Policy Research

WORKING PAPERS

World Development Report

Office of the Vice PresidentDevelopment Economics

The World BankOctober 1992

WPS 988

Background paper for World Development Report 1992

Economic Valuationand the Natural World

David Pearce

Economic valuation can help improve decisions about protect-ing the environment. By imputing values to unpriced goods, itcan make public choices more cost-efficient and thus allowlimited public income to be optimally spent.

Policy ResearchWoriong Papers disseminate the findingsof work in progress and encouragethe exchange of ideas among Bank staffandall others intersed in developmnent issues.T7hesepapers. distributed by the ReseatchAdvisory Staff.,carr thenames ofthe authors, reflect,ihy theirviews,andshouldboused and cited accordingly.Thefindings, interpretations,andconorlusionsaretheauthors'own. Theyshouldnot be attributed to the Wotdd Bank, its Board of Directors, its management, or any of its member countries.

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Policy Research

World Development Report

WPS 988

This paper- a product of the Office of the Vice President, Development Economics - is one in a seriesof background papers prepared for the World Development Report 1992. The Report, on development andthe environment, discusses the possible effects of the expected dramatic growth in the world's population,industrial output, use of energy, and demand for food. Copies of this and other WorldDevelopmentReportbackground papers are available free from the World Bank, 1818 H Street, NW, Washington, DC 20433.Please contact the World Development Report office, room T7-101, extension 31393 (October 1992, 63pages).

Economic valuation is controversial largely present, and against future, generations wherebecause its purpose has not been clearly con- future costs and benefits are both distant andveyed to non-economists, says David Pearce. significant.The purpose of valuation of the natural world is But economic valuation is useful in severalto elicit measures of human preferences for, or contexts, says Pearce. Project and programagainst, environmental change. As a procedure, appraisal cannot be comprehensive or adequateit thus faces two immediate limitadons, he without it. National environmental policyargues. priorities will be better informed if economic

First, economic values are not the same as values are known with some degree of certainty."intrinsic" values - values "in" things rather The entire objective of sustainable developmentthan values "of' things. Economic valuation almost certainly cannot be interpreted withoutmakes no claim to measure intrinsic values, some idea of the value of environmental servicesalthough through the concept of "existence" and assets.value it may be capable of capturing humar. Empirical work on valuation remainsperceptions of intrinsic value. limited, even in the developed world. It is fairly

Second, measuring preferences focuses on new in the developing world, although manyefficiency gains and losses from environmental project evaluations have used some form ofchange. It says little about the distribution of indirect valuation. Its importance for the devel-costs and benefits within a time period or opment process is that revealed economic valuesbetween time periods. for environmental conservation and environmen-

Within a time period, the use of efficiency tally improving projects and policies have oftengains and losses as a guide to policy or project been found to be large.evaluation assumes that the prevail.ng distribu- Valuation demonstrates that there is antion of income is socially acceptable, since it is economic case for protecting the environment,that distribution which "weights" the measures and can help improve decisionmaking. In soof willingness to pay. doing, it could make public choices more cost-

Between time periods, the use of another efficient, thus allowing limited public income toefficiency concept - the discount rate - biases be optimally spent.the outcomes of evaluation in favor of the

The Policy Research Working PaperSeries dissemninates thefindings of work under way in theBank. Anobjectiveof the seriesis to get these findings out quickly, even if presentations are less than fully polished. The findings, interpretations, andconclusions in these papers do not necessarily represent official Bank policy.

Produced by the Policy Research Dissemination Center

Economic Valuation and the Natural World

David Pearce

Centre for Social and Economic Researchon the Global Environment

London and Norwich, UK

Prepared as a Background Paper for theWorld Development Report 1992

Not for quotation without permission of the author. Contact: David Pearce, 90 Kimbolton Rd,Bedford, MK40 2PE, UK. Fax (44) 234 215528 (0234 in UK)

The World Development Report 1992, "Development a.d the Environment," discusses thepossible effects of the expected dramatic growth in the world's population, industrial output, useof energy, and demand for food. Under current practices, the result could be appallingenvironmental conditions in both urban and rural areas. The World Development Reportpresents an alternative, albeit more difficult, patl- - one that, if taken, would allow fLturegenerations to witness improved environmental conditions accompanied by rapid economicdevelopment and the virtual eradication of widespread poverty. Choosing this path will requirethat both industrial and developing countries seize the current moment of opportunity to reformpolicies, institutions, and aid programs. A two-fold strategy is required.

* First, take advantage of the positive links between economic efficiency, income growth,and protection of the environment. This calls for accelerating programs for reducing poverty,removing distortions that encourage the economically inefficient and environmentally damaginguse of natural resources, clarifying property rights, expanding programs for education (especiallyfor girls), family planning services, sanitation and clean water, and agricultural extension, creditand research.

* Second, break the negative links between economic activity and the environment.Certain targeted measures, described in the Report, can bring dramatic improvements inenvironmental quality at modest .ost in investment and economic efficiency. To implement themwill require overcoming the power of vested interests, building strong institutions, improvingknowledge, encouraging participatory decisionmaking, and building a partnership of cooperationbetween industrial and developing countries.Other World Development Report background papers in the Policy Research Working Paperseries include:

Dennis Anderson, "Economic Growth and the Environment"

Dennis Anderson and William Cavendish, "Efficiency and Substitution in Pollution Abatement:Simulation Studies in Three Sectors"

William Ascher, "Coping with the Disappointing Rates of Return of Development Projects withEnvironmental Aspects"

Edward B. Barbier and Joanne C. Burgess, "Agricultural Pricing and EnvironmentalDegradation"

Robin W. Bates and Edwin A. Moore, "Commercial Energy Efficiency and the Environment"

Wilfred Beckerman, "Economic Development and the Environment: Conflict orComplementarity?"

Richard E. Bilsborrow, "Rural Poverty, Migration, and the Environment in DevelopingCountries: Three Case Studies"

Charles R. Blitzer, R.S. Eckaus, Supriya Lahiri, and Alexander Meeraus,(a) "Growth and Welfare Losses from Carbon Emission Restrictions: A GeneralEquilibrium Analysis for Egypt";(b) "The Effects of Restrictions of Carbon Dixide and Methane Emissions on the IndianEconomy"

Judith M. Dean, "Trade and the Environment: A Survey of the Literature"

Behrouz Guerami, "Prospects for Coal and Clean Coal Technology"

David 0. Hall, "Biomass"

Ravi Kanbur, "Heterogeneity, Distribution and Cooperation in Common Property ResourceManagement"

Arik Levinson and Sudhir Shetty, "Efficient Environment Regulation: Case Studies of Urban AirPollution"

Robert E.B. Lucas, David Wheeler, and Hemamala Hettige, "Economic Development,Environmental Regulation and the International Migration of Toxic Industrial Pollution:1960-1988"

Robert E.B. Lucas, "Toxic Releases by Manufacturing: World Patterns and Trade Policies"

Ashoka Mody and Robert Evenson, "Innovation and Diffusion of Environmentally ResponsiveTechnologies"

David Pearce, "Economic Valuation and the Natural World"

Nemat Shafik and Sushenjit Bandyopadhyay, "Economic Growth and Environmental Quality:Time Series and Cross-Country Evidence"

Anwar Shah and Bjorn Larsen,(a) "Carbon Taxes, the Greenhouse Effect, and Developing Countries";(b) "World Energy Subsidies and Global Carbon Emissions"

Margaret E. Slade,(a) "Environmental Costs of Natural Resource Commodities: Magnitude andIncidence';(b) "Do Markets Underprice Natural Resouce Commodities?"

Piritta Sorsa, "The Environment - A New Challenge to GAIT?"

Sheila Webb and Associates, "Waterborne Diseases in Peru"

Background papers in the World Bank's Discussion Paper series include:

Shelton H. Davis, "Indigenous Views of Land and the Environment"

John B. Homer, "Natural Gas in Developing Countries: Evaluating the Benefits to theEnvironment"

Stephen Mink, "Poverty, Population and the Environment"

Theodore Panayotou, "Policy Options for Controlling Urban and Industrial Pollution"

Other (unpublished) papers in the series are available direct from the World Development ReportOffice, room I7-101, extension 31393. For a complete list of titles, consult pages 182-3 of theWorld Development Report. The World Development Report was prepared by a team led byAndrew Steer; the background papers were edited by Will Wade-Gery.

Table of Contents

I. Introduction . .................................................... I

1. Scarcity and Choice ........................................... 12. Choice and Value ............................................ 13. Projects, Programs, Policies ...................................... 24. Whose Values Count? .......................................... 35. Valuation and the Developing World ................................ 5

II. Economic Valuation: What Is It? ....................................... 6

III. Total Economic Value . ............................................. 7

IV. Why Derive Economic Values? ....................................... 12

1. The Importance of Environment in National Development Strategies .... ........ 12A. Mali: Soil ErosionB. Burkina Faso: Biomass

2. Modifying the National Accounts .................................. 143. Setting National and Sectoral Priorities .............................. 174. Project, Program and Policy Evaluation ............................. 195. Valuation and Sustainable Development .............................. 24

V. Valuation and Discounting .......................................... 27

1. Sustainability Criteria . ......................................... 282. Modifying the Discount Rate .................................... 29

VI. Valuation in Practice . ............................................. 32

1. Setting Priorities . ........................................... 322. Identifying Willingness to Pay for Conserving Environmental Assets .... ........ 343. Compara.tive Economics of Environmental Conservation ................... 46

VI. Valuation and Global Environmental Problems ............................. 47

VIII. Conclusions . ................................................. 48

Annex I Environmental Policy as a Constraint on Economic Growth ................. 50Bibliography . .................................................... 59

I. Introduction

l. Scarcity and Choice

If the Earth's resources were available in infinite quantities, and deployable at zero cost, therewould be no economic problem. Everyone could have everything they wanted without compromisinglater generations' wants and needs. It would not be necessary to choose. Choice becomes a necessityonce it is recognized that resources are finite in terms of their absolute quantity, and involve costs ofextraction and use.

The oceans, for example, have a finite capacity to assimilate waste before the process ofeutrophication sets in. Going beyond that capacity means that the further benefits of disposing of wasteto the ocean have to be weighed against the costs associated withl eutrophication - e.g. the loss of fishstocks. This kind of resource constraint is an instance of Malthusian scarcity - after the ReverendThomas Malthus. The limit can be exceeded, but only at a cost.

The other main form of scarcity is Ricardian scarcity - after David Ricardo. Absolute limits arenot breached, but the cost of harvesting, extracting and using a resource rises. The global atmospheremight be an instance of a scarce resource in the Ricardian sense. As its capacity to receive andaccommodate gaseous wastes from fossil fuel combustion, land conversion and chlorofluorocarbons(CFCs) is exceeded, so the surface temperature of the Earth may warm up with (disputed) deleteriouseffects to human wellbeing. The "price" of using the atmosphere as a waste sink is effectively risingthrough time as greater and giteater demands are put on it.

Given that resources are scarce in relation to human demands upon them, choices or "trade-offs"have to be made. In the market place the individual has fairly clear information on which to base anychoice; the product tends to be visible, its characteristics are generally well known, and it has a marketprice. The individual's choice is then based on a weighing up of the quantity, quality and price on offer,subject to some uncertainty arising from incomplete information. But when envir( qmental assets andservices are involved there is often very limited information about the nature of the product in question,and, invariably, there is no price posted in the market place. In the case of global warming, forexample, there is extensive uncertainty over its likely impacts; hence there is only limited informationabout the environmental benefits of controlling global warming - the "product" or "good" in this case isthe damage avoided by undertaking control measures. Moreover, the global atmosphere is not boughtor sold in the market place; it has no perceived "price." An additional complication is that manyenvironmental goods are "public," rather than "private," goods. Public goods generally have thecharacteristics of joint consumption and non-exclusion; when the good is consumed by one person thisdoes not, and cannot, diminish the amount consumed by another person. A's consumption of clean airdoes not diminish B's consumption, nor is there a way in which it could, since A cannot prevent("exclude") B from consuming the resource. This "publicness" is one reason why markets forenvironmental goods and services often do not develop naturally.

2. Choice and Value

Making choices in the context of environmental quality, therefore, is more complex than makingchoices in the context of purely private goods and services. What is being compared is one priced good(the private good) and one unpriced one (the public good) - as when deciding to invest in air pollutioncontrol rather than new output capacity. Alternatively, the comparison may be between two or moreunpriced public goods - air quality versus water quality, for example. In this context it is necessary toimpute a value to the environmental good or service. The discipline of environmental economics hasdeveloped techniques whereby such values can be imputed. In the

1

I 1COWBOY AND SPACESHIP ECONOMIES

EconM1t Krenneth Bouldiao eoined the phases *eowboyeconomy' and 'spaceship e¢onomy' t characterize the 20th ceniutytansidon in human perception of the natural environmont. The cowboy symbolizes man's view of the natural envi.onmtent asa new domain, a roOier, to be conquerod and civilized, The cowboy economy is an open system which Is maintained byresoure and anergy inputs which then become w'aktes, or outputs of the system. This contasts starkly with tho economy as aclosed system, in which inputs ao, as :ar as poasible, transformed into outputs wbich are then returned to the system throughreycling and reuse, As mankind peroelvea the 'Iinits' of economic activity in tertms its environmmental effets, so econotucactivity shoulth be reorganized to increase recycling and reuse of materials, and to substitute unlimited energy flowsbased onsolai enegy for tho embodied solar energy of fossil fuels.

Boukling's vision has done much to influence the nature of environmental thinking. In its most provoking sense it can be takento imply that thoe throughput" of the economy Is not something to be ntaxinized, but something to be mninimized. What rusttersis not throughput (the economic analogue of which is GNP) but the stock of wealth, including the toek of knowledge andhum,a wellbeing and the stock of environmental assets. The idea that it is this dtock that needs to be maintained and expandedunderlies a good paitt ofmoder thinig about 'sustainablc development". However, the idea of conceatadng on stocks ratherthan fows may be justified for a rather different reon. It is not necersay to accept thO view that stocks s than .flowsdetermine wellbeing. Thw ock of wealh determines the canabilitv to generate real income. Ifreal income isawbat creates mosthuman wellbeing a and in the poorer worid it is difricult to see it otherwise - then incteasing the oapability to secute real incomeinvolves increasing the stock of wealth; This iS consistent with the World Commission on Environment and Development's viewof sustainable development (s= the text).

K.Boulding, 'Mre Economics of the Coming Spacehiip Earth', in H.Jarrett (ad), EnviroUnmnts! Quait in a 3rowtna EconoJohw Hopkins University Prss, Baltimore, 1966.

market place, individuals exercise choice by comparing their willingness-to-pay with the price of theproduct; they purchase the good when their willingness-to-pay exceeds the price, and not otherwise.Imputing values thus requires finding some measure of willingness-to-pay for environmental quality. Thisis the essence of the process of economic valuation: it involves finding a willingness-to-pay measure incircumstances where markets fail to reveal that information.

This kind of "market failure" affects the allocation of resources within an economy. If theproduction of specific crops involves using agricultural technologies which give rise to soil erosion, thenthe damage done by the soil erosion may well not be reflected in the choice of crop or technology. Thismay be so even where the costs of damage are borne by the farmer growing the crops, since futuredamage to crop productivity through soil erosion may be imperfectly reflected in present choices. Marketfailure is even more pronounced when the costs are borne by agents other than the farmer - possibleexamples include the siltation of rivers, ports and reservoirs. Failure to account for these external costsgives rise to a misallocation of resources in the economy, in this case through the choice of the wrongagricultural technology. Avoiding this misallocation of resources involves understanding the value of theexternal costs, and then finding a mechanism for integrating those values back into the original selectionof a technology. Valuation may be imperfect but, invariably, some valuation is better than none.

3. Projects, Programs. Policies

The putpose of economic valuation is to reveal the true costs of using up scarce environmentalresources; choosing "instruments" is the mechanism whereby the resulting values are reflected indecision-making. If the disposal of sewage to inland waters, for example, gives rise to a loss ofwellbeing, then the value of that loss should be reflected in the private costs of disposing of the sewage.This might be achieved by taxing the sewage discharger, by setting some environmental standard for theeffluent or the receiving waters, or by requiri,ng the discharger to buy permits for the effluent.

2

In general, the choice of instrument - tax, standard or permit - will not be affected by the valueof the damage done. The virtues of economic instruments - taxes, permits and other incentive systemsbased on altering market signals - remain even if valuation is not carried out. But valuation is essentialif the ca_le of the tax or strength of the regulation is to be determined. In practice, valuation is theexception anu not the rule. Environ,nental standards are often set by criteria that incorporate somefeatures of the valuation process; health criteria, for example, determine many enu/ironmental standardsin the developed world. Damage to human liealth should be an integral part of any valuation process,since people will be willing to pay to avoid health risks from pollution and waste. But as there are oftenmany other forms of damage 't,esides health effects, using health criteria alone could impose its owndistortions on resource allocation. In many other cases, environmental standards are set without any clearor detailed rationale. Many regulations, for example, are responses to environmental scare stories andmisinformed perceptions of hazard and risk. In such circumstances, economic valuation is helpful as acheck on the criteria implicitly being used.

