefits estimates lead to double counting of some benefit components?
As noted for the previous RIA,  these issues do not necessarily imply problems in the
RIA, but do imply an expanded  scope of benefits needs to be considered in the design
and reporting of RIAs.
Summing Up
      The solid waste RIA appears to be much closer to the ground water valuation
framework in Section II than is  the injection well RIA.  This difference  may be due to
reporting or the injection well RIA may indeed have  taken  too narrow of a scope when

-------
                                                                                  44
considering potential benefits of the action.  Given the applications, we ask whether
important benefit categories were omitted in both RIAs and whether values may have
been underestimated for benefit categories considered.  Both of these issues, if present,
will lead to under estimates of total benefits vis a vis total costs of the implementing the
regulations. No information is reported regarding what components of values were
considered, but not analyzed for the RIAs.
       The  injection well RIA explicitly considered uncertainty and did implicit sensitivity
analyses by considering different levels of regulation.  No comparable analyses were
reported for the solid waste RIA.  Given the complexity of ground water resources and
services, uncertainties regarding ground water conditions, and  difficulties in measuring
benefit categories, we strongly urge that all ground water RIAs should consider potential
sources of uncertainty and  conduct sensitivity analyses to investigate the robustness of
assumptions employed in analyses.
       Finally, neither analysis even opened the door for considerations of
intergenerational equity issues.  Although this is a very difficult issue, which we did not
attempt to  solve in  Section II, consideration should be given to the fact that all benefits
and costs arise from the preferences of the current generation given available technology.
Simultaneously, either implementing or not implementing ground water policies can have
substantial  implications for ground water resources available to future generations.
       There are several key points to consider when addressing the issues raised in our
overview of the two RIAs and developing systematic ground water evaluations for  future
RIAs and other  policy assessments.  These key points are illustrated by our valuation

-------
                                                                                45

framework summarized in Figure 1 and Tables 1 and 2. An RIA must first assess the

biophysical condition of a ground water resource "with" and "without" the proposed policy

change.  It appears that the RIA addressed effects where biophysical data were available

and did not report potential effects that could not be documented with available
                                                                l
technology or data. More research is needed to develop data bases and models to assess

the effects of ground water policies on biophysical changes in ground water resources

(Boxes 1-7 in Figure  1). The  next issue faced  in conducting an RIA is to identify how the

policy-induced changes in the  biophysical condition of a ground water resource will


change ground water service flows (Box 8,  Figure 1). The two RIAs we reviewed only


accounted for a portion of the service flows suggested in Tables 1 and 2.  Future RIAs


should identify potentially affected service flows that were considered and dismissed

because no effect was identified, or the identified effect was quite small, or there was no

data to quantify the effects. Both RIAs used several valuation methodologies (e.g.,

averting behavior, contingent valuation, hedonic price), but are weak because value


estimates can not be clearly linked to specific biophysical changes in the ground water


resources.  The application of economic value  estimates in RIAs  can be improved  by

precisely defining changes in ground water service flows in terms that are relevant  for

economic analysis (using Figure 1, and Table 1  and  2 as a guide).




    V. A STRUCTURE FOR CONSIDERING THE VALUE OF GROUND WATER

      In this section, we  discuss a general process or protocol for EPA offices to follow

when incorporating the economic value of ground water in RIAs.  The  overall goal of

-------
                                                                                46
this protocol is to generate and apply economic value estimates consistently across policy
issues and offices within EPA. Our valuation framework begins to develop the protocol
for this consistency.  In addition, following the protocol may help  EPA Offices to avoid
duplication of efforts and potential double-counting of values.  For example, concise
summaries of previous RIAs would be available enabling future RIAs to explicitly build
on the knowledge developed and experience gained in conducting previous RIAs.  This
effort may be particularly fruitful for transferring knowledge and information through
time, across policy issues within EPA Offices, and across offices within EPA.
       Another useful application of our protocol is that it will provide information for
building EPA's Regulatory Impact Analysis (RIA) Benefit-Cost Database.  Our protocol
is different from the RIA Benefit-Cost Database  in that it provides guidelines for
conducting and reporting benefit assessments in RIAs. The  RIA  Benefit-Cost  Database
is a general reporting of all information contained in RIAs.  It probably is not  practical to
include all of the detailed information about procedures used to assess ground water
values  in the  RIA Benefit-Cost Database.  We  recommend, however, that  all of the
information generated by our protocol be available to supplement the RIA Benefit-Cost
Database, and the Database include information  about where more detailed information
regarding valuation procedures can be obtained.