Valuation is relevant at all levels of public choice:

* in proiect appraisal the environmental impacts of any investment need to be estimated andcompared to the other costs and benefits;* in program appraisal the value of environmental impacts similarly need to be integrated intothe evaluation process;* in policy appraisal enviromnental factors need to be treated on an equal footing with other costsand benefits so that sectoral priorities are not distorted. This is as important in choosing betweenmarginal expenditures on, say, transport as against energy, as it is in choosing betweenconservation and development projects. Similarly, as discussed above, the setting ofenvironmental standards should be informed by valuation analysis. In short, environmentalvaluation should be an integral part of

- sectoral priorities;- the balance between conservation and development;- the choice of environmental standards.

4. Whose Values Count? Intergenerational and Intragenerational Incidence of Costs and Benefits

Economic values reflect individuals' willingness-to-pay for benefits or their willingness-to-payto avoid costs. Typically, the values that count belong to those actually exercising the choice: the currentgeneration. But a particular feature of environmental costs and benefits is that they often accrue to peoplein generations yet to come. How are their values to be counted? Counting only the current generation'spreferences biases the choice against future generations unless there is some built-in mechanism to ensurethat current generations choose on behalf of future generations and take their interests into account. Thisbias arises because future generations are not present to have their votes counted; this is the problem ofintergenerational incidence. Whether they are present or not, future gains and losses tend to be playeddown in economic decision-making because of the practice of discounting the future.

An analogous form of bias arises even within a generation: willingness-to-pay is weighted by theincomes of those expressing their willingness to pay. The economic votes of the poor count for less inthe market place than the economic votes of the rich. This is the problem of intragenerational incidence.

Both inter- and intra-generational bias are present in the willingness-to-pay criterion for

3

JqX COUNTING ON FVTUR VALUES

On the approach of the World Commnisilon oa Etwitonmont and Development ( 04r Cpmllon utur0, sustainable developmnentimplies soma general rule about not impairing the capability of future gonerations to achieve the same level of wellbeing as thecurrent generation. But this is in faot a particulat ethical rle for treating future generations. lTere are others. .noosing betweenmlea Is for from straIghtforward. Yet which rule is chosen will have potentially major resource allocation implications.Philosophers -and economists have analyzed the isues in detail. In broadest outline the alternative views might be summarizedas follows.

Teleology

Teleoiogy involves weighing up goods and bad* and aims to maximize what is good. Goods and bads are broadly consred.Equality might be good, so that maximizing equality would be a teleological approach. Maximizing the economist's notion of'tiulity' (preference satisfacton) would be a particular form of teleology - utUiitargnism. The essence of toelology is that itpermits a balancing of goods and bads or of one good against another - equality against utility, for axample. 1he benefit-costapproach is teleologicat, being a form of tfilitarianisni based on preference satisfaction as a 'good thing',

On the teleological approach it would be consistent to adopt a polic' that made futume generations worse off compared topresont genorationa if the gains to the present are deemed to be greater than the costs to the future. Teleology Is not thereforeconsistent with the broad definition of sustainable development entertained by the Brundtland Commission.

Theoe ame sevetal theoies of justioe. Some have been applied to the issue of how to sccOunt for the f tegeroeationaldistnrbution of goods and badsa

-Conteaealism. Contractualis atgue tht people will come together to deteramine Mules of social behavior becausait-is to their mutual advantage to do so. Laws and their implementation exist for this reason. This doctrine mu::tual,,advaA&e will only arise in socialt contexts where the patties to the 'contract are of mughly equal power. Otherwisethe powerful would not secuxe any advantage from an agreement and they Would not allow one to eerge> But fure

: generations not yet born have no power at all, so the requirement of roughly equal power is not met. On the.conttatatln sppeoach,the, thete appears to be no basis for a theory of inteneional justce.Eventfthee" were,it is inconsient .with tleololg justice would take precedence over the good.

-IWO. the rights appsrosh, justice Implies a duty to behave in a curtain way,nd coners a night onA ho thpo-whoisa the subject of the duty.to expect that behavior. The. rights approach is also inconsistent with tldeologyfbcause::whiat is,right takes preedin over what is'good. The rights are. like constrats which mius ae mnet first bcfore anyother rule of behavior is applied. The notion of 'maintaining wealth 'see Box. '. 1) fits this approach siice it ispredicated on the 'view that fiiture geneations have a right to at least the same leVel of wellbeing as'cuurrentgeneratlons. But there :are pvroblems of definition since it is not clear, who holds tho righlts. Obeying tbe rule g:-.would arguably zaltrthe p,attern of resour allocation which, in -turn, would alter the behavior of individuals evento the point 'of alterg their decisions about firntly size or timing. The future' population arsing'fom the " ruIesituation 'mt differ from the population arising from the "without rule" situation. 'Thte Is tino oie in particular to..whom the rights belong. This is the w-called "nonidentity problemt.

Resoures. On this approach each generation should have the same level of resources or productive capacity aF eachother. lheir wllbeing mfay then differ, depending on what each generation makes of this stock of resources. But theirpa2abiliw to genefrae wellbeing would be the same. Alternatively, since wellbeing depends on consumption an i totproductive capaeity, what matters is productive capacity less savings-

Strict Egalitarianims. Strict egalitariais insist on equality of somne characteristic for each generation. It might beresoure- endowments (as with resourcismn), or wellbeing itself. Or the rigidity May relite to the wellbeing of a targetgroupf say iom poorest group in society. No change would be permitted if the wellbing of this poorest gtoup wasreduced, regardless of gains to other groups (John Rawls's "difference principle"). Clearly, such approaches ameinconsigsent with the teleological view since none of them allows gains' and losses to be weighed up indepetfently ofto whom they atome.

Philosophical discussion nmy appear out of place in a discussion relating to world development. But it is fundamental to the way

4

in which environMnma: isules are treated in economic appraisal. First, the conventional economic appro3ch ia based on abenefit-cost framework. Second, envimutmental problems often involve long-lasting or distant inpacts. Applying the benofit-costframework to environmental Issues thetefore poses a potential intergenerational impact prblem. Rosource allocation decisionswill differ, pediaps markedly, according to the rule adopted for the treatment of future generations. The benefit-cost approachwould permit gains and losses to be balanced between generations. Rules based on justice will produce different resourceallocatiosm sinco they invariably do not permit the tradeoff betwoon generations to take place.

The definitions used here have becn borrowed from J.8Dome, Tbo _nterenoratlotnl Asnca of Cglimag '9innng, Departmentof Economics, University of Bristol, UK, June 1991,m.

eliciting economic values. Both biases are strongly debated by economists; yet their significance maybe overstated, for two reasons:

(a) Generations overlap. The current population includes three generations: parents, children andgrandchildren. Parents care for their children and grandchildren and make sacrifices for them.Current children care for their children and will care for their grandchildren. In formal language,the rate at which current parents discount the future is likely to incorporate a "coefficient ofconcern" for the future through the direct effects of childrens' wellbeing on parents' wellbeing.But whether these concerns are consistent with the kinds of discount rates used in practice (often10 per cent in real terms, or more) is open to serious question;

(b) Redesigning projects and programs to allow for distributional fairness within a generation maybe an inefficient way of serving the goal of fairness. It is often preferable to secure the gain tooverall development by concentrating on efficiency gains and losses, and then correcting thedistributional impacts in some other way (e.g through lump sum transfers). Moreover, integratingdistributional concerns into project and program appraisal has been tried (Squire and van der Tak,1976), but is not widely practiced because of operational and informational difficulties. Care hasto be taken not to use this argument to ignore distribution in the appraisal process altogether.More seriously, the same rationale for ignoring distributional considerations cannot be advancedso firmly in the context of jlicy choice.

There is no consensus on how to integrate inter- and intra-generational considerations intoeconomic decision-making about the environment. While economists would typically favor the use ofpositive rates for discounting the future, some argue that there is no particular rationale for discountingfuture wellbeing. Most economists would probably focus on efficiency gains and losses in project andprogram appraisal, but others favor the explicit recognition of multiple social goals or 'multi-criteria' andseek some form of calculus for trading-off between such goals when they conflict.

5. Valuation and the Developing World

Valuation is fi"damental to the notion of sustainabledevelopment, which has been loosely definedby the World Comm.-iJon on Environment and Development as development that "meets the needs ofthe present without compromising the ability of future generations to meet their own needs." This isbecause, development paths which ignore the environmental consequences of economic change may wellbe unsustainable. As environments deteriorate, so human health will suffer from environmentally-induceddiseases, and long-term labor productivity may decline. Degraded environments also impose costs interms of: foregone crop output due to soil erosion; additional energy imports, as biomass energy is

5

exhausted; diverted labor time to collect water and fuelwood from more and more distant sources, andso on. Moreover, when properly valued, investment in natural resource augmentation is often found toyield rates of economic return comparable to that earned on conventional capital investments.

Demonstrating that "conservation pays" in terms of economic development, is a process onlyrecently b.3gun. But it is already possible to point to significant findings. Far from environmental andresource conservation being inimical to sustained economic development, it is in a great many casesintegral to the development process.

11. Economic Valuation: What Is It 9

It is important to understand what is being done when economic valuation is carried out.Economic values are measured by the summation of many individuals' willingness-to-pay for a particulargood. In turn, willingness-to-pay (WTP) reflects individuals' preferences for the good in question. Thus,in an environmental context, economic valuation is about "measuring the preferences" of people for anenvironmental good or against an environmental bad. Valuation is therefore of preferences held bypeople. The valuation process is anthropomorphic. The resulting valuations are in money terms becauseof the way in which preference revelation is sought - i.e. by asking what people are willing to pay, orby inferring their WTP through other means. Moreover, the use of money as a measuring scale permitscomparison that is required of "environmental values" and "development values." The latter areexpressed in money terms, either as a dollar amount or an economic rate of return. Using other unitsto measure enviromnental values would not permit the necessary comparison with development values.

The language of economic valuation is often misleading; studies speak of "valuing theenvironment" or "pricing the environment." Similarly, because changes in the environment affect health,it is necessary to define valuations of changes in health status, the ultimate change, of course, being thecessation of life itself; hence, references to "the value of life." All these terminologies generate anunfortunate image of economic valuation.

But in practice, what is being valued is not "the environment" or "life", but people's preferencesfor (and against) changes in the state of their environment, and their preferences fc (and against) changesin the level of risk to their lives. There is no dispute that people have preferences for and againstenvironmental change. Likewise, there is no dispute that people are willing to pay to prevent or securechange: donations to conservation societies alone demonstrate this. The problem arises when this WTPis taken as the value of environmental change. Many people believe that environmental assets areintrinsically valuable: they are of value in themselves, not merely because individual human beings havepreferences for them. Yet, there is no reason to reject the idea of intrinsic values because the idea ofmeasuring preferences is adopted. What is being assessed are two different things: the value of people'spreferences for or against environmental change (economic value), and the value that intrinsically residesin environmental assets (intrinsic value). Economic valuation is essentially abat discovering the demandcurv for environmental goods and services.

Once it is accepted that both forms of value exist, the issue becomes one of which values shouldinform and guide the process of making public choices. The answer is that, since both values are"legitimate', both are relevant to decision-making. Making decisions on the basis of economic valuesalone neither describes real world decision-making, nor would it be appropriate given that governmentsand other agents involved in the development process have multiple goals. One difference between theeconomic and intrinsic value approaches is that while economic values can, in principle, be measured,intrinsic values cannot. If decision-makers do not feel the need for quantified assessments of gains andlosses, then lack of quantification may not be an obstacle to decision-making. Otherwise, it will oftenprove difficult to make choices between competing projects or alternative policies with differing

6

environmental impacts.The practical problem with economic valuation is one of deriving credible estimates of value in

contexts where there are either no apparent markets or very imperfect ones. If it is possible to derivesuch values, then it may well be that some measures of individuals' preferences will, in any event,capture at least part of what might be called intrinsic value. This will be the case, if those peopleexpressing values for the environmental change in question themselves possess some concept of intrinsi.value; they may then be partly valuing "on behalf' of the environment as an entity in itself. Althoughthese kinds of issues may seem remote from tMe concerns of the deyelopment process, they are not.Many of the environmental assets that people generally feel are very important are in the developingworld; notable examples include tropical rain forests, ecologically precious wetlands, and many of theworld's endangered species. Many people feel these environmental assets have intrinsic value. niey mayexpress that view by speaking of the immorality of activities which degrade these resources, and of the"rights" to existence of trees and animal species.

Yet, at a more practical level, the "development and environment" debate often centers on thehigh relative value of development in a context of malnourishment and underemployment. Theenvironment tends to be viewed as a luxury to be afforded later, not now while the struggle fordevelopment is under way. In such contexts, it may prove counterproductive to introduce notions ofrights and intrinsic values into policy dialogue; honouring such notions may be perceived as a sacrificeof developmental benefits. If, on the other hand, conservation and the sustainable use of resources canbe shown to be of economic value, then the dialogue of developer and conservationist may change -environmental and development benefits may come to be seen,hot as necessary opposites, but as potentialcomplements of each other.

For this to happen, ways must be found for the developing world to captur conservation benefits.If environmentalists in rich countries perceive value in conserving a rain forest in a poor country, thisis of little consequence to the poor country unless there is a potential cash flow or technology transferto be obtained. Economic valuation is therefore a two-part process in which it is necessary to:

edemonstrate and measure the economic value of environmental assets;* find ways to capture the value.

III. Total Economic Vaie

The economic value of environmental assets can be broken down into a set of component parts.Decision-making aboat alternative land uses for a tropical forest, can provide a useful illustration.According to benivfit-cost rules, decisions to "develop" a tropical forest can be justified by showing thatthe net benefits from development exceed the net benefits from "conservation." Development, in thiscontext, is taken to mean some use of the forest that would be inconsistent with retention of the forestin - at the minimum - some approximation of its natural state. Conservation, here, can have twodimensions: preservation, formally equivalent to outright non-use of the resource; and conservation whichwould involve limited uses of the forest consistent with its retention. These definitions are necessarilyimprecise. Some people would argue, for example, that "ecotourism" is not consistent with sustainableconservation, others that it may be. Accepting the lack of precise lines of differentiation, the benefit-costrule argues for development only if net development benefits (benefits minus costs) are greater than netconservation benefits. Put another way, development benefits minus both development costs and netconservation benefits, must be positive.

Typically, development benefits and costs can be fairly readily calculated because of attendantcash flows. Timber production, for example, tends to be for commercial markets and thus prices are

7

directly observable. Conservation benefits, on the other hand, are a mix of associated cash flows and"non-market" benefits. This results in two biases. The first is that the components with associated cashflows are made to appear more "real" than those without such cash flows; decisions are likely to bebiased in favor of the development option because conservation benefits are not readily calculable. Thesecond bias follows from the first. Unless incentives can internalize non-market benefits into the land usechoice mechanism, conservation benefits will automatically be downgraded. Those who stand to gainfrom, for example, timber extraction or agricultural clearance will not be consumers of non-marketedbenefits, and the resultant "asymmetry of values" imparts a considerable bias in favor of the developmentoption.

Conservation benefits are measured by the total economic value of the tropical forest. TotalEconomic Value (TEV) comprises several types of value. These are explained in greater detail in Box3. Briefly, however, four main types of value should be identified:

Direct use values are conceptually fairly straightforward, but are not necessarily easily measured ineconomic terms. Thus minor forest products' output (nuts, rattan, latex etc.) should be measurable frommarket and survey data, but the value of medicinal plants is more difficult to measure.

Indirect use values correspond to the ecologist's concept of "ecological functions." A tropical forestmight help protect watersheds, for example, and removing forest cover could therefore result in waterpollution and siltation. Similarly, tropical forests "store" carbon-dioxide. When they are burned forclearance much of the stored CO2 is released into the atmosphere, thus contributing to greenhouse gasatmospheric warming. Tropical forests also store many species which in turn may have ecologicalfunctions - one aspect of the value of biological diversity.

Option values relate to the amount that individuals would be willing to pay to conserve a tropical forestfor future use. That is, no use is made of it now but use may be made of it in the future. Option valueis thus like an insurance premium to ensure the supply of something the availability of which wouldotherwise be uncertain. While there can be no presumption that option value is positive it is likely to beso in the context where the resource is in demand for its environmental qualities and its supply isthreatened by deforestation.

Existence value are those valuations of an environmental asset that are unrelated to either current oroptional use. Its intuitive basis is easy to understand because a great many people reveal tneir willingnessto pay for the existence of environmental assets - through donations to wildlife and other environmentalcharities - without taking part in the direct use of wildlife through recreation. To some extent, thiswillingness to pay may represent "vicarious" consumption, i.e. consumption of wildlife videos and TVprograms, but studies suggest this is only a weak explanation for existence value. Empirical measuresof existence value, obtained through questionnaire approaches (the contingent valuation method), suggestit can be a substantial component of TEV; this finding is even more pronounced where the asset is unique(see Box 4), suggesting high potential existence values for tropical forests, and in particular, for luxuriantmoist forests.

8

TOTALECONOMIC VAJ1JEIN THE TROPICALFOREST CONTEXT

lotat Economic Value =

_Use Value ____on-Use Value(1) - (2) (3) (4)

DTrect +Indirect Value +Option Value +Existence Value

Sustainable Timber

Non-Tnmber Products Nutrient Cycling Future Uses as Forests as objects ofper (1) + (2) Intrinsic value, as a

gift to others, asRecreation Watershed Protection responsibility (stew-

ardship)Medicine Air Poltution Reduction

Inctudes cuttural andPlant Genetics Microclimate heritage vatues

Education

Humuan Habi tat

Total economic value coimpries use and existence values. Use value comprises direct uses (e.g. timber production), indirect uses(e.g. the protective effects of forests on watersheds), and 'option' values -aldn to an insurance payment to reflect the value ofa future use if the option to use "the resource is exercised. Existence values comprise willingness-to-pay for an environmentalasset's oonservation even though no use value is present.

*onom!c; Valzs and Topical Forest Fun)tions: the Korup National Park

KOMp Ntional Park lies in Southwest Province, Canrun. It contains Africa's oldest rainforest, over 60 miltion years old, WMhigh speciesendemism.There are over 1000 species of plant, and 1300 animal species including 119 mammals and 15 primates.Out of h toW listed species; 60 occur nowhere else and 170 are currently listed as endangered. Con'tinued land use chaAgesare putting substantial pressure on the rainforest. The Worldwide Fund for Naturc CWWF) inid a program of conservation,centered on a management area of 126,000 hectares plus a ourrounding buffer wund of 300,000 hectares. A similar programwas initiated for Oban nationat Park just across the border in igeria (see nap), -

Econotnic valuation of the rainforest's benefits was carried out in order to assist with the proceas of raising development aidfinds to consetve Vhe faea. Benefitg of conscwvatio were then compared to the cost4 of th conaservation -project plus'theforegone timber revenues. While the- framework for analysis was the total economic value concept, oxisence and option valueswere not directly estimated. The procedure involved estima'ting direct and indirect use values to the Cameroun and then seeingwhit the existence and option valuowould have to be in order to justify the project. Since it was thought that the non-use valueswould maiinly reside with people outside the Cameroun, the focus of attention for non-use values was on seeing whatInternational transfers tight be needed. Briefly summarized, the results were as shown in the table below.

Prom tei standpoint of the Cameroun, the project appears not be worthwhile because there is a negative net present value ofsooi 18$2 miliott CFA at 5% discount rate, although there is a modest positive net present value if the discount rate is loweredto 6%. But the analysis covers only some of the components of total econotnic value. What of existence and option values?These were not estimated directly. Istead, the isuue therefore beconmes one of asking whether the rest of the world would bewilling to pay 1852 million CFA (in present value terms) to the Cameroun to reflect these option and existence valucs, One wayof testing this is to look at existing conservation transfer through debt-for-nature swaps. Translated into a per hectare badis,the tequired trensfeil for the Cameow is just over 1000 ECUs per km2. Debt-for-nature swaps have implied various valuationsranging from as low as 15 ECU per knl (Bolivia) to around 1600 ECUs per km2 (Costa Rica), Given the high species mndemismand diversity of Korup, values of t000 ECUs or rmore would seem justified. The conservation of Komip forest becomes justiftedin economic terma provided this transfer actually occurs.

9

Box 3 (cont.)

Senefits and Costs to the Caneoroun (Present vatues, MilLion CFA, 1989 prices)

(Discount Rate a 8%)

Costs of conservation protect:

Resource cost8: - 4475Foregone forest benefits

tintber: - 353forest products - 223

-5051

Benefits of Conservtion Project:

Direct Use Benefits

Use of forest products + 354Tourism + 680

.Indirect Use Benefits

Protection of Fisheries + 1770ftood control + 265

*Soil productivity + 130

+ 3199

Net enef its to Cameroun - 1852

Economic Rate of Return 6.2X

Ilet Benefits to-Cameroun ifthe discount rate is 6% + 319

The resource cos are based on budgets and plans in the Korup Nat;onal Park Master Plow, net of compensation payments(Wich age internal tratnsfrs) and other costb regarded as being not auributable to the consetvation project. The florgone forestbenefits .cludes timber from potential comMfrcial lo8ging (the #%# nmillion CPA) and sone foregone traditional uses oftheforeat, na mnlhunfing, that would be fojiidden witin a designated national park, and which cannot be offset by,diverting idtivityelsewhere (the 223 million CFA). This proscrtiption of traditional uses affeots some 800 villagers within the national parkboundaries. In the long run, however, other residenta, mainly some 12,000 people on the peiphery willbe able to continue theirtradWional use of the ftrest, which they would Pot be able to do if deforestation continued. Thus, while one grwp loses beneftsanther, larger, group gaius (the 354 million CFA). The tourism figure is conjectural and is baed on an eventual 1000 visitorsp.a by the year 2000 ind their expected expenditure adjusted for the shadow wage rate. 'Te fisheries item is impotlant. Rainfallia the fotest feeds several rivers which feed into large mangrove areas tich ia fish. The mangroVes ptosper on the basis offrehwater inundation in high water periods and saltwater in low water periods. If the forest was to disappear, peak flows fiomthe forest would increase and thete would be added sediment and less salinity. Basically, the tangrove swamps would no longerfinction as the habitat for the rich fish species that nake up both the on and offasore fWuetiee. Since the link between thorainforest and the offisore fishery is less establishod than the link to the inshore fishery, ondy damage to the onshore fisherywas estimated. WIhis was valued at the market value of fish and, as a check, at the income derived frort the fishery.

The flood alleviation benefits were calculated by looking at the expected value of the income losses that would accrue if therewas a flood, The soil fekrilty benefits were based on a broad brush assessment that, if the forest disappeared, cash crop yieldswould decline by 10%.

Tle implicit miniimum requirement for an international transfer (the so-called "rainforest supply price') was estimated by takingthe present value ofnet costs (the 1852 million CFA) and dividing by tha present value ofthe hectarage that could be identifiedas being protected by the conservation project - some 500,000 'hectare years'.T1is produces the value of 3600 CFA pet hectareper year, or some 1060 ecus/kmn.

10

Box 3 (cent.) Notable omissions from the study arn twofold: no attempt was made to assess the value of thd forest to localpeople over and above its use value; and no attempt was made to estimate the net contribution to C02 emissions ftomdafostatoio Both oiss;ions are likelyto reduce the net present valte deficit shown in the table. But only the former will lowerthe rainforest supply price because COZ benefits are likely to attract a negligible If not zero willingness to pay on the part ofCanmtrowz cize, The C02 benefits will, however, make it more likely that the rest of the world will pay for rainfotestconservation (i.e. it affects the rainforest demand price).

Shurces: J.4RitenbZeek, MTe Rainibrest Supply Pric:- a Step Towards Estimating a Coat Cve for Rainfores ConsetvatIon',SuntoryToyota Internationl Centr for Economics and Related Disciplines, Paper 29, London School of Economnics, September190: L.Ruitenbeekl Evaluating Economnic Policies for Pronmoting Rainforest Conservation in Dewtoinst Countries, Ph.D theila,Lonon School of Economics, 1990; J.Ruitenbeek, Economic Analysis of Tropical Forcst CoQtsrvation ,litiatives: xM&lefrom West Afric,. World Wide Fund for Nature, Godalming, 1990.

Total economic value can be expressed as:

TEV = Direct Use Value + Indirect Use Value + Option Value + Existence Value

While the components of TEV are additive, in practice care must be taken not to add competing values.There are trade-offs between different types of use value, and between direct and indirect use values.For example, the value of timber from clear felling cannot be added to the value of minor forest products,but timber from selective cutting will generally be additive to forest products.

BOX 4 - VALUINGPREFERENCES FOR UNIQlE ASSETS: VISIIBILITY AND THE GRAND CANYON

Calculating existence valuo is likely to be important in contexts where (a) many people are faniiar with toe attributes of thesse to be valued, and @f) the aset i unique. Some evidence to support this view can-be found in an analysis of valiationsf torimproved visibility in the Grand Canyon region. By using surveys to assess both users and non-users' willingness to pay forimproved visibility, one study found that user values were some 7 US cents per monts, whilst existence values were $4.43 permonth (I98) prices), over 60 ties higher, Signtficantly, distance from the site did not affect presetvatio values, a -ict that theresearchers put down to:the unique nature of the Grand Canyon, a 'wonder of the world'. Since distance was not rlevant tothepreservation bids, it is legitinate to extrapolate the nean preservation bids to the nation as a whole.The ratio of 60+ Isn-tich higher than othet studies have found between total values and use values. But i arises partly fom the uniqueness of theasset and pastly because two different questions are being asked. The user value question asked how much users would be willingto' ay th,ro entrace eharge increases. whereas the total value question related to monthly electricity bill increase.

Respondents were shown photographs ofthe Grand Canyon region with each photograph revealing different degrees of visibility.Pereeol9los of visibility could of course differ fom some scientific mcasure, so tests were carried out which suggested a linearelationshp between perceived visibility (on a scale of I to 10) and apparent target contrast measured by a niultiwavetelradiometer. Respondents were asked one of two questions: how much would you be willing-to-pay for improved viibtlity,withthe 'vehicle'of payment being hypothetical 'additions to the existing entrance fee. Other respondents were asked how much theywould be willing to pay to presetve visibility if the vehicle was increases in the monthly electricity bill. The first group shouldtherefore ptovide uset values. The second group would provide a 'total preservation bid', i.e. user plus existence values. Ifexistence values 'exissthen the latter vahlions should be greater than the former. This was the finding. By showing how bidswere related to Income, age, and distance from the Grand Canyon. the WTP estimates could be extended to the nation as awhole and compared to the costs of controing air pollution. The annualized preservation benefits for the nation as a whoecame to $7.4 bilfion (1980 dollars) and the costs of control came to S2.8-3.1 billion in annualized form. Hence the costs ofcontrol were outweighed by the benefits of control by a factor of about 3.

W.Schulzeet al.,'The Economic Benefits of Preserving Visibility in the National Parklands',Natural ReOUrCes Joumal, Vol.23.January 13; and D.Bmokshire, W,Schulze and M.Thayer, 'Some Unusual Aspect of Valuing a Unique Natural Resource",Department of Economies, University of Wyoming, Februasy 1985, mimi.eo.

11

IV. Why Derive Economic Values?

There are at least five major reasons why economic valuation of environmental goods and servicesis important.

1. The Importance of Environment in National Development Strategies

Environmental damage shows up in two ways as a eost to nations. First, it impacts on GNP;GNP is less than it would be if at least some environmental damage were avoided. Second, it generatescosts which are not currently recorded as part of GNP, but which would be if GNP accounts weremodified to reflect comprehensive measures of aggregate well-being rather than economic activity.

With regard to the former cost, some evidence now shows that environmental degradation resultsin appreciable losses of GNP (see Box 5). The kinds of impacts that give rise to such costs include:

* foregone crop output due to soil erosion and air pollution;* foregone forestry output due to air pollution damage, soil contamination and soil erosion;* impairment of human health with consequent lost labor productivity* diversion of resources from high productivity uses to uses such as maintenance of buildingsdamaged by pollution.

The empirical investigation of these losses at a national level is in its infancy. In the case of croplosses, for example, what is required is some measure of change in the overall level of economicsurpluses (consumers' plus producers' surplus), rather than a more straightforward estimate of crop lossvalued at market prices. Two examples follow.

A. Mali: Soil ErosionSoil Erosion is endemic to many developing countries. Soil erodes "naturally" but lack of investmentin conservation, poor extension services, inability to raise credit and insecure land tenure all contributeto poor management of soils. A standard approach to estimating the costs of soil erosion is to estimatesoil loss through the Universal Soil Loss Equation (USLE). The USLE estimates soil loss by relatingit to rainfall erosivity, R; the "erodibility" of soils, K; the slope of land, SL; a "crop factor", C, whichmeasures the ratio of soil loss under a given crop to that from bare soil, and conservation practice, P,(so that "no conservation" is measured as unity). The USLE is then:

Soil Loss = R.K.SL.C.P

The next step is to link soil loss to crop productivity. In a study of soil loss effects in southern Mali,researchers applied the following equation to estimate the impact.

Yield = C"bx

where C is the yield on newly cleared and hence uneroded land, b is a coefficient varying with crop andslope and x is cumulative soil loss. Finally, the resulting yield reductions need to be valued. A crudeapproach is simply to multiply the estimated crop loss by its market price if it is a cash crop. But theimpact of yield changes on farm incomes will generally be more complex than this. For example, yieldreductions would reduce the requirement for weeding and harvesting. The Mali studyallowed for these effects by looking at the total impact on farm budgets with and without erosion (see Box6).

12

VW, VALItJNGOF THE EFFECTS OF DOUBLING CARBON DIOXIDE LEVELS ON WORLD AGRICULTURE

Eonomic Models are-being developed which attempt to measure the likely impcts of global warming on the wodd economy.oni: study shows th effeots of doubling C02 concentrations on world agricultur, using a patial equilibtium world food modelshicb ~snsee ehas1gee in consiommrse and protucera' suplus. Provisional results are shown below.

ounttyf/Reeioi - .nWelfare Charge (Om 1986) As a X of GDP

:XSA +194 0.005(COfldt - -167 0.047

-673 0.022CI.Europie 51 0.010Japat . .*-1209 0.062Australita +66 0.038Clsilne x2882 0.141USSR +658 0.292oraz'i -47 0.017

btorld TotE +1509 0.010

Two cees am of htt iere Fist, tho impacts are genetalty very sanfl when exproesed as a pecentage of national income.Seodnd, some atoas :ain from global warming due to the effects of the warmer climate on crop growth and suitability of land.'The differenes aris bcause of diffemet climatic Irnpaet in different regions: global warmnog will not regult in the sameemperlttc increases throughout Ibe w orlnd pteeipitation willalso change. The notable gainers are China and the USSRMThe ov'erallimpact on the world is very small.at around 0.01% of world G1DP.

Aayses i.sf t,. kind assist in identifying the intrests taat each country would have in a global wanrinagreemrenmL ose who-gain -are o lik'ly t show much inierest in signing an agreement, whereas those who tlo' might. On the odher hand, the studys ' i.siw.:iidi8¢c !inity' in the elaIotshlp' between warming and damage done. Sudden oatastrophes and other ecological

Adocks and siis arA not allowed for. These might not be correlated with the losinS'nations shown above. i.e. some of theexpectel gins mtay be.offet by 'to 'aejo et h Uevents, while some of the losse might be even bi;gger.

.So*ur-. Resources and Technlogy yDivision, Econoaic Research Service, US Department of Agriculture, Climate Change:. c.no, ic Imlleatloos br World Ariculture,Washngon DC, November 1990 (_rafl).

The procedure described above is an example of a "dose-response," or "production function"approach to valuation. The "dose" is soil erosion, the "response" is crop loss. Another approach wouldbe to look at the costs of replacing the nutrients that are lost with soil erosion. Nutrient losses can bereplaced with chemical fertilizers which have explicit market values.

B. Burkina Faso: BiomassWhere it is not possible to engage in detailed assessment of the costs of resource degradation it is stilluseful to obtain "best guess" calculations. In Burkina Faso estimates were made of the total amount ofbiomass lost each year in the form of fuelwood and vegetation. The resulting losses show up as foregonehousehold energy (fuelwood) which can be valued at fuelwood market prices; foregone millet andsorghum crops which can be valued at market prices; and reduced livestock yield due to fodder losses.Fuelwood losses amount to some 47 CFAF billion, livestock a further 10 CFAF billion, and cereal lossesa further 15 CFAF billion. The grand total amounts to some 9% of Burkina Faso's GNP.

It cannot be deduced from this that Burkina Faso's GNP is 9% less than it otherwise would be.This is because resources would have to be expended in order to rehabilitate eroded areas and to preventfurther damage. But if the resources required are small, then the 9% figure is a ballpark estimate of thedirect loss to Burkina Faso.

Provided they are credible, national environmental damage cost estimates can play a useful role

13

N.A THIE COSIS OF SOIL EROSION IN MALI

The oUos of soil crosion in Mali ame shown using both the dose-response approach and the nutrient replacement approach.Because sil loss in any one year has effeots in subsequent years the data show both an annual loss and a present value loseexpressed as a loss In a single year. Several conclusions emerge:

(a) economic losses from soil erosion are high enough to warrant conservation investments in some areas in the southOf the countty;

(b) investing in additional agricultural output may be less profitable than a sinple financial appraisal would suggest.It is necessary to build in to the analysis some estinate of expected soil erosion, and this will lowor rates of return;

(e) most importantly, it is necessary to ask whv soil erosion occurs. Restrictions on access to informal redit andinsecure find tnure are imnportant factors. High risks also contribute to high fanner discount rates: measures canbe taken to reduce risks.

The nutrient replacement approach. which values the soil loss at the cost of replacing the natriets, shows higher values thanthe crop.retponae estimate ($7.4 million p.a conpared to $4.6 million p.a.).

Farm Inome Losses in Nati Due to Soil Erosion. 1988

Based an USLE and Farm Bdgets:

Vationwide Annualt hcome Losses US$4.6 mitlion = 0.2X GDP 0.6% Agrieulturat CDP

Wscounted Present Valtue-of Income Loss USS31.0 mitlion ~ 1.5% GDP - 4.0% AgricuLturet-GDP

Nettionwidi Arnwat-toss BOsid onNutrient 6Reptaceent USS7.4 million = 0.4% GOP 1.OX Agriculturat &DP

Sotre:. IJishop ad J.Allen, e lOn -Site Costs of Soil Erosion in Mali, Envitonment Deattmen Workiog P"nr No.21November 1989, Wod dBa, Wahi DC.

in assessing development priorities. Because environmental damage costs do not show up explicitly inmeasures of national product, planners have no obvious incentive to treat environmental damage as apriority in development plans. Increasingly, however, environmental concerns are entering intodevelopment plans as the GNP costs of degradation are shown to be significant and sometimes verysubstantial (see Box 7). Arguments of this kind are particularly appropriate at the level ofmacroeconomic management; it may be more important that the Ministry of Finance appreciates the costsof environmental degradation than that the Ministry of the Environment does.