Protocol  Components
       The first component of our protocol is for the  RIA analyst to record answers to
the following important questions.

-------
                                                                      47
 Name of Proposed action?
 What is the current ground water condition?
       Contaminated —>
       Uncontaminated —>
What are the
Contaminant
contaminants?
Geographic
Concentration Extent




What are the
Contaminant
potential contaminants?
Geographic
Concentration Extent



      Unknown
What is the proposed action?
      Protection —>
      Remediation --->
What are the proposed
policies or rules


What are the proposed
policies or rules?


-------
                                                                       48
What are the sources of contamination?
           Known —
List of sources:
                              Source  Contaminant
           Unknown
What would the ground water condition be over the study time frame without any
action (reference condition)?
          Quantity Quality
  Year 1
  Year 2
  Year 3
  Etc.
What would the ground water condition be over the study period with action
(subsequent condition)?
       Quantity    Quality
  Yearl
  Year 2
  Year 3
  Etc.

-------
                                                                                  49
Answers to these questions relate to Boxes 1 - 7 in Figure  1 and comprise the technical
data necessary for estimation of benefits, Boxes 8 - 9.
      The next component of the protocol is to identify affected services that give rise to
benefit estimates.  This issue relates to both the stock and  surface water discharge
functions (Tables 1 and 2).  Assessment of potential changes in services can be facilitated
by completing matrices such as those shown in Tables 5 and 6. These tables are partially
filled out for a hypothetical regulation.  The first step in completing the tables is to assess
the reference condition for the services listed under each function  in Tables 1 and 2.  For
example, affected services for the stock function are documented in Table 5. The
"Reference Conditions" indicate that the aquifer provides an adequate supply of drinking
water through public or private wells and is uncontaminated.  These quantity and quality
dimensions are known with certainty. The  aquifer is not directly utilized for crop
irrigation, livestock watering, or food processing services, as indicated by the "no" entries
in the second column of Table  5. To clarify interpretation  of  the table all other entries
for  these services are left blank. Thus, the body of the table only documents affected
services.  Completing the first column indicates that a service was considered and
purposely excluded. The information in the first column also  briefly notes why a
potential service is  excluded.
      The entries for the discharge function in Table 6 indicate that the aquifer indirectly
provides water for crop irrigation and livestock watering, but surface water is not used
for  human consumption. Again, quantity is assumed to be  adequate,  but  the

-------
1 TableS. Changesin Ground WaterServices- Stock Function ||
SubsequentConditions |
1/9
'•5
§
jO
u
Uncertainty
T3
0
c
O
Uncertainty
CO
O
c
CB
o