2. Modifying the National Accounts

Macroeconomic management makes extensive use of the national economic accounts which recordmonetary flows and transactions within the economy. Their primary purpose is to record economicactivity, not to measure aggregate wellbeing in the nation. None the less, national accounts are widelyused to indicate wellbeing, and rates of change in national aggregates such as GNP are widely construedas measures of "development." Whether the accounts are designed to record economic activity ormeasure wellbeing, or both, they are deficient in respect of their treatment of the environment. Economicactivity involves the use of materials and energy, and, once transformed

14

2 1!SUMMARYNATONALENVIRONMENTALDAMAGEEIMATES

Countrv Eorm of Enviorwmentat Damase Year

Burkina Faso Crop, livestock and fueltood losses 1988 8.8due to land degradation

Ethiopia Effects of deforestation on fuetwood 1983 6.0- 9.0sUppty and crop output

Madagascar Land burning and erosiOn 1988 S.0-15.0

MalawTi Soll erosion 1988 0.5- 3.1

Mati On site solt erosion tosses 1988 0.4

Nigeria Soil degradation, deforestation 1989 17.4water potLution, other erosion

Irdonesia Sofl erosfon and deforestation 1984 4.0

Hungary PolLution damage (mainly air pollution) late SOs 5.0

Potand Pollution damage (mainly air poltution) 1987 4.4- 7.7'

goermany* PotLution damage 1990 1.7- 4.2

Netheiftands Some pollution damage 1986 0.s- 0.8

USA**' Air pottution control t191 0.8- 2.1Water pollution control 1985 0.4

*e-pili6cati°n

*0 benpfits of envirnettal poly, I.e.ovoided deamas rather than actual damage osts. Whilet1# eates use diftrnttachnlqes, ielh*. to difomnt yesas and vaty in the quality of' the underlying res"th they sugge some broad ierctatioIn e devloped world total gis environmunt damage may be around 24% ofONP; in the East Eur n unties pes5.10%- of ONP, d in i the poot deveiojg nations poaps 10% and above. 'rThi does not nean it is. wordi avoiding all thisdame. Consrvatiion costsetiftes as w6ll. But since many eonservtion measures involve te wmoYal of cononicdistotions(such ias pric control6, subuidiesf, undefined esource dghts the cods of consrvration will, in many wses, be ow.

oursces: Burkbut Pasi: World Bank data Ethiopa: K.Newcombe, *AnEcononic Justificeton ibt Rutal Affore.sttio. th C.aseFolflpthip -G.-clirnmi and .Wsribid, Enbmen e m imtad &onon bevewon -jnt, 1n fHauiveutItyrees, Ralote, 1f89;Ms'a soar Wod B Mad ar - EWOmnv tal Ac*t-n Plan July 1988Malawh Word Dank

data- Mali: J.Dishop nd I.Allen, 1he On-Site Cost of Soii Erosion ia MJlI. World Ba-En nronnnt Depare WorinPaper No.21,Noven9ber g98q; fige.wa. World Eankg Towards the Deve et of an t rl Action Pl i aWodd Bank, eca.18, 199 ..htdonesia: R.Rpettoi HWsting Asse - Natural esourcess in the National lwgmiein A cOunta, World Resore Instuute,Washaington 19W9,ansd W.McOratIt and P.Armns.ie Cots of Soil Er1sion on Iv a .iuaI 1Rce Aec i jW;World Bank, Enaviroment Department Working Paper IS, Washington, August 1989; ft pungry Wodd -ank data; Polnd:D.W.Peatse and J.Werford, E0nv1irnent and 1coanmic Develonment: Mana2lna Natura Resources ln the DevePioting Woldfo4coming 1992; Oenmenyr Federal Minisg of the Envitonnwernt Advantases of Envisonmenta Protection. tElonntil Pbllution. Bonn. September 1991; Ndethedands: J.Opsohoor, A Review .of Monetary Estimates 'OfBeefita ofEafVitnmeital bnprovent in the Netherlands', 'OD, Paris, October 1986; USA, P.Potey,, 'AlI'Pollutdon lcy,,iisP.Por¶ney (cd). Pub5cPolci for l4iYiton~aLEmli2g. Resources forthe Future, Washigton DC, 1990j.nd A.FPeemuaa,NWater Poxlltion Policyt.ia Porny. On.ct.3 -

into products, those same resources become, sooner or later, waste products. Any measure of economicactivity which ignores these materials and energy flows will fail to record important activities which affectthe sustainability of economic activity. In the same way, any measure of wellbeing which ignores thesesame flows will fail to measure sustainable wellbeing. For these reasons, there is a widespread consensus

15

that the national accounts need to be modified - at least with respect to the way in which environmental"stocks" and "flows" are recorded.

Material and energy flows begin at the point of natural resource extraction, harvest, or use, Theyend as waste products: emissions to ambient environments, discharges to water, and solid waste to landor sea. Logically, then, GNP needs to be modified to account for:

* any depreciation of natural capital stocks, in the same way that net national income equals grossnational income ls estimated depreciation on man-made capital. This is a measure of the 'drawdown" of natural capital;* any damage losses accruing to human wellbeing from the extraction, processing and disposalof materials and energy to receiving environments.

Both adjustments involve economic valuation. The first involves valuation of the natural capitalstock, and the second, valuation of such things as health impairment, pollution damage to buildings, cropsand trees, aesthetic and recreational lossws and other forms of "psychological" damage. Nationalaccountants are not agreed on how best to make the appropriate adjustments. At the very least, grossmeasures of national income should be adjusted so that:

Modified GNP = Conventional GNP + Value of Environmental Services - Value of EnvironmentalDamage.

In this way, additions to national parks, for example, or improvements in pollution levels, would beincluded as positive entries for modified GNP, and damage done would enter negatively. The way inwhich damage should be measured is disputed. Some experts measure it in terms of expenditurenecessary to offset the damage in - so-called defensive expenditure. Others propose using the kinds ofvaluation techniques which attempt to elicit willingness-to-pay to avoid damage or to improveenvironmental quality. Under certain circumstances it happens that defensive expenditures ar perfectmeasures of WTP, but the general use of defensive expenditures is strongly disputed in the nationalaccounting literature. Moreover, defensive expenditures include both final and intermediate expenditures,breaking the equivalence between factor incomes and expenditures that is fundamental to conventionalnational accounting. Defensive expenditures by firms tend to be intermediate expenditures; those byhouseholds, final expenditures. It is significant that the literature showing how expenditures can beperfect measures of WTP relates only to a household context.

Depreciation on stocks of iiatural capital must also be valued in order to measure sustainableincome - the income a nation can receive without running down its capital base. In conventional nationalaccounting this is partly accounted for by estimating net national product (NNP), defined as:

NNP = GNP - Dk

where Dk is the depreciation on man-made capital (machines, roads, buildings etc.). The furtheradjustment that is required is:

NNP = GNP - Dk - D.

where D. is the depreciation of environmental assets.Box 8 illustrates both types of adjustment: deducting environmental costs from GNP, and

estimating the depreciation on natural capital stocks. However the debate about modified nationalaccounts develops, there is a clear role for economic valuation.

16

DQ.. MODtFIED NATIONALACCOUNTS: AGRICULTURE AND FORETRV IN THE UNITED KINGDOM

rvlaiotal but noftfcial adjustaeta have betet mado to one sector of the UK's national accounts, agriculture and fo(hly,Ea line with-the requirefient that positiwv environmental effects (bonefits) be added to GNP for this sector and Ohat negstiveffects are deducted, tho filiowing adjustments can be made

UKs'gtgrCa., Mitlions (1988)

Pinat marketed output .11,161-Input 5,663

mGoss product 5,498

Depreciation 1,470

Net Product

t nvironnentat servicesbiodiversity 94amentty -sreen bett 642-amenity; national parks 152

G Oovt. -expenditure to maintain- 58tandscape end comserved areas,

-d ean-up pol tut 7on-

1 Household defensive expenditures n.a.expenditures

- Depreciation- {D,,>-- - - - --carbn -fixig --- (+)146mater : -.- - 11-- . - - -- :-i : : -

* sustainable Not Product : - - - -

To fmake the -adjuatents to netp.roduct, Wbinatesaw.ete made of the per hectare recreatioma Ind 'iAenity values obtained &-'-samplealuatuiiion sdies. These were edin applied to tie whole area under consevaiadon desig-ations of o-neform.or .nother.,Wsilingness-so-ay to avoid daiimge wast estim directly: rather, the defWv expenditurt -approach weatis, iing-comaniesJ' oepvaditure and including government anti;pollution expenditures:..:No.est imiates 1weea!vilable or householdexpenditures. Natualia capivta depreciation- invol,ved estimates for the net aceretion or releas osf abon di omde anid -h vauaioof waetr pollution.- Beiase -thi sctoi has a net lixeton tate of C02 this item appears positive ia the 4l. st, ftousebold-e.penditures can be ignoreW then -theisetor'a accoutss how an upwards revaluation by 24% 'a signficant adjustnt'

oM0es: N.Adger and Ml.Witby, RNatiooI Accounting for the Externalities of Aidculttiwe and Fe eatr, cOtiti 0de ChUange- Unit,-University of Newcsle-up*nTyne, Working Paper 16. April 1991.

3. Setting National and Sectoral Priorities

Information on the economic value of policy changes can greatly assist governments in settingpolicy and sectoral priorities. Estimating damage or benefit figures alone will not be sufficient for thisprocess; it is necessary to compare the benefits and costs of policy. The presence of net benefits issufficient to establish that existing or planned policy is potentially worthwhile, though not sufficient toestablish that resources devoted to that end would not be better used elsewhere (net benefits may begreater still if the resources were put to alternative uses). But if benefits are less than costs then it canat least be inferred that resources should not be devoted on such a scale to that particular goal. Thisgeneral requirement to review sectoral priorities in terms of benefits and costs, has perhaps even greater

17

force in the developing world where government income is at a premium. Indeed, this has always beenone of the motives underlying the development of cost and benefit valuation techniques for developingcountries. Despite this, sectoral benefit-cost techniques have been used in fairly limited ways in thedeveloped world, and hardly at all in the developing world. Although there are a great many benefit-coststudies of specific policies in both developed and developing countries, there are only a limited numberthat analyze overall sectoral expenditures. These few attempts, however, have been revealing.

United States air pollution regulations probably cost some $13-14 billion in the single year 1981.Beyond that date, annual costs probably rose fairly fast as standards were better and more extensivelyenforced, and regulations grew in number. Benefits in 1978 were probably around $37 billion, and alittle above this in 1981. Thus, for 1981, the overall air pollution control policy almost certainly yieldednet social benefits. With a benefit-cost ratio of nearly 3, the regulations would seem to have beeneminently worthwhile, and it is unlikely that the resources involved would have yielded higher returnselsewhere. That conclusion needs to be qualified in several ways.

First, what was probably true of 1981 may not be true of years after that, especially as regulatorycosts probably rose faster than benefits. Second, the conclusion assumes that all the improvements in USair quality in 1978 were due to prior legislation. In practice, as evidence from a number of countriesshows, underlying structural changes in the economy have also contributed to improved air quality:switches from polluting fuels to less polluting ones due to ordinary market forces; reductions in heavyindustry in favor of lighter, less polluting industry; changed consumer habits, and so on. Third, thepicture changes somewhat if the regulatory policy is looked at in parts. It seems likely, for example, that1970s US policy on air pollution from stationary sources did achieve net benefits, while policy on mobilesources (vehicles) probably generated net costs. A similar result emerges for federal water pollutionpolicy. Costs of around $20-30 billion for 1985 compare to a best estimate of benefits of only some $14billion.

Not too much can be derived from such comparisons, but the results for mobile air pollutionsources and for water pollution suggest the need to look carefully at the costs of policy. It has to beborne in mind, for example, that the costs quoted are estimates of the actual costs involved, not the coststhat could have been involved if the most efficient policies had been pursued. One of the attractions ofmarket-based approaches (taxes, charges, tradeable quotas and permits) is that they have the potential tokeep compliance costs down. Savings may well be large, perhaps by a factor of two or more, comparedto the costs of traditional "command-and-control" costs (Portney, 1990).

In the real world of political decision-making, priorities are rarely set by reference to measuresof costs and benefits. In part this reflects a lack of understanding of the techniques involved, but it alsoreflects the fact that decision-makers have multiple criteria for deciding on policies (not all of themrational from the social standpoint, of course: chance, favoritism, patronage, whim and corruption arejust as important). Benefit and damage estimation are therefore likely to be V= of a wider package ofcriteria including distributional concerns, human health, and concerns over the quality of environmentalimpact and the sustainability of resource use. Box 9 illustrates one possible ranking of environmentalissues in Nigeria according to various criteria.

18

SErlVING 1MVRONMENTALPRIORITIES IN NIGERIA

To nk envinntental ptioritiOes in otrde to obtain guidance for development aid to Nigeria, th World Bank adopted thofollowing oriteda.

* ihpact of environmental degradation on GNP*.size of population affected by the environmentat issue

I incidence by income group of the degadationM a aure of tesourco integrity based on tho retationship betwoen waste (W) and etwironmnenal assimilativecapacity (A)* a inilar measure of rsourc Integrity based on a comparison of harvest and uge rates (H) compared to

* rgoneration rates R) for renewable resources.

Itc retults afe shown below (figures in square brclkets are indicative only). While elio data ate clearly imper&ct, the appoachyields sonm cohmre priorities. For exanmple, soil dogrdation, deforestation and water pollutioa ali rank high on each of thegena) criteria of GNP impact, distnbutional incidence and resource integrity. Such rankings can assist national ptiority setting.

ilslul GP Ponuttton e rdk Incidence WfSo/yr (miltions) (higher scores worae)

Soil degradation 3000+ s0 2-3 3 3-4.-

Water pollution 10W+ 40+ 3-4 3-4

Oeforestation 750+ 50 2-3 (2] 4

Coastat erosion c. 150 <3 3 2-3 2

Gulty erosion c. 1 00 10 2-3 2 2

Fish Loss c. 50 5 3 n.a i

Wildlife C. 10 ci 2 n.a 4

-A-irpotlUtion n.8 35 4 2.3 E13

-Water hyacinth c. 50 5 23 2-3 n: .

Source: Worldi Bank, Towards the Develonmert of an Envirornental Action"Pian for I4itia Western AfricaDepartmn*t Deeember 1990.

4. Project. Program and Policy Evaluation

The traditional role for environmental damage and benefit estimation is in project appraisal.The main manuals that have influenced theoretical and practical work in economic project assessmenthave not, however, addressed environmental issues. Issues relating to the treatment of environmentalfactors are not, for example, discussed at all in the main project appraisal technical manuals (Littleand Mirrlees, 1974; Squire and van der Tak, 1975; UNIDO, 1972). They receive very sketchytreatment in Gittinger's agricultural appraisal manual (Gittinger [1982J). In contrast, assessingenvirommental impacts has been the subject of a wholly separate set of procadures known asEnvironmental Impact Assessment (ETA). EIA is important in drawing decision-makers' attentionto the many forms of environmental impact. To some extent EIA also permits an assessment of theimportance of impacts. The main problem, however, is that EIA tends to be pursued either as anadjunct to conventional economic appraisal, or as a precursor; in neither case is EIA integrated into

19

economic appraisal.Extending project appraisl to account for environmental impacts, or to the assessment of

pure conservation projects, presents no conceptual problem for benefit-cost approaches. The typicalbenefit-cost assessment (BCA) calculates measured benefits and costs and converts them into aneconomic rate of return (ERR). In this process, market prices are adjusted for distortions, usingshadow prices. Environmental impacts are simply additional costs or benefits; economic valuation ofenvironmental impacts is essentially a matter of shadow pricing also. The traditional BCA rule forthe potential acceptance of a project can therefore be re-expressed as:

Et(B,-Ct-EJ).(I +r)" > 0

where B, is non-environmental benefit at time t, C is non-environmental cost, r is the discount rate,and E is environmental cost. Economic valuation is concerned with the monetary measurement ofE in this inequality. Environmental issues do, however, raise a further problem, namely the selectionof r, the discount rate, in the above inequality.

Projects

Box 10 shows how project economic rates of return car. be transformed when due accountis taken of the detailed environmental consequences of planting trees. The analysis makes extensiveuse of data on the various physical interlinkages in environmental and agricultural systems. Treeshave many functions, from producing timber for poles, to fuelwood supply, leaves for animal fodder,crop wind protection and, in some cases, the fixing of ambient nitrogen. The principles of BCArequire that all impacts be accounted for.

Programs

Just as project appraisal requires comprehensive environmental valuation so, logically, doesprogram appraisal. Programs tend to be amalgams of often interrelated projects, policy measures anddevelopment plans. As with single projects, the environmental implications of a program should beevaluated, and the overall return to the program should be assessed with reference to the inclusionof environmental enhancement components - tree planting, soil conservation, water supply etc. Inprogram analysis, ERRs should still be estimated wherever possible, especially where the intermixingof policy changes and projects is liable to make ERRs higher than if projects alone were beingevaluated. Box 3 illustrates some of the kinds of benefits from environmental conservation in atropical forest context.