V
So
o c
U
Decrease
Increase
u
oo
o c
Decrease
Increase

Conccn
-tration
Contam
-ination

"S "8 -~
2 i/s u Q ^/
O _o* tx"5 c &*
"5 (X ^
1
B <—
llll
511.1

— o
c 
-------




























o

0
C
3
te/t
£?
CO
"5
«M
D
i
u
w

J3
c
1
O
c
00
6
u
£>






1/9
C
o

'•5
c

c
u
or
u
ri
3
V3
















(A
_g
1
rencflCOr
4J





•Jl*
*s
1
p



>•»
13
3
O










C
re
3
O



C
'5
n

0
c
D


^^
=
0

^>
03
O



o c
6
«
re
u

Q

re
o
u
c

u

o c
U
1

O
Q
CO
u
c







•-t
K =
8.2
C •*•'
6E
Contam
-ination



-o ~o
J*Wf
< (X ^

U
U
'E
u
00










































t-
0
§'o"S
0 o o -w Z a?
z^.5 £ S.3

00
^c
*o

P) C ^^ .M
311.1


^CO w5
*JUUJ (Q
C! W

cud








X











O
^
o
Z


^*
Z
0
I

« ^ _,
cr D ^

-------
                                                                                  52
 quality is threatened by contaminated ground water. The extent and timing of the
 potential contamination is unknown.
       A starting point for assigning monetary values to changes in ground water services
 is an assessment of available valuation data, e.g., the studies reported in Table 3.
 Available value  estimates would be graded as to their suitability for transfer to the
 current ground water valuation issue. For discussions of criteria for selecting value
 estimates see the special issue of Water Resources Research (Vol. 28, No. 3, 1992)
 dealing with benefits transfer. We do not envision this process as being purely qualitative
 (e.g., good, average or poor), but dealing with specific issues of how the available value
 estimates relate  to the current situation being evaluated in the RIA at hand.  For
 example, are the same contaminants involved?  Are the magnitudes of contamination
 comparable? Were the valuation studies conducted adequately, e.g., are estimates biased
 or have large variances?
      As an example, suppose there is a potential decrease  in the quality of drinking
 water provided directly by the aquifer. This change is represented by an increase in the
 concentration of Chemical Z of 30 ppb. As indicated in Table 7a, the proposed
regulation will not affect  the quantity of ground water available for human consumption,
and the aquifer  is not directly used for the other services listed  in Table 5. The
"Increment Evaluated" under "Quantity Changes" is listed as "no effect" in Table 7a.  The
value columns for the quantity change, therefore, are left blank to  facilitate interpretation
of the table.  The increment of contamination  to be evaluated is documented under the

-------
                                                                                    53
Table 7a. Available Data for Valuing Changes in Ground Water Services - Stock Function
Services
Drinking
Water
Crop
Irrigation
Livestock
Watering
Food
Product
Processing
Etc.
Quantity Changes
Increment
Evaluated
No effect
No effect
No effect
No effect

Value
Estimaie(s)





Valuation
Method





Quality Changes
Increment
Evaluated
30 ppb
Reduction
No effect
No effect
No effect

Value
Estimate(s)
None
Available


•

Valuation
Method
N/A




"Quality Changes" heading in Table 7a.  We assume that the water can be made safe for

drinking, but expenditures must be made on water purification.  For our hypothetical

example we assume value data are not available to assign initial values to the reduction

in quality.8

      After assessing available data, additional data needs are identified. This covers

services for which available value estimates are not appropriate and services for which

value estimates do not exist. Continuing with the example, value estimates are only

needed for a reduction in water quality for human consumption under the stock function.

We identify averting cost as a minimum  estimate and contingent valuation as a procedure

for estimating the full value the public places on avoiding potential contamination
   8 A number of Meta analyses of environmental values are being developed.  These
studies could, if developed for ground water valuation (Boyle et a]., 1994), can be a
source of initial value estimates for RIAs (Smith and Huang, 1993; Smith and Kaoru,
1990;  Smith and  Osborne, 1993; and  Walsh et a].,  1988).

-------
                                                                                 54
(Table 7b).  Values included in contingent valuation estimates, but excluded from
averting costs, include disutility from having to invest and maintain filtering systems for
private wells and potential nonuse values.  The question mark in Table 7b for value
estimates indicates the values to be  estimated. After the study is completed, the question
mark would be replaced by the estimate(s).
      Tables similar to 7a and 7b can be developed for function of ground water
discharge to surface water.  We omit this step here for expositional convenience.
      The final step is to identify services that will not be monetized and the reasons for
these decisions (Table 8).  We assume there are no  effects that are not monetorized in
this simplistic example.  We do assume there is a 50% chance of the 30 ppb
contamination actually occurring.  The expected change can be monetorized  in some
instances using appropriate measures of economic value under uncertainty (e.g., option
Table 7b. Needed Data for Valuing Changes in Ground Water Services - Stock Function
Service
Drinking
Water
Crop
Irrigation
Livestock
Watering
Food
Product
Processing

Quantity Changes
Increment
Evaluated
No effect
No effect
No effect
No effect

Desired
Valuation
Method





Value
Estimates





Quality Changes
Increment
Evaluated
30 ppb
Reduction
No effect
No effect
No effect

Desired
Valuation
Mel hod
Contingent
valuation or
averted cost




Value
Estimates
7





-------
                                                                                 55
Table 8. Other Valuation Considerations for Changes in Ground Water Services - Stock Function
Services
Drinking Water
Crop Irrigation
Livestock Watering
Food Product
Processing
Nonmonetorized Effects
(Reason Why)
None
None
None
None
Treatment of
Uncertainly
50% chance of
contamination



Sensitivity Analyses
Geographical extent
of contamination



price described previously).  However,  in some cases this will not be possible.  In such




instances, sensitivity analyses conducted with plausible value estimates can be utilized to




consider the effect of the uncertainty on the outcome of the entire benefit-cost or cost-




effectiveness analysis. Another source  of uncertainty in the current example is the




geographical extent of the contamination.  It is assumed that this factor is not known and




can not be  accurately predicted. Thus, several scenarios of damages might be




investigated to consider the impact on aggregate value estimates.
                          VI. CONCLUDING COMMENTS




      Preparing an RIA that adequately considers the full range of effects of a proposed




ground water regulation is a major undertaking.  Benefit estimation can be facilitated by




carefully identifying, measuring, and documenting the  linkages and "chain of events"




shown in Figure 1, using Tables 1 and 2 as guides for tracing specific linkages between




policies, changes in ground water services and value estimates. These tables guide




identification and quantification of linkages between a proposed regulation, changes in

-------
                                                                                 56
services provided by ground water functions, and the effects of service changes on
economic activities and values. This information reveals the gainers and losers of a
proposed regulation, over both time and geographic space.  Using Table 5, 6, 7a, 7b and
8 will facilitate clear and concise documentation of valuation analyses for RIAs. This
documentation will report service effects valued as well as those dismissed as not
relevant.  It will also insure all RIAs considering ground water values begin at the same
starting point, consider the same issues and provide uniform reporting.  Establishing
structure and consistency within and across EPA offices is important for producing
accurate benefit estimates, avoiding double-counting problems, and eliminating
duplication of ground water valuation efforts.
     We envision these tables as comprising a concise form for reporting all benefit
analyses conducted for RIAs.  The list of questions would comprise a cover sheet to
identify the RIA and ground water issue. Each of the tables would then follow to
complete the documentation.  This reporting framework would provide a systematic way
of documenting and reviewing RIA benefit analyses.  It may also be helpful to document
studies used as secondary sources of value data as has been done by Boyle (1994) for
ground water contingent-valuation studies, and we abbreviated in Tables 3 and 4.

-------
                                                                                57

                                 References Cited

Arrow, K., R. Solow, P.R. Portney, E.E. Learner, R. Radner, and H. Shuman. Report of
     the NOAA Panel on Contingent Valuation.  National Oceanic and Atmospheric
     Administration.  U.S. Department of Commerce. Washington, D.C., January, 1993.

Abdall, C. W., What we know about groundwater valuation: Results and implications
     from avoidance cost studies, Amer.J. Agric.Econ., 76(5), forthcoming,  1994.

Abdalla, C. W., B. A. Roach, and D. J. Epp, Valuing environmental quality  changes using
     averting expenditures: An application to ground  water contamination,  Land  Econ.,
     68(2), 163-169, 1992.

American Institute of Professional Geologists,  Ground water:  Issues and answers, Arvada,
     CO, 1985.

Bishop, R. C,  Economic efficiency, sustainability and biodiversity, Ambio, 22(2-3),  69-73,
     1993.

Bishop, R. C,  Option Value: An exposition and extension, Land Econ., 58(1), 1-15,
     1982.

Bishop, R. C.,  and M. P. Welsh, Existence values in benefit-cost analysis and damage
     assessments,  Land Econ., 68(4), 405-417, 1992.

Boyle, K. J., A review of contingent-valuation studies of the benefits of ground water
     protection, Research Triangle Institute, Cooperative  Agreement CR 818760-01-0,
     1994.

Boyle, K. J., and R. C. Bishop, Valuing wildlife in benefit-cost analyses: A case study
     involving endangered species, WaterResour.Res., 23(5),  943-950, 1987.

Boyle, K. J., G. L.  Poe, J. C. Bergstrom, What do we know about ground water?
     Preliminary implications from a meta analysis of contingent-valuation studies, Amer.
     J. Agric. Econ., 76(5), forthcoming, 1994.

Braden, J. B., and  C. D. Kolstad, Measuring the demand for environmental  quality.
     Amsterdam:  North Holland, 1991.

Cadmus Group.  Class V injection well regulatory impact analysis and  regulatory
     flexibility analysis,  Report to USEPA, Office of Water, Contract  68-CO-0020, Work
     Assignment No. 2-0-27, 1993.

-------
                                                                                58

 Caudill, J. D., The valuation of ground water pollution policies: The differential impacts
      of prevention and remediation, Unpublished Ph.D. dissertation, Department of
      Agricultural Economics, Michigan State University, 1992.

 Caudill, J. D., and J. P. Hoehn, The economic valuation of ground water pollution
      policies: The role of subjective risk perceptions, Staff Paper No 92-11, Department
      of Agricultural Economics, Michigan State University, 1992.

 Edwards,  S. F., "Option prices for ground water protection". J. of  Environ. Econ. and
      Manage., 15(4), 465-87, 1988.

 Freeman, A. M. Ill, The Measurementof Environmental and Resource Values: Theoryand
      Methods. Washington, DC: Resources for the Future, 1993.