Choice of Technology

Within a program the issue of choice of technology usually arises. A given developmentobjective may be met by selecting among a range of technological options. The program objectiveof meeting a given increment in electricity demand, for example, involves selection of incrementalelectric power sources which contribute to the overall objective of meeting demand at least cost.Whereas least cost power system planning has typically been couched in terms of the private costsof generation and distribution, environmental considerations require that the criterion be modifiedto become least social cost, and therefore inclusive of the environmental impacts of different energy

20

-RATES OF RETUN TO AFFORESTATION IN NORTHERN NIGERIA

;CareM e,ix;ntiion and' teaswetlt of the eviVonmental benefits of afforestation can greatly increase the ERR to foestryinvesameni-. Oni study in northom Nigeia assessed the benefits of afforostation in norhem nNigeria as:

* batting the fotwte decline of .oil Irtility (since trees typically reduce soil etosioo);-* rising current levels of soil fertity;* produoing tree products -Iheiwood, pole, fruits;* pteiuclng- oddew both ftom ratied produedvity of soils and from forest fodder.

The net prsent values (NPVs) and econoWic rates of return (ERs) that resulted for Shettetbeits (planting treS mainly for.winddprawetion) and farm forestry (nenniJng tres and crops) woee:

Shette-betts farm Forestr

-c ease 170 14.9 129 19.1

*tow' field, Hfi h cost 110 13.1 70 14.5

Nigh yietd 221 16.2

.ao erosfon . 100 13.5 75 16.6

-tore r:apid erosion - 109 13.6 .60 15.5

. rQIsto $red +` yeld JuIP 263 16.9 203 21.8

benfits Ony(* -95 4.7 -14 7.4

-* wa' and ftuit for farm forestry)

.:alXiculation of i;~~ costs and benefits inh Kano aea have tended to show Jates of return of around 5%. which has to be: pae it us=ualy ttfe hge t t 0%.In other wo,afforeston does not pay. &K onmsthe other.bbnbfts ore iAadled1 dratif icnreaies "lin-res of eitun on be secured.

.Tbe nalysis os t cowtin -wood beteft'" oly ptudces negative net prmt value.and ootrespondigly low eConomic..a-es of realtut. But if allowane .is made 'for th efcts of trees on crop yields; and for exqpected rates of soll erosion a theahbsence of aflotatlon, t te Is transfonred fol oth farm forestry and sheltobelts.

Sources: Adown Te Eomics, of Aff We in- a Cse Study in Africa. Johns Hopkins Universty Press, 1987 andiAirduao',Qconomic Asects of Affo and oi Conservation Prqjectsb,lnG.Sehimind lWadEviron iental

M*at Jobsm h opkins Universty Press, Balt(imor, 1980, 172184 -

technologies. In some developed economies this redefinition has resulted in the estimation of"externality adders." These are the surcharges or credits to be attached to specific energytechnologies according to their relative environmental impacts. Expressed in this way,the credits anddebits have to be measured in money terms, so that the monetary value of environmental impacts isused to calculate price adjustments. Box 11 illustrates the kinds of calculation that are involved. The"adders" are then added to, or subtracted from, the private costs of generation. As an illustration,nuclear power might be credited with avoiding carbon dioxide and acid rain emissions, but it wouldbe debited with a surcharge for any routine or accidental radiation risks. Several countries areexperimenting with estimation of externality adders. If applied in practice, the choice of energytechnologies in a least-cost planning system could change markedly.

21

.XI EXTERNALITYADDERS FOR ELECTRICITYGENERATION SYSTEMS IN THE USA

Suthaegea aro ealculatW accoing to the tometary value of impacts relating to sutphur dioxide, nitrogen dioxide, carbondioxide and paticulate emitsions. Nuclear power costs are based on the value of damage done by romtine emissions, accidentsand the coats of decouinissioning.

Slectricity TeAhnotogy Surcharge(USc/Kwh gen.)

ao"

Convbntional 0.058Fluidised Bed Combustion 0.028Integrated Gas Combined Cycle 0.025

Ltow sutphur 0.027:High sulphur 0.067

N;tural Gas

Steam plant 0.010Combined cycte 0.010

*00:;;~~M t:::cear Power-

'Routine emissfons 0.110Accidents 2.300.. -0 -: 0- Deconmisafoning . .0.500

Fof this p a uilar tdy, the rng would be (from most to least dantaging nuclear power, oi, coal and gas.ThisIt i bly iacc Iti puliW p t tedeveloped world. The-very1high penlty io nuclear power Ia een tib e

: Aeialy .afinct ofitht e esdi* t ed mma teni costs In in f plant choica, :ctore, the tretevai;e of this' penalty wouldde oi - nodilations to safety designs which would affect risk factors.

-S: trge R.Ouins et al, Envioi eta C of Etecgi-ty, PACE University Center for MEvironmental Legal Studies, Whitiuan. New York. September 1990.

The Polluter Pays Principle

The externality addition approach extends beyond choice of technology. Existing sources ofsupply and service can be priced to reflect environmental damage, as the general principles of optimalresource allocation would require. Adding surcharges in this way is consistent with the Polluter PaysPrinciple, to which OECD member countries subscribe.' The 'PPP' requires that those emittingdamaging wastes to the environment should bear the costs of avoiding that damage or of containingthe damage to within acceptable limits according to national environmental standards. As stated bythe OECD, the PPP does not require that environmental damage be valued in monetary terms,although it could be. Whatever the cost of achieving the national standard, that cost should, in thetirst instance, be borne by the emitter of waste. That the emitter's increased costs may then be

I For a full statement of the PPP see, "The Polluter Pays Principle; Definition, Analysis,Implementation" Paris: OECD, 1975.

22

passed on partly to the consumer is not inconsistent with the PPP. The costs borne by the emitterand the consumer can be thought of as a form of valuation. Regulatory agencies set standards onbehalf of the voting population, and the cost of meeting those standards becomes, in effect, aminimum estimate of what the regulator regards the damage to be. Nor is it essential for the generalPPP to be implemented via taxation or some other form of economic instrument (tradeable permit,product charge, tax-subsidy etc.). The PPP is consistent with traditional standard setting via"command and control" policies.

However, economic instruments do have many attractions relative to command and controlpolicies. If this approach is used, then it is essential that any charge or tax should be at leastproportional to the damage done. Valuation therefore becomes important in giving guidance to thesetting of such environmental prices. Box 12 shows how a tax on greenhouse gases might becomputed using the economic valuation of globai warming damage as a base. The analysis suggeststhat, if global warming produces an impact on global national product of around 1% , then a verymodest "carbon tax" of about $7 per ton of C02-equivalent would secure an optimal reduction ofgreenhouse gas emissions of about 10%. But if damage rises to 2% of GWP, then the tax is above$60 per ton. Taxes computed so that they secure the level of pollution reduction that yields thegreatest net benefits are "optimal pollution taxes" and are special examples of the PPP.

IMQX1 lDERIUVING A CARBON TAX FROM GLOBAL WARMING DAMAGEESTIMATES

S-ver s4oe. have atteted to caloulate the tnonetauy value of damage from globe* waaming. Estimates arc bighly uacenainbut can be-expetd i- improve ai the underiying'physical data and global circuation models improve. Nordhaus ostimates -thatdamage- ight am-ont to some I pe cent- of groas word prduct (GWP) expresed in present val.ue tems Comparing hi to-the robabla costs of reducing grO004s gis emissions, he eslimates that th reduction in gaes tht Would bong the greatest:net benefit to the wrid would bc some 10 per ent off a baseline trend of pryected emissions.

- -greenbhoise-. total -ost of - total benefit net totatgas eimtiss-ion reduction of reduction benefits--redut - -- MSb)Y tSb) ($b)

- :-1 --- .--- 4.b0.04.-0.-56-. ::- : 2. -- --- :0.12 1.20 1-.08 -:-- -- .3 -0 .Z4 --- 1.80 1.56

4 0.40 - - 230 1.905 :0.61-- 2.90 2.-3910- 2.20 -5,90. 3.70

-- 11 - -2.9:- 6.40 3.50--- 15 - -: -6.80- 8.80 2.0025 - 30. 70% - - 14.70 -16.00

Ihe stimates- suggeist that gree'Aous gases -.aggregated and measured in tes of COeuivalent - should -be reduccd in -agegait by litile over 10%. To find tho surchargc newessay to achieve this optimal reduction-t it mi neasaay to calculate thedaage d*otw by eao s 4tr6t -of jfrakaltt T is is $7.30 per ton C0)2-equivaata X the datrage- cwg sb*n. But itwould .sue to $6per ton if damages were twice the estimated level - corresponding to 2% of GWP. .These istimates are ento many reservation. If it is conmpaatively easy to control emissions, then th Cats of contol miay be less than owq, dictat.nga hihe optimal e*del of greenbouse gVs reduction. If tere are, as many scientists beleve, certai thresholds byond wichdamage would become -vy svere, tfihen the benefit estimates would nsee also justifyng sticter conols. Cendainly, thosecountrie -, commtotd to grenbouse gas oonitms, over and abovo those for 0hlotofluotoeatboe ae talking aou fmore e*tenswoe levest of control.Source for data: W.Nordhaus, "A Sketch of the Eonomics of the Greenhoue Effect', Ameican Economic Review01o.51,No.2,199, 1464250.

23

Policies

Policy changes can also be evaluated using the benefit-cost framework with special referenceto environmental implications. Box 13 shows the computations used to determine whether or not theEnvironmental Protection Agency (EPA) would recommend reducing lead in gasoline in the USA.

5. Valuation and Sustainable Development

The need for economic valuation of environmental impacts and assets arises quiteindependently of the definition of sustainable development. Simply pursuing efficient policies andinvesting in efficient projects and programs requires valuation to be pursued, in order to be credible.At the most general level of intergenerational concern, valuation is still required. If transfers ofresources are to be made between generations - with the current generation sacrificing for the future,or future benefits being lost for the sake of present gain - then it is essential to now MIMI is beingsacrificed and how much it is that is being surrendered. It is not necessary, therefore, to invoke thephilosophy of sustainable development, however it is defined, to justify a focus on economic valuationin a development context.

However, if one or more definitions of sustainable development are to be espoused, the roleof economic valuation needs to be investigated. An efficient use of resources need not be asustainable one. The optimal rate at which an exhaustible resource should be depleted, for example,still requires the rate of use to be positive; and in the absence of repeated discoveries of furtheridentical resources, the resource must be exhausted eventually. Every unit used today is at the costof a foregone unit tomorrow. Global warming is another example of an activity that impairs thewelfare of future generations. "Sustainability" therefore implies something about maintaining thelevel of human wellbeing so that it might improve, but at least never declines (or not more thantemporarily, anyway). Interpreted this way, sustainable development becomes equivalent to somerequirement that wellbeing does not decline through time. The implication for valuation is nowsomewhat different to what is implied by consideration of efficiency alone. It now becomes necessaryto measure human wellbeing in order to establish that it does not decline through time, and, sinceenvironmental assets contribute to wellbeing, it is necessary to measure preferences for and againstenvironmental change. The problem from the point of view of development planning and aid is thata simple "trends continued" cannot be assumed. This is particularly true if the environmental changesin question risk harming future wellbeing in any significant way. In terms of Box 14, a developmentpath such as A appears to be sustainable; B is non-sustainable (but could be "efficient"); whilst C isboth unsustainable and non-survivable because average wellbeing levels fall below some minimumlevel (e.g. a poverty line). But from the vantage point of 0 in the diagram, it may not be possibleto tell whick development path a country is on. Hence, defining sustainable development as sustainedwellbeing is of only limited help in real world development planning. The declining rate of growthof wellbeing (path C) might provide an early indicator, but would not detect the unsustainability ofpath B. Moreover, A and B look very similar to begin with; what matters is knowing whether theconditions for sustainable development are fulfilled or not.

If the focus is on identifying the conditions for achieving sustainable development, then it maybe that wholly non-economic indicators will suffice. For example, computations of the carryingcapacity of natural environments could act as early warnings of non-survivability (path C). Carryingcapacity measures the number of people whose livelihoods can be sustained by a given stock ofresources if each of them consumes the minimum level of those resources necessary to survive. Ifthe numbers resulting are less than the actual population, then the situation appears to be non-survivable and therefore certainly non-sustainable. Carrying capacity indicators would need to be

24

iSC: THE USE OF BENEFIT C COST ANALYSISIN DECISION-MAKING, LEAD IN GASOLINE

Under Exeoutive Ordetr 1291 of 1951 US govemment asgencies were required to use wRegulatory Impact Analysis (RIA) andt adopt egulatory poceas6se that would maximnize net benefits to society'.The Order was the first to eabish the net benefitobjectivo as the. oriteiron for adoptng regulatory processes, although its adoption has been circumscribed byexlistnglawgltedng-t* oather -objectives ., .' . ,.,-

Benefit - cost afalysis played -a important role In the adoption of retgulaon con g lead m gasoline. Ambient leadcoenztrtion weoe thott to b lnked to serious health effects, including retardation, kidney disease and even death. TheEnvrnmntal Proection Ageacy conducted a benefit-cost study with the reults own below,

'The regulation involved reducing lead in'gasoline from 1.1 grams per gallon (jpg) to0.1 gpg. The costs of the rule ar shownAS 'total refiiag coasbt. Rlfinery osts increase becawe lead his traditionally been used t6 boost octane levels in fael, and other

-manis woud have to be found to achieve tiis. ITe benefits included:

(e) iflDptOVn childreti'a health-(b) impreved blood pressure in adults(c) reduced damges fom misfuelled vehicles, arising from hydrocatbon, ltOx and CO emissions(d) impacts on maintenance and fioel economy.

Childrenes H .'The EPA study found tht blood kad levels closely tracked trnds in gasoline lend. Medical coats for the careof obldren would be redued by' roducing lead concentrations, and themr would be lass need for compensatory eduction forIQ-imaiired children. These 'savings afe shown as 'children's health effects' in the able.

Adult ~od B lesauro. Blood lead levels ware thought to be associated with blood presure and hypertendon. Medica coatswould'be savdIfthese illnesses could be' reduced. Moreover, some heart attacks nad okes would be avoided. A value of a*stastifof $S 4mllion was used br the at. The tesulting values show up in the 'adult blood ptares' row of Table 'X.T1ey are seen to be high because of the involvement of'mortality-avoidance in this benfit.

,eC, p,oliqpWk. Reducing lead' in gasoline ao reduces other pollutants. Tis is becaus -making unleaded ficl the 'noert'reduces the risk of 'ml4iselling - l.e. uing leaded fuels in vehicics designed for unleaded fueis. The mechanism wherebysniselllng is rdced is throuagh the hi8her cost of leaded fuels at the new low b- kad concentration. This deers dtiver* fiompurchsing the leaded fuel. As misfuelling is reduced, so emissions of HC, Nox and CO are rduced. Damage don by thespolltags was esWmted by studis of ozone pollutioo damage (ozone arises from He and CO emisionlo), but estdmates weoas made of the value of tm equipment destroyed by mnisfuelling. The figutes appean in the tow 'conventional pollutants'in the table are in fict averages of the two methods.

M-ia&weance costs for yehles were expected to fl due to the reduced corosive effects of lead and its saveVnges on eaginesand oxhaust systems. Fewer engine tonw-ups and oD changes would be needed, exhaust systems would last longer. Fuel economywas expeted to 'liS as the new technologies, to ralse octane levels to what they were previously, also 'increased the energyGontsat offitels., here would also be rieuced fouling ofoxygensensors. Maintenance benefits outweighed ftel economy benefitSby around 6 to I. Tne totals are shown in the table.

The net bnefits from reducing lead in gasoline are seen to be substantial, even if the blood preswre benefits (which dominatethe aggregate benefits) are excluded. Indeed. it can be seen that the regulation is worthwhile even if all health beneMits areg4ode in the event. the blood prestue benefts were excluded from the final decision becau tIhe tesearch establishing thisLink was judged too rec to permit adequate review. The lead regulation was also of interest because of the introduction ofa 'lead permits system' to reduce the finanwial burden on the refining industry. Eseritially, this alowed 'kad qotas' to be tmdedbetween refiners. Refies who found it easy to get below the lImit were allowed to sell theit 'surplus'lead rights to refiners'whofound it expenve to get back to desirable octane levels without lead. The particular featur of the lead-in-gasoline benefit-coststudy that miade it a powerful aid to decision-making was the clear-cut nature of the net benefits even when unertainties aboubenefits wore allowed for, But it wao also executed carefully and in comprehensive detail,

Soce; US Eavironmental Protectioo Agency,EPA's Use of Benefit-Cost Analysis 1981-1986,EPA-230-0-87-028, WashingtonDC,'August, 1987: and US EPA 119851, Costs and benefits of Reducine Lead ig Gasolineg Einal Regulaton InnacT AnalysiXEPA-230-05845006, Washington DC, Febuary.

25

POX3 t4 =I= DEVELOP14ENTPATHS'00; ;S'

.. . . - .. . A ' ' O'' 0 - ' '';~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. .

WeUbefin'

Minimum Sustainable L - - -

calculated on a regular basis for the measure to be useful in this context. Typically, carrying capacity

measures are produced on an ad bo basis and for a single year only. None the less, they offer one

anticipatory measure of sustainability. Other physical measures could include assessments of the rate

of resource use relative to the rate of resource regeneration, and the rate of waste emissions relative

to the assimilative capacity of the environment. It may be, therefore, that some light will be shed on

sustainability indicators by non-economic approaches, especially if they can be developed to include

other measures of stress and shock to underlying natural resource systems.The literature on environmental economics tends to suggest that the clues to sustainability

lie in the quantity and quality of a nation's capital stock. Part of the intuition here is that nations

are like corporations. No corporation would regard itself as sustainable if it used up its capital

resources to fund sales' and profits' expansion. As long as capital assets are at least intact, and

preferably growing, any profit or income earned can be regarded as "~sustainable". Analogously,

nations are no different. Sustainable growth and development cannot be achieved if capital assets

are declining. Indeed, some economic growth models suggest strongly that, if capital assets are kept

intact, one concept of intergenerational equity - that of equalizing real consumption per capita - can

be achieved providing population growth does not outstrip the rate of technological change.

If a condition for achieving sustainable development is that capital stocks be kept intact, then

the problem of how to tell whether a nation is on a sustainable development path is partially resolved.

It is not necessary to observe real levels of wellbeing as such, but instead to look at the underlying

condition and amount of the capital stock. Unfortunately, while this approach solves one problem,

it raises many others. First, it is necessary to know what counts as capital. Second, this must be

measurable, otherwise "constancy" cannot be tested (constancy, throughout, should be read as

"constant or increasing").Mte national accounting issue arises again in this context of defining and measuring capital.

Capital assets in the national accounts are typically confined to "man-made" capital - machines, roads,

26

factories. Some accounits include mineral wealth as part of capital. The depreciation on man-madecapital is then deducted from GNP to give NNP. A more comprehensive definition of capital andincome would include human capital (knowledge, skills etc.) and natural capital (environmentalassets). The primary condition for sustainable development would then be that this aggreg stockof capital should not decline. Put another way, depreciation on this capital stock should not exceedthe rate of new investment in capital assets.

But how is the capital stock to be measured? For some economies, heavily dependent on oneor two natural resources, it may be possible to use a physical indicator of reserves or available stocks.But in the vast majority of cases it will be necessary to find another measuring rod for capital.Typically,that means monetary units; it becomes necessary to value capital, including environmentalcapital. Valuation and sustainable development are - again - intricately linked. How much this linkmatters depends in large part on how likely unsustainable development paths are, and, of course, onthe value judgement that sustainability "matters". Opinions differ. Past development suggests thattechnological change and the expansion of human knowledge will make resource use more and moreefficient with consequent benefits to subsequent generations. Some technologies have, of course,brought their own damage costs (chlorofluorocarbons, for example): technology is not a free good.How far future development will be sustainable perhaps revolves around the issue of irreversibility.The more irreversible the damage done by the current generation, the fewer degrees of freedomfuture generations willhave to expand their own wellbeing. Sustainable development certainly looksas if it should be partially guided by the need to avoid significant irreversible damage.

If securing sustainable development has something to do with monitoring and measuringaggregate capital stocks and not allowing them to decline, then there need be no particular role forenvironmental protection in sustainable development. Environmental assets could decline in quantityas long as depreciation in these assets was offset by investment in other man-made assets or humancapital. But even if this view of sustainability is accepted, then valuation is stil central to the process.For it is not then possible to know whether offsetting investment has taken place unless there is somemeasure of the rate of depreciation on natural assets and their foregone economic rate of return. Stillothers will want to make a special case for the environment. The acceptability of "running down"environmental assets provided other assets are built up will depend on relative valuations and onjudgments about other measures of sustainability, as well as the moral view about destroying theenvironment.

Discussing sustainable development in broad terms risks giving the impression thatphilosophers and economists fiddle while the Rome of under-development burns. But there isnothing in the idea of sustainable development that lessens the emphasis on development now, oron targeting the most vulnerable. If it is used to justify large sacrifices of real income and wellbeingnow for very long term gains that are highly uncertain, this emphasis becomes lost. Elicitingeconomic values can help guard against the latter risk by showing, as far as possible, where and whenenvironmental protection yields the highest returns.

V. Valuation ane Discounting

Many envirommental problems - nuclear waste storage, nuclear power stationdecommissioning, the release of long-lived micropollutants, ozone layer depletion, global warming -are likely to have their major impacts well into the future. Their costs are therefore likely to beborne by people alive 50 years, and longer, from now. Conventional benefit-cost approaches wouldregard $1 of future damage as being less important that $1 of damage now because of thephenomenon of discounting. The underlying value judgments of benefit-cost analysis are that

27

"people's preferences count" and that preferences are justifiably weighted according to existing incomedistribution. If the sovereignty of preferences is to be applied consistently, then the bias of thecurrent generation's preferences towards present as opposed to future benefits, and against presentas opposed to future costs, needs to be reflected in decision-making. This is the essential rationalefor discounting. Typically, any benefit (or cost) accruing in T years time is recorded as having apresent value of:

PV(B) = BT/(I + r)T

where r is the rate at which future benefits are discounted.The problem that arises with discounting is that it discriminates against future generations.

In one sense this discrimination is not a problem - the discount rate is meant to discriminate in thisway: this is its purpose. But this discrimination presupposes that meeting the current generation'swants is more important that meeting future generations' wants. Discounting is consistent withimposing a major cost on the future for the sake of a relatively small gain now. The usualjustification for this is that future generations will be better off anyway - their incomes will be higherbecause of economic growth. They will therefore attach less value to an extra $1 of income than thecurrent generation (the "diminishing marginal utility of income" argument) and will perhaps be betterplaced to counteract any ill effects of the current generation's activities that spill over to them.

Yet a cost accruing in 100 years time and amounting to $100 billion would, at a 10 per centdiscount rate, have a present value of

$100 billion / (1.1)1"

which comes to $7.25 million. That is, any benefit-cost study of a project that imposed this futurecost would record the damage done at only $7.25 million, even though the actual damage done isnearly 14,000 times greater than this. If intergenerational equity is to be a genuine concern, thendiscount rates of the order of 10 per cent - which are typically applied to investments in thedeveloping world - would be inconsistent with that concern.

Intergenerational considerations would seem therefore to call for some fairly fundamentalrevision in the way project and policy appraisal is carried out. Two broad categories of modificationhave been suggested, although it is as well to note that all the arguments are the subject of extendedcontroversy. The first set of modifications requires what a two tier approach. Allocations ofresources over time are treated differently to allocations within a period of time. Some kind of"sustainability rule" is applied to the inter-generational allocation, and fairly conventional rules, suchas maximizing net present value of benefits, are applied within the time-frame. The second set ofmodifications are made directly to the discount rate itself; the framework of maximizing net presentvalues is left intact, but the actual rate of discounting is changed to reflect intergenerational concerns.

1. Sustainability Criteria

Some authors argue that simply changing the discount rate - usually by lowering it - is amistaken procedure. Sustainability argues for a major redefinition of the underlying objective -namely intergenerational fairnss - and it is therefore inadequate merely to modify a procedure basedon efficiency gains and losses; an issue of fairness cannot be handled by modifying efficiency criteria.Furthermore, an appraisal procedure that evolved from concerns mainly with localized and certainlymarginal changes to the state of the economy, cannot be applied to issues that are global in a non-marginal sense, where significant changes in wellbeing are involved. A tool for fine-tuning decisions

28

should not be applied to contexts where fine tuning is not the issue. More fundamentally, transfersbetween generations should not be treated in the same way as decisions about how to use resourcesavailable to the current generation. Equity issues within a generation can be treated by makingresource transfers between individuals. Equity issues between generations need to be treated thesame way: pursuing efficiency within a generation does not guarantee a fair distribution of resourcesthrough time.

One way to avoid some of the concerns about discounting is to impose a sustainabilityconstraint. Essentially, this amounts to formulating some rule which would maximize gains towellbeing now, provided this does not reduce the wellbeing of future generations below that of thecurrent generation (a rule very much in line with the Brundtland Commission's definition ofsustainable development). This is a departure from benefit-cost analysis because it requires thatwellbeing be constant or increasing over time. Benefit-cost analysis would be consistent with reducingcurrent wellbeing if it yields a greater benefit for future generations, and vice versa. Rules of thiskind have been formulated in terms of maintaining overall stocks of capital of all kinds - man-made,human and natural. The basic idea is not to ensure equal (or rising) wellbeing through time, butequal or rising capabilities for generating wellbeing through time. The stock of capital is the meansof raising wellbeing, and hence it is this stock that has to be maintained or improved.

In practical terms, such rules would require monitoring and measurement of capital stocks,and an investment policy that sought at all times to ensure that net investment offset depreciation("compensating investments"). The main difficulties would lie in the issue of measuring capital;physical units would not be adequate, due to the heterogeneity of capital (the "adding up" problem).Hence a valuation procedure would be needed. Then either the total value of the capital stockwould be monitored and adjusted so that it is constant or rising, or perhaps the price of the capitalstock would be used as an indicator - resource prices, for example, would be monitored and demandand supply adjusted so as to secure constant real prices through time.

As yet, little advance in this area has been made beyond attempts to recompute GNP toreveal net investment levels that allow for depreciation on some natural capital assets. Suchprocedures are promising, but to be all-embracing they would have to be extended to all forms ofnon-marketed, and especially environmental, capital. At the global level, substantial problems arise -some forms of capital will depreciate because of pM action (the ozone layer, earth's surfacetemperature). How, therefore, is the "stock"of such assets to be measured? Valuation can assist butthe prospect is fairly daunting; it becomes necessary to know not just the "price" of global warming(the marginal damage done), but also how that price will change over time. The same holds true fortropical forests, wetlands, and so on.

Economists, philosophers and ecologists are only just beginning to tackle the ways in whichsustainability might be measured. Fairly clearly, many of the implications of sustainability - howeverthe latter is defined - will be the same. Non-marketed assets must not be treated as if they have azero price. Environmental impacts must be fully accounted for. The national income accounts mustbe modified. But whether all this is enough to raise significantly the probability of securingsustainable development is not clear.

2. Modifying the Discount Rate

Environmentalists have traditionally been more concerned to see actua discount rateslowered. Four approaches to modifying discount rates may be considered. These are:

* setting the discount rate equal to zero* computing a consumer discount rate

29

* computing a producer discount rate* computing some weighted average of consumer and producer rates.

Zero Discount Rates

The argument for zero discount rates is essentially as follows. The point in time at which anindividual existsshould not determine that individual'swellbeing; there must be an "impartiality"abouttime. Wellbeing at one point in time cannot count for more than wellbeing at another. Thisargument has a long tradition in utilitarianism, being clearly stated by Sidgwick, for example. Anotherdefence of impartiality with respect to time is given by Rawls in terms of his "original position"argument. An imaginary group of people coming together to determine an allocation of individualsto social groups and particular time periods would not choose to favor one group or one time periodover another, since they would not know to which group or time period they themselves would beallocated. Thus there must be no discrimination between time periods if there is to be a "just"allocation.

There are two components of discounting. The first relates to the discounting of consumptionstreams and is usually justified by assumptions about diminishing marginal utility of income. Thesecond relates to the discounting of utility itself. This latter is perhaps what is meant by true "timepreference", the former being due not to time but to differences in levels of consumption. It can beshown that if time preference is zero and interest rates are positive (for the first reason noted above),then any individual would rationally reduce current consumption levels to zero in order to make themarginal utility of such consumption infinite. Everything would be transferred to the future.Adopting a zero rate of discount for utility - which is what pure equality of treatment for generationswould imply - would imply a policy of total current sacrifice. It would appear that zero rates mayhave implications contrary to the purpose advocated by their supporters.

Consumer Discount Rates

The standard formula for discounting future consumption is:

d-de- a + A.g

where dc is the consumer discount rate, a is the "rate of pure time preference" (i.e utility discounting),it is the elasticity of the marginal utility of consumption function, and g is the growth rate of percapita consumption. If the function linking utility to consumption is logarithmic, then y = 1. If,further, pure time preference is rejected on ethical grounds, then a = 0 and we have

d = g

The discount rate becomes equal to the (expected) rate of growth of per capita consumption. Takingpast growth rates as a guide to expected rates, Table 0.1 shows estimates of d; for selected countries.

30

Table 1 Estimates of Consumer Discount Rates (% p.a.)

Country Growth of Real Growth of Discount RatePrivate Consumption Population

(1) (2) (1) - (2)

USA 3.3 1.0 + 2.3UK 2.8 0.2 + 2.6Japan 5.0 1.0 + 4.0Ethiopia 2.4 2.8 - 0.4Ghana 1.7 2.6 -0.9

Chile 0.8 1.7 - 0.9Thailand 5.8 2.5 + 3.3

Notes: assumes a 0, = 1 ; growth rates computed 1965-1988 from WorLd Bank, WorLd Devetounent Rewort 1990,OUP, Oxford.

One result of this approach is that discount rates for the poorest countries become negative.Yet behavior towards natural resource endowments in those countries is clearly inconsistent with thisoutcome; resources are depleted as if personal discount rates were very high. Moreover, marketlending rates are positive and high. The application of the income utility approach in these contextsmay be questioned. It implies, for example, that as incomes double households enjoy only one-halfthe utility from the extra units of income.

Table 1 suggests that rates for industrialized and industrializing countries are in the range 2-4%. Estimates will, however, be conditioned by the past period used to make the calculation.Moreover, while the value of unity for p is convenient, some empirical work suggests values of around1.5 as more accurate. The effect of p = 1.5 in Table I is to raise effective discount rates to 4% forthe UK and USA and above 6% for Thailand and Japan. The exclusion of a from the estimates hasalso to be questioned. Little evidence exists about pure time preference rates in the industrializedworld: a rate of 1.3% for the UK has been suggested, for example. Added to the rates in Table 1this would suggest a consumer discount rate in the UK inclusive of pure time preference of about4% for u = 1 and 5.3% for p = 1.5.

Producer Discount Rates

If capital markets were perfect, rates of return on capital would be equal to the rate d. above.In practice, a number of distortions in the market place give rise to divergences between d0 and theproducer rate of discount dp. Corporation taxes, for example, mean that a company must earn r%if it is to pay its shareholders s%, where:

r = s/(1-t)

and t is the corporation tax rate. Company taxation necessarily makes producer borrowing rateshigher than the rate at which consumers discount the future.

Many economists argue that r% is the "correct" rate of discount because it measures theopportunity cost of using up $1 in public expenditure: it is the foregone rate of return on marginalinvestment in the private sector. To find r, one might take the weighted rate of return on equity anddebt. The resulting long run, weighted, average cost of capital to the private sector in theindustrialized world would be perhaps 7% in real terms.

Clearly, if a discount rate of 7% is used, damages from distant environmental impacts suchas global warming would appear insignificant in any benefit-cost comparison.

31

Synthetic Discount Rates

Any public expenditure on environmental controls would not occur simply at the expense ofprivate investment; it is more reasonable to suppose that it would be at the cost of some privateinvestment and some consumption. If so, a "synthetic" rate of the form:

s = wp.dp + w,.d,

would be appropriate. If it could be assumed that the weights for marginal investments are the sameas those for existing expenditures, then the shares of consumption and investment in national incomecould be used. If a long term consumption growth rate of 1.5% is used, together with a = 0 and IA

= 1, and dp = 7%, then a typical synthetic rate for an industrialized country, with 85% investmentshare in national income, would be around 2.3%. It is difficult to argue that it could be any lowerthan this.

Conclusions on Discounting

There are two broad options for accommodating the distant nature of the effects of globalwarming and other environmental costs. The first requires that some intergenerational criterion ofsustainability be imposed, leaving the conventional discount rate unmodified as a means of allocatingresources within a generation. The second involves seeking some quantitative adjustment to theconventional discount rate. The problem with the first approach is that, as yet, few specific rules forpractical operation have emerged. Indeed, it may be that there is no requirement for special rules:each concerned individual simply argues for a "fairer" allocation of resources to the future. Theproblem with the second approach is that it takes fairly heroic assumptions to make a guantitativeadjustment that "is other than arbitrary. In terms of real-world conclusions, the "discounting problem"is not resolved either way. If discount rates above 1-2% are used, an issue such as global warming isvery unlikely to be seen as significant. Future generations would simply have to bear the costs itwould impose. Rates of perhaps 2% M be justified if utility discounting is rejected as unethical(which seems valid given that the whole idea is to account for intergenerational equity), if opportunitycost discounting is ignored, and if specific restrictions are placed on the nature of the income-utilityfunction. Use of the opportunity cost rate alone does not appear justified, so that the appropriaterange of estimates appears to be perhaps 2-5%.

VI. Valuation in Practice

1. Setting Priorities

The economic valuation cf environmental change and natural resources is a comparativelynew activity for the developing world. Much valuation relies for its credibility on the existence ofwell-functioning property, goods and labor markets. In so far as these markets operate given theextensive government intervention in the developing world, the scope for valuation is more limitedthan in the developed world. This fact has not inhibited valuation studies in Eastern Europe where,perhaps surprisingly, valuations of the economic damage from pollution, especially air pollution, havebeen carried out reasonably regularly. But the problems of credibility are substantial; damage by acidrain to buildings, for example, has been estimated to be as high as $1.8 billion in Poland, or around2.7% of GNP. Such massive losses, if true, would justify significant expenditures on pollution control,regardless of any impact of air pollution on human health, crops and forests. But the methodologies

32

used to arrive at such figures are primitive. Moreover, the prices used to value increased repair andrebuilding due to foreshortened building life, are administered prices. The theory of valuation,however, requires the use of domestic market-clearing or border prices. As a result it is difficult toassess the reliability of such estimates, making their policy relevance very doubtful.

Nor are valuation exercises in the developed world sufficiently advanced to give many insightsinto the setting of overall environmental priorities for the developing world. Few studies havecompared benefits and costs for a single environmental medium (air, water etc) and fewer still havecompared different media. As a result, it is difficult to say, for example, whether $1 in Europe isbetter spent on controlling air or water pollution. Even if this information were available, itsimplications for the developing world would not be readily transferable.

From the limited information available, the following speculative conclusions may be derived.First, it seems reasonable to focus environmental policy on two broad targets: increasing net gainsto GNP, and improving human health. Methods of raising GNP through environmental conservationand improvement need to be developed. Wildlife conservation may pay handsomely if associated withtourist revenues. Conserving tropical forests may involve some "development" sacrifices, but couldbe rewarded if international resource transfers were to compensate for losses incurred: investmentby the Global Environmental Facility would be an example. The focus on GNP ought to becontingent upon avoiding significant irreversabilities. That is, a GNP gain should not be sought atthe cost of major environmental damage that is irreversible and unrecorded in conventional GNP.

Second, and focusing on GNP gains, investment in soil conservation and forms of afforestationwould appear to have potentially high rates of return. Broad-brush calculations on soil conservationby the FAO suggest that unchecked erosion could cost some 19% or Asian, African andCentral/South American crop output between 1975 and 2000. Some individual country studies ofdamage done produce very high estimates of damage. In Zimbabwe, for example, valuing nutrientlosses from soil erosion in terms of the artificial fertilizers to replace them produced a staggering $1.5billion estimate for 1985, one third of Zimbabwe's GNP. Clearly, the policy implication to be derivedfrom this is not that GNP would rise by one-third in the absence of soil erosion. An anti-erosionpolicy would clearly costs significant resources itself. But the magritudes are indicative of the kindsof gains to be had. Economic rates of return to soil conservation are also often high, although finaljudgement requires a far more substantial body of literature relating to the developing world ratherthan the well-buffered soils of some developed economies, and careful accounting for all costs andbenefits.

In cases of soil conservation or afforestation much of the return is likely to be in the formof damage avoided rather than visible net gains in the form of increased production. This presentsa problem of perception for farmers and others: investment designed to maintain economic activitywill tend to appear relatively unattractive until the dramatic consequences of failing to preventdamage actually arise.

To some extent, the expectation that the damage resulting from soil erosion and biomass losswill be high may reflect the fact that, in a limited literature, these areas have actually been studied.By comparison, the economic rate of return to improved water quality in the developing world ismuch under-researched. The number of work-days lost from water-borne diseases in Africa, Asia andLatin America, for example, may have totalled some 250 billion in the late 1970s. At just 50 centsper day this would amount to a staggering $125 billion lost output, around 10% of gross product forthe relevant regions. In the developing world, high rates of return to water quality investment wouldalmost certainly exceed rates of return to air pollution control. In the developed world, wheredrinking water quality largely precludes the presence of waterborne diseases, the balance might switchback towards air pollution control. In truth, however, valuation studies have not progressed farenough to underscore these conclusions.

33

2. Identifying WIllingness to Pay for Conserving Envi onmental Assets

The notion of economic valuation rests on the concept of willingness to pay. The range oftechniques available for eliciting willingness to pay is fairly wide (see the attached Ainex), althoughtheir application in the developing world is very recent.

(i) Yaluing ProtedAreas

Protecting wilderness areas tends to be a low national priority in many developing countries.This is especially true where the demand for land for agricultural extensification threatens protectedlands, and where domestic values for land are low relative to the "global" value afforded it byresidents of other countries. Notable examples of the former include many of the national parks inAfrica and of the latter, tropical forests, unique wetlands, coral reefs and mangrove swamps whichtend to be rich in biological diversity.

Some sense of economic value can be obtained by looking at the implied valuations in existinginternational conservation schemes. This is particularly relevant to debt-for-nature swaps wheresecondary debt is bought by a conservationist concern and then traded with the host government fora domestic currency liability and a conservation package. Box 15 shows the results of translating thesums paid in four debt-for-nature swaps into per hectare values. The other projects shown involvejoint ventures in conservation without debt-for-nature swaps.

!0L0 IMPlCiT WJL1: JNGNE& TO *PAY iFR RAINMI}REST ASSETS IN 7N-NALTRANS

-ro1e~t Pretud A - : 1e NPV Actual Trenofero PV Area ;rootetedJJ t ow -s}.7.u,..........d 9t)¢6,

: -e84i1 R,eaeNe lot lvfa :-:-0' . :::0 ' -15420- 26- ' 0 b7 . ::: ' ' ,15.42: - '.

Amazonla~ -Pa'e 1 Eu 0 - 10.00 d3' .- 7.:

-St; Pauit's Park, -Phfl f -;\ U;ne 402-- 0.4 1 037.

Santa l oa Pm1k, Costa Odora 160 302 - 430.69.aMonte Vere FN& estN COOUt R - 36. 279 : .16 .:

Oban Park,; Mtgera 25 - - - 00W 9 -1y8 ;- : t1173 50 W. 2

.- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ U- . ,w ... Y- j 1s--X:

-bioljicul diveriJ is se.n to h: vesi imall fIb ies irsi two proJects, with th or ying in:th range :400.t80Deqare::Itout.ter. Th ow v X the: e tw e proe may have arsen beause dy We -te 8 . swpiifences m ay also affet the ty' of- th assets: tu Monte vede V area for eaple>i scoudforest. And- te :y- as

.ellectjudcenta by to s to how suec'asf*i the venture is likely t- :beI

Source: JARuitenbeek Evaluatig Economic Policies for Promotine Riinfo ret ation m n Develo ontin es. P,D Thess, Uliverity ofh Lndon 199 :

More explicit valuations of protected areas are comparatively few in developing countries. Anexercise in Khao Yai national park near Bangkok in Thailand suggested recreational benefits of some10-25 million baht per year, and possible "existence" benefits of more than 120 million baht/yr. Thesemight be compared to the sum of management costs and foregone farm income - about 30 million

34

baht. Provided existence values can be "captured", for example through raising entrance charges tothe park, the analysis suggests a high return to conservation. Similar analysis of a wildlife sanctuaryin Khao Sol Dao, Thailand, where tourism is not encouraged, produced a series of 'indeterminate"values which, in prin"-iple, could be estimated with further data and resources.

(ii) Valuing the Ecological Functions of Wetlands

The world's wetlands are under threat from agricultural, residential and industrialdevelopment, and from pollution. Wetlands comprise areas of marsh, fens, mangroves, and other wetareas, usually, but not always, at the interface between aquatic and terrestrial environments. Theyaccount for some 6% of global land area. Wetlands are especially fragile ecosysten-s because theyare "open" and fed by river systems which are themselves subject to pollution and man-made changesin flow. Because their economic functions have been so poorly understood, they also tend to beregarded as relatively unimportant. But there is now a wider appreciation that wetlands aremultifunctional and that many of their unpriced functions are economically important (Box 16).

Table 2 shows some estimates of the economic values of wetland,. In themselves, they areof little interest, although they do show that wetlands have economic value and that that value is notnegligible. Of more relevance is the relationship between these economic values and the values ofalternative uses for wetlands. It is often assumed that water feeding a wetland is not serving a usefulfunction; yet, as Table 2 indicates, natural wetlands serve a number of direct economic functions suchas supporting agriculture and fisheries. The draining of wetlands, therefore, although enabling boththe diversion of water resources to the irrigation of adjacent areas and the reclamation of wetlandsoils, should be debited with the foregone benefits of the natural system. In the case of the Hadeja-Jama'are floodplains of Northern Nigeria, for example, it has been possible to show that even apartial valuation of natural functions reveals that it is better to employ the wetland as an agricultural,fishery and fuelwood supply system than to dam its feeder rivers for other purposes. A useful wayof presenting such findings is in terms of the net economic value per cubic meter of water suppliedto the wetland system. In the Hadeja-Jama'are case the resulting comparison showed net benefitsof $45 per 1000 m3 of water flow for the natural system, but only 4 cnts per 1000 m3 for an existingdiversion of water through the building of a dam. A similar analysis of Ichkeul National park inTunisia, which is also threatened by dams, showed fishery and grazing benefits of $134 per 1000 m3

of water compared to negativ returns for the diversionary use (see Thomas et al, 1990). It cannotalways be assumed that there is profit in nature, nor that, when there is, it wIll exceed man-madealternatives, but the evidence is sufficient to show that the alternative mistake of assuming naturalsystems to have low economic value, is a serious one.

35

.

-4~1 a U JITOGICALPNCTIONSO F .WETLANDS

emp and " * 'fYitkCttoev4t Wetland Tweatistow oyf*ing nd otw'egew fourtho 0 W)ct0lyinpo~l water fliThy siiprovemenz:

atwape and iechar 1bncon

Wioviioa of a 4eXy n m neiam to thc .0IeaF -All eteet po.ibjy (> fflo w:t. ot toa toiiB Q

tidal s*irges and winds

Shoie$ine gmnobdn (oatatand :rtvede) - . -. a;b.Z ai- : : :;:: -- F

prv4i1* -ue a.na t eeodw -:. t -:- - : --.

(carboh sinks,et6

Food We.9bX support AUKliS.-Se:.-: to .a~y~ d :: - : - .:

Oilier(n14es + wlhil. habitts. la 4e.p.l *sea* AI.* . . yin d eg ra e (b) Inland salkie mardi..~~~~~~~~~~~~~~~~~~~~~~~ m

C0P;f, Mot! ' ' -0 .,W , ,,-,,O W,ff0

(), e.ru, Swamp ;' o' ' g .. ..... .. ' "'h...' '

',' im a ''''"'.."" " " ''f .....I...............

,.,,, f; t' ' 0,') x 0,,,' ;'sE'' sL A.,,, fE ,0 g . VL;'' ,.''V.' .. r. i. W.. .t.f~.. .......

.(b) NWooe wm

"~) Wet mea , 'ta .a'nd' " ., *i oter riparian hab'ats'

.0) Mang. roe wzp-,: Tmidal es,iw,ater''Miarses' . .: , - - - --Ntoois vais likely t - veb y s fio tmlw high-mking o*rdevwt.andk.

-Soue ,RX.Tut Mauket "ad In_vifionFa-ibWu in th Maemt of Wland , cof owUniversity of ast Angla, Norwici, UX; 1990, mlmn.-:

36

Tablb 2 Economic Values for Wetlands Functions

Area Source of Value Valuation Per acre (1990 $ prices. 8% discount rates)

Louisiana" CommerciaL Fishery 400Fur Trapping 190Recreation 57Storm Protection 2,400

Total 3,047

Louisiana(2 Recreation 103

Charles River, Mass.'3 Recreation 3,400Water supply 80,000

Total 83,400

Hadejia-Jama'are Agriculture 41Floodplain, Nigeria'41 Fishing 15

Fuelwood 7

Total 63Mangrove: Tt6

Trinidad Mainly fisheries 15,000Fiji 11,000Puerto Rico 13,000

(1) R.Costanza, S.Farber and J.Maxwell, "Valuatlon and Management of Wetland Ecosystems", Ecotogical Economics,Vol.1, No.4, December 1989, 335-362;(2) J.Bergstrom, J.Stoll, J.Titre, V.Wright, "Economic Value of Wetlands-Based Recreation", EcologicalEconomics, Vol.2, No.2, June 1990, 129-148.(3) F.Thibodeau and B.Ostro, "An Economic Analysis of Wetland Protection", Journal of Environmental Management,12 (1), January 1981;(4) E.Barbier, W.Adams, K.Kimnage, Economic valuation of Wetland Benefits: the Hadeia-Jama'are Floodolain.Nigeria, London Envirorunental Economics Centre, Paper 91-02, London, 1991;(5) Handbook for Mangrove Area Management, Section IV.

(iii) Valuing Preferences for Peace and Ouiet

Noise nuisance afflicts all societies both in the workplace and in the open, where the maincauses are traffic and, in the richer world, aircraft noise. Attempts to value people's preferences forpeace and quiet have centered on the use of the hedonic price approach (see Annex), in which thedeterminants of house prices are analyzed. A residential property price will vary with thecharacteristics of the property - its location, size, neighborhood, nearness to business districts andshopping, and so on. In this way the house is seen more as a "bundle of attributes" than as bricksand mortar. By statistically analyzing the prices of different properties according to their attributes,it is possible to separate out the factors that influence prices, factors that will include the local noiselevel.

Table 3 shows the results of various studies of the relationshiru between noise levels and houseprices. They are presented in terms of a price elasticity - i.e. for each unit change in the noise level,measured in standard noise units, the percentage change in property price is shown. For aircraftnoise the estimates suggest that for every unit change in NEF (noise exposure forecast), propertyprices might change by around 1%, and for every unit change in NNI (noise and number index), thechange is around 0.5%. For traffic noise, measured in Leq, a one unit change again producesproperty price depreciation of 0.5-1.0%. Clearly, using property price changes to measurepreferences for reducing noise nuisance does not encompass all the benefits of noise reduction. Highand continuous levels of noise are probably associated with health impairment (through stress, for

37

t~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

example). It is unlikely that individuals will be sufficiently aware of health risks to capture their valuein the form of house location choice. Never the less, the hedonic property price approach offers areasonable approach to the valuation of the dominant benefit of noise reduction - reduced irritationand nuisance.

TablI 3 The Value of Reducina Noise Nuisance

Studv; Imoact of 1 Unit Chanae in

NEF NNI Le9Airoraft Noise: Traffic Noise:

USA USALos Angeles 0.8 N.Virginia 0.1Englewood 0.8 Tidewater 0.1New York 1.6-2.0 N.Springfield 0.2-0.5Nimeapolis 0.6 Towson 0.5San Francisco 0.5 Washington DC 0.9Boston 0.8 Kingsgate 0.5Washington DC 1.0 North King County 0.3Dallas 0.6-0.8 Spokane 0.1Rochester 0.6-0.7 Chicago 0.7

Canada CanadaToronto 0.2-0.6 Toronto 1.0Ednonton 0.1-1.6*

SwitzerlandUK BaseL 1.3Heathrow 0.2-0.3Manchester 0.0 Norway

Oslo 0.8*Australia

Sydney 0.0-0.4 Average 0.5

SwitzerlandBasel 0.2

NetherlandsAmsterdam 0.3-0.5

NorwayBode 1.0 (per Db)

Average: 0.6-1.3 0.2-0.5

Sources: OECD, Envirorvuental Polict Benefits: Monetary Valuation, OECD, Paris, 1989; J.Nelson, "Airports andProperty Values: a Survey of Recent Evidence", Journal of Transport Economics and Policy, XIV, 1980, 37-52;J.Nelson, "Highway Noise and Property Values: a Survey of Recent Evidence", Journal of Transport Economics andPolicy, XVI, 1982, 117-130; S.Navrud, "Norway", Ch.5 of J.Ph Barde and D.U.Pearce, Valuing the Environent,Earthscan, 1991.

(iv) VaYuing Preferences for Unique Habitat

The "existence" value component of total economic value can be important, particularly where theobject of valuation is unique (as with the Grand Canyon - see Box 4) or, if not unique, the subjectof extensive familiarity to people some distance from the asset. The Kakadu Conservation Zone innorthern Australia is a 50km square area surrounded by the 20,000 square kilometer Kakadu NationalPark. The Park is visited by over 200,000 people every year and has outstanding scenery, wildlife,wetlands and Aboriginal archaeological sites. When mining operations threatened to disrupt theConservation Zone, Australia's Resource Assessment Commission determined to elicit economicvalues for the Zone in order to compare them to the benefits of mining development. The approach

38

used was contingent valuation (see Annex), whereby questionnaire respondents reveal theirwillingness to pay to conserve the area. The resulting "market" is hypothetical and the consequentproblem with CVM is to test for "hypothetical bias" - i.e. the extent to which answers given tohypothetical questions would be borne out if there were a "real"market in conservation. Part of thisbias-minimization process involves asking "discrete choice" questions in which respondents answeryes/no to specific questions about willingness to pay, rather than answering questions about what theirwillingness to pay is.

The Kakadu CVM produced the following results:

Type of Mining Valuation: $A p.yr for 10 yearsImpact National sampie Northern Territory sample

Major $A 124-143 $A 7-35

Minor $A 53-80 $A 14-33

with the analysts showing a preference for the lower end of the range, so that valuations are between$A 50-120 per year for the national sample, according to whether the mining development wouldhave a minor or major impact, and between $7-14 for the Northern Territory sample. Extrapolatedto the whole Australian population, the total willingness to pay to conserve the area against miningranges from $650 million to $1750 million, greatly in excess of the net benefits from mining (seeImber et al, 1991).

CVM is controversial partly because of its use of hypothetical questions, but also because itis the only valuation technique capable of capturing the option and existence value components oftotal economic value. No attempt was made in the Kakadu study to separate out the componentparts of value, but it is clear that much of the stated willingness to pay was made on behalf of peoplewho were very unlikely to visit the area. How far the valuations recorded would be validated if therewere a real market in the conservation of the Kakadu Conservation zone is unknown. There aresome reasons for supposing that so-called "framing bias" arises in highly targeted valuation studies ofthis kind: individuals state a willingness to pay for a single purpose without reference to the manyalternative uses of that money they indicate a willingness to pay. Some commentators feel thatframing bias is particularly relevant when it comes to valuing endangered species.

(v) Valuing Preferences for the Conservation of Endangered Species

Contingent valuation techniques currently provide the only available technique for eliciting preferencevaluations for environmental assets that have no related market. Endangered species provide onesuch example. The problem with CVM is that because the market is created experimentally - throughthe use of interviews and questionnaires - there is no obvious way to valida the estimatedwillingness to pay (WTP) for conservation. A great deal of the CVM literature is thereforeconcerned with procedures for validation (see Annex 2). Broadly speaking, validation tests include(a) checking the CVM results against other valuation techniques (usually the travel cost method - seeAnnex 2), (b) checking fir biases in responses to the questionnaire, and (c) checking, where possible,against actual market-revealed willingness to pay.

A unique virtue o the CVM approach is that it can capture existence and option values; allother valuation techniques focus on use values. Table 4 shows the results of CVM studies for

39

endangered or rare species, and for highly valued ecosystems. The various estimates have beenconverted to per person WTP in 1990 prices. The data are interesting because of their broad

Table 4 Per Cafita Preference Valuations for Endangered Species and Prized Habitats

Species Valuation Habitat Valuation

Norway: brown bear, wolf 15.0 conservation of rivers against 59.0-107.0and wolverine hydroelectric development

USA: bald eagle 12.4 Grand Canyon (visibility) 27.0emerald shiner 4.5 CoLorado wilderness 9.3-21.2grizzly bear 18.5bighorn sheep 8.6whooping crane 1.2blue whaLe 9.3bottlenose dolphin 7.0california sea otter 8.1Northern elephant seal 8.1humpback whales' 40-48 (49-64)

Australia: Nadgee Nature Reserve, NSW 28.1Kakadu Conservation Zone, NT 2 40.0 (93.0)

UK: nature reserves3 40.0

Notes: (1) respondents divided into two groups one of which was given video information; this group indicatedhigher valuations; (2) two scenarios of mining development damage were given to respondents; the major damagescenario produced higher valuations for the habitat then the minor damage scenario; (3) survey of informedindividuals only.

Sources: Norway - L.Dahle et al., "Attitudes Towards and Willingness to pay For Brown Bear, Wolverine and Wolfin Norway", Department of Forest Economics, Agricultural university of Norway, Report 5/1987, (in Norwegian);A.Hervik et al., "Implicit Costs and Willingness to pay for Development of Water Resources", in A.Carlsen (ed),Proceedings of UNESCO S mosium on Decision Making in Water Resources Plannina, May 1986, Oslo; USA: K.Boyleand R.Bishop, "The Total value of Wildlife Resources: Conceptual and Empirical Issues", Paper presented toAssociation of Envirornmental and Resource Economists, Boulder, May 1985; D.Brookshire et al., "Estimating OptionPrices and Existence Values for Wildlife Resources", Land Economics, 59, 1983; R.Stoll and L.Johnson, "Conceptsof Value, Non-market Valuation, and the Case of the Whooping Crane", Department of Agricultural Economics, TexasA&M University, 1984; R.Hageman, "Valuing Marine Mammal Populations: Benefit Valuations in a Multi-SpeciesEcosystem", National Marine Fisheries Service, Southwest Fisheries Center, Report LJ-85-22, La Jolla,California, 1985; K.Samples et al., "Information Disclosure and Endangered Species Valuation", Land Economics,62, No.3, 1986; W.Schulze et al., "Economic benefits of Preserving Visibility in the National Parklands of theSouthwest", Natural Resources Journal, 23 (1983); R.Walsh et al., "Valuing Option, Existence and Bequest Demandsfor Wilderness", Land Economics, Vol.60, No.1, 1984; Australia - D.Imber et al., A Continsent Valuation Surveyof the Kakadu Conservation Zone, Resource Assessment Comfission, Research Paper No.3, Canberra, February 1991;J.Bennett, "Using Direct Questioning to Value Existence Benefits of Preserved Natural Areas", School of BusinessStudies, Darling Downs Institute of Education, Toowoomba, 1982; United Kingdom - K.Willis and J.Benson,"Valuation of Wildlife: A Case Study on the upper Teeside Site of Special Scientific Interest and Comparisonof Methods in Enviromental Economics", in R.K.Turner (ed), Sustainable Environmental Management, BelhavenPress, London, 1988.

consistency. Valuations of preferences for species conservation, for example, cluster around $9 if therelatively high value for humpback whales is excluded, and $13 if included. The range is $1-18excluding humpback whales and $1-48 including humpback whales (see note to Table 4). For prizedhabitat, the range is $9-107 per person per year. While a great deal more work is needed in this area,the results are suggestive in that (a) they do not represent large proportions of respondents' income,and (b) habitat appears more highly valued than species which, given the role that habitatconservation would play in species conservation, is a difference one would expect: a wider array ofbenefits is secured through conservation of habitat than through targeting species. Clearly,framing bias presents some problem. The sum of the species valuations in the USA, for example, is

40

much higher than average personal contributions to conservation societies, although this may reflectthe "free rider" phenomenon (many who value the environment do not pay because they know otherswill). The international comparison of per capita values is also problematic. There are no particularreasons to suppose that "unit values" of this kind would be the same across countries or even acrossdifferent regions of the same country. But where there are reasons to suppose that environmentalawareness exists on approximately the same scale, which is testable by opinion polling, then, allowingfor variations in income, one might expect similar valuations. As yet little work has been done to testthis "transferability" of values.

(vi) Willingness to Pay for Rural Water Supplies

Valuation techniques have also been applied to the more immediate human environment -notably water supply and sanitation. Traditionally, water supply investments have been evaluated byrules of thumb related to an assumed willingness - to - pay for basic services. Since the service isusually supplied to the poor, the assumption has been that only the most basic provision - public tapsand hand pumps - is warranted; no-one is willing to pay for better, more elaborate services. This"basic needs" philosophy would be satisfactory if resulting public supplies were reliable. Yet at anygiven point in time, perhaps one in four public supply systems are not working, while use rates ofthose that do work are low - only one-third of people connected to public supply systems in Coted'Ivoire and Kenya actually use them. Since the benefits of such systems, in terms of public healthand time-saving, are clearly substantial, it is worth estimating households' true willingness to pay.

In the absence of real markets, the challenge is to find the underlying demand for the service.In terms of time saving, one approach is to observe how people choose between alternative sourcesof supply. In Ukundu, Kenya, villagers can choose between water supplied by vendors who visit theirhouse, water sold at "kiosks" in the village, and water from the well (Mu, 1989). In terms ofcollection time , and relative to use of the well, house delivery saves the most collection time andcollecting from kiosks the least amount of time. In terms of expenditure, household vending coststhe most, then kiosk water, with well water being the cheapest. By looking at actual choices, thetrade-off between money and time can be determined. Time-saving is one of the benefits of watersupply improvement; in this case, if water quality is invariant between sources, time-savings willgenerally define total benefits. The Ukundu study found that users of vendors and kiosks wererevealing high WTP for time-savings, of the order of 8% of their incomes.

Another study in Brazil, using the contingent valuation approach (see Annex), asked thosesurveyed: "If you are required to pay X, would you connect to the new supply or use an alternativesupply?" Three different areas were surveyed, some with improved services available, to whichhouseholds might or might not be connected, and some without. Some of these latter areas hadservices planned (with an announced tarift), others expected a service but did not know of what kindor what the tariff would be. From the survey the probabilities of being connected were estimatedand found to behave as predicted. The higher the price and the greater the distance to the source,the less likely a household was to be connected. WTP estimates were also obtained from thequestionnaires. The results provided not just an estimate of the average WTP, but also indicated howhouseholds would respond to higher prices, an important consideration if revenue-raising is a concern.Maximum WTP for a yard tap was around 2.5 times the prevailing tariff and some 2.3% of familyincome. Some "strategic bias" - deliberate under-reporting of WTP - was probably present (seeAnnex); true WTP was probably higher than this. Equity considerations could be taken care of byproviding relatively high-priced services to the better-off, and by using revenues to cross-subsidize thepoor's need for free public taps.

41

(vii) The Benefits of Improved Sanitation

As urban populations continue to grow rapidly, sanitation needs in developing countries willbecome a greater and greater burden on public revenues. In 1950, less than 300 million people livedin developing country urban areas. Today the figure is over 1,300 million. By 2000 it will be 1.9billion; by then there will be 200 cities with populations over 1 million people, of which 150 will bein developing countries. The cost of the necessary infrastructure for this urban development isenormous. As with water supply generally, sanitation systems tend to be primitive for the poor, whilemiddle and upper income classes benefit from systems that are both subsidized and less primitive.Willingness-to-pay is generally assumed rather than estimated; charges above 3 per cent of householdincomes are thought not to be affordable.

In Kumasi, Ghana, WTP was estimated through a contingent valuation approach. The optionswere water closets with a piped sewerage system and ventilated pit latrines ("KVIPs"). The latterrepresent a far cheaper option for sanitation than connecting sewers and installing water closets.Some households already had water connections and could therefore be asked their WTP for a watercloset and a KVIP. Households with water closets could be asked how much they would be WTPfor a connection to the sewer, and so on. KVIPs can operate without water connections. The resultsshowed that households without water closets were WTP roughly the same sum for a WC or a KVIP.In terms of WTP for KVIPs, households with bucket latrines bid the lowest price; those using publiclatrines bid significantly higher prices (around 30-35% more), reflecting the inconvenience and lackof privacy of the public system. Overall mean bids of around $1.5 per month compare to averageexisting expenditures of about $0.5 per month. Comparing WTP with the costs of provision of KVIPsand WCs, WTP was found to be -Is than costs of supply. Given that sanitation systems yieldextensive external benefits in the form of public health )although these benefits were not measuredin the study), a subsidy would presumably be justified. The study showed that the required subsidyfor a WC system for Kumasi would amount to some $60 million. The required overall subsidy forthe KVIP system would amount to some $4 million (Whittington et al ,1991).

(viii) Valuing the Benefits of Fuelwood Planting

In the developing world, wood still accounts for the major part of energy consumption.Planting trees for fuelwood is thus an inherently valuable activity, but how valuable exactly? Sincesuch a large quantity of fuelwood is collected, rather than purchased in the marketplace, there areno market prices at which to value the commodity. Moreover, growing trees yield benefits besidesfuelwood; trees provide poles for building, leaves for fodder, protection for crops, and so on.Economic valuation techniques are therefore essential if the benefits of investing in tree growing areto be demonstrated.

Typical approaches to valuing fuelwood benefits involve estimating what other source ofenergy would be used if increased fuelwood were to be unavailable. This might involve supplies ofkerosene, coal if available, and cow dung. For kerosene or coal, market prices are available. Cowdung may also be marketed, but this will typically be the case where fuelwood is also marketed (i.e.in conditions of considerable scarcity), so that fuelwood market prices are then available. The valueof non-marketed cow dung can be estimated by looking at the responsiveness of crops to cow dungas a fertilizer and soil conditioner. The market value of the crops then provides the relevant link tomarket values. Care has to be taken that the predicted substitution is credible. In Korea somefuelwood investments have been justified on the grounds that the alternative fuel would be coal. Inthe event, the fuelwood did not displace coal; rather, coal displaced fuelwood in rural areas. As with

42

all project appraisal, predicting tastes and preferences can be hazardous.Estimating fuelwood's value relative to cow dung involves the following procedure. First, find

the energy content of fuelwood and dung. Second, estimate the weight of fuelwood in a cubic meter.Third, compute the "dung equivalent" of I cubic meter of fuelwood by multiplying the weight by theratio of the energy values of fuelwood and dung. Fourth, estimate the amount of manure that agiven amount of dung produces, so that the cubic meter of fuelwood can be expressed in "manureequivalent". Fifth, estimate the crop yield response to this amount of manure and the monetary valueof this yield increase. Sixth, normalize the value of the yield increase per cubic meter of fuelwood:this is then the economic value, or shadow price, of the fuelwood.

The validity of the final result is crucially dependent upon the crop yield response estimates.As with so much economic valuation, it is not the economic stages in the process that give rise to theproblem, but the underlying "production function" - i.e. the links between the environmental andeconomic variables. Approaches similar to the "dung-equivalent" method have been used in a numberof economic appraisals and in estimating soil erosion damage. Sometimes costs and benefits areestimated in terms of chemical equivalents. For example, instead of the "manure" equivalent, it ispossible to estimate how much commercial fertilizer would be required to compensate for thatquantity of cow dung diverted from manure to fuel use because of fuelwood scarcity. This was theapproach used in a World Bank study of afforestation in Ethiopia (Newconibe, 1989). Suchapproaches do not capture all the benefits of fuelwood since the chemical nutrient status of dung isonly part of its value as a manure. However, the procedure does reveal that environmental costs andbenefits invariably have an analogue somewhere in the private market system. The challenge forvaluation is to make that link and translate available market values back to the environmental asset.

Fuelwood investment yields other benefits. The proximity of the wood to the point of usemeans that valuable labor time is saved. Past studies have typically valued the saved time at theruling wage rate if there is no surplus labor, and at the minimum wage where there is. Strictly,neither approach is correct in terms of the criterion of willingness-to-pay; what is required is somevaluation based on actual choices, as with the Ukundu study above.

Trees also provide leaf fodder for animals and this is often included in project evaluations.Again, if fodder is not actually marketed, a "production function" link can be made to marketedoutputs by estimating the effects of increased fodder on livestock weight, and hence the market valueof livestock. Care has to be taken to incorporate full costs; if trees are grown especially for fodder,then the loss of output from the existing use of the land has to be deducted. Grass yields foregone,for example, would be deducted from tree fodder yields.

Trees may also be important as inhibitors of "desertification" (which should be understood asgeneral land degradation rather than as the more popular and unwarranted concept of "spreadingdeserts"). All tree planting tends to reduce pressure on naturally forested land. One method ofvaluing the resultant gains is to estimate the fuelwood yield from plantations (X) and compare it toyields from natural forest areas (Y). Each hectare of plantation can then be said to "save" X/Yhectares of natural forest land. The "avoided damage" can then be accorded a monetary value; byprojecting the rate of soil erosion on cleared natural forest land, the latter's productivity decline canalso be projected. (After some threshold year, all crop and livestock production would be lost.) Bycalcuiating the present value of lost output, a surrogate value for the benefits of planting trees isobtained.

The fuelwood valuation issue reveals several important points. First, valuation ji possible.Second, the underlying ecological interlinkages are the vital element in the valuation process; thehighest rewards are likely to be obtained from expanding our knowledge of these interdependencies.Third, it is essential to look at all potential benefits. In the event some may turn out not to beimportant, but the fodder and anti-desertification examples show that some of the "incidental" benefits

43

could be substantial.

(ix) Valuing the Benefits of Biological Diversity

Probably the greatest challenge for economic valuation is to derive values for people'spreferences regarding biological diversity. "Biodiversity" is frequently used as a shorthand for boththe guantity and range of species, and equally frequently as a catch-all phrase for wildlife and habitat.

The African Elephant

Kenya is visited by about 250,000 foreign adult tourists every year. Safaris are the main focusof this tourism, accounting for an expenditure of some $200 million p.a. in Kenya (and for perhapstwice that for the visits overall, since much of the generated income accrues to the industry in thetourists' countries of origin). Until a recent ban on the ivory trade, the Kenyan elephant wasdisappearing very rapidly. From 65,000 elephants in 1981, numbers were probably down to 16,000by the end of the decade. An analysis of expenditures by tourists (the travel cost apgroach - seeAnnex) and a contingent valuation assessment together suggested that tourists would be willing topay an extra $25 million p.a to ensure they saw elephants during their stay. Points of comparison arethat (a) this represents at least a 10% increase in actual expenditures, and (b) it is substantially higherthan even the peak value of (largely illegal) ivory exports in 1979 at $3 million, and higher still thanthe estimated 1988 value of only $17,000 (Brown and Hall, 1990). In policy terms it suggests thatcountries with significant wildlife resources and a demand by tourists to see them, could extract someof the "rent" tourists obtain.

Rirds

Few studies exist of the economic significance of birds. One Canadian study looked at thedirect benefits from recreational and other activities associated with birds (Jacquemot and Filion,1987). Over 100,000 people were surveyed in order to determine their actual participation in bird-related activities and their willingness to pay for participation. Expenditures by participants amountedto $C 1.9 billion (1986 C$) and incremental benefits (the excess of WTP over actual costs) was some$C 350 million. For all wildlife (birds and mammals) the total net benefit was $780 million p.a. Birdsthus accounted for around 45% of all net benefits derived from wildlife-related activities.Expenditure on bird-related activities, which results in direct income and employment to others,accounted for some $2.4 billion of Canadian GDP, half of which was accounted for by non-consumptive activities (i.e. birdwatching), and for $C 870 million of government revenues