 Jordan, J. L., and A. H. Elnagheeb, Willingness to pay for improvements in drinking
      water quality, WaterResour. Res., 29(2), 237-245, 1993.

 Just, R. E., D. L. Hueth, and A. Schmitz.  Applied Welfare Economics and Public Policy.
      Englewood Cliffs, NJ:Prentice-Hall,  Inc. 1982.

 Kask, S. B.,  and S. A. Maani, Uncertainty, information, and hedonic pricing, Land Econ.,
      68(2),  170-184, 1992.

 Lichtenstein, S., P. Slovic, B. Fischoff,  M.  Layman, and B. Canbs, Judged frequency of
      lethal events."  J. of Exper. Psych.: Human Learning and Memory4, 551-578, 1978.

 McClelland, G. H., et al., Methods for measuring non-use values:  A contingent valuation
      study of ground water cleanup.  Final Report, Office of Policy, Planning and
      Evaluation, U.S. Environmental Protection Agency, Cooperative Agreement  #CR-
      815183, 1992.

Mitchell,  R.C., and R.T. Carson, Using surveys  to value public goods: The contingent
      valuation  method. Washington, DC: Resources  for the Future,  1989.

Office of Management and Budget, Guidance and discount rates for benefit-cost analysis
      of federal programs,  Circular A-94,  1992.

Poe, G. L., Information, risk perceptions and contingent values: The  case of nitrates in
      ground water, Unpublished Ph.D. dissertation, Department of Agricultural
      Economics, University of Wisconsin-Madison, 1992.

-------
                                                                                 59

Powell, J. R. The value of ground water protection:  Measurement of willingness-to-pay
     information, and its utilization by local government decision-makers, Unpublished
     Ph.D. dissertation, Cornell University, 1991.

Powell, J. R., and D. J. Alice, The estimation of ground water protection benefits and
     their utilization by local government decision-makers, Final report to the U.S.
     Geological Survey, award  number 14-08-0001-G16494, 1991.

Raucher, R. L., The benefits and costs of policies related to ground water contamination,
     Land Econ., 62(1), 33-45,  1986.

Ready, R. C, and R. C. Bishop, Endangered species and the safe minimum standard,
     Am. J. of Agric.Econ., 73(2), 309-312, 1991.

Sassone,  G., and W. A. Schaffer, Cost-benefit analysis: A handbook, New York:
     Academic Press, 1978.

Shultz, S. D., Willingness to pay for ground water protection in Dover, NH:  A
     contingent valuation approach, Unpublished M.S. thesis, University of New
     Hampshire, 1989.

Shultz, S. D., and B. E. Lindsay, The willingness to pay for ground water protection,
     Water Resour. Res., 26(9),  1869-1875, 1990.

Smith,  V. K., Option value: A conceptional overview, South. Econ. J., 49(1), 654-668,
     1983.

Smith,  V. K., and J. C. Huang, Hedonic models and air pollution: Twenty-five years and
     counting, Environ.and Resour.Econ.,  3, 381-394, 1993.

Smith,  V. K., and L. Osborne, Do contingent valuation estimates pass a scope test? A
     preliminary  meta analysis, Discussion paper, Department of Economics, North
     Carolina State University,  1993.

Smith,  V. K., and Y. Kaoru, Signals or noise? Explaining the variation in recreation
     benefit estimates, Amer.J. Agric.Econ., 72(2), 419-433, 1990.

Sun, H., J. C. Bergstrom, and J. R. Dorfman, Estimating the benefits of ground water
     contamination protection." South. J. of Agric. Econ., 24(2)* 63-71, 1992.

Sun, H., An economic analysis of ground water pollution  by agricultural chemicals,
     Unpublished M.S. Thesis.  Department of Agricultural and Applied Economics,
     The University of Georgia, 1990.

-------
                                                                              60

United States Environmental Protection Agency, A Guide for Cost-Effectivenessand Cost-
     Benefit Analysis of State and Local Ground Water Protection Programs, Office of
     Water Report WH-550G, 1993.

United States Environmental Protection Agency, Guidelines for PerformingRegulatory
     Impact Analysis, Office of Policy Analysis, EPA-230-01 -84-003, 1991.

Varian, H. R., Microeconomic Ana lysis.  New York: W.W. Norton & Company,  1978.

Walsh, R. G., D. M. Johnson, and J. R. McKean, Review of outdoor recreation demand
     studies with nonmarket benefit estimates, 1968-1988, Tech. Rept. 54, Colo. Water
     Resour. Res. Inst., Colo. State Univ., Fort Collins, 1988.

-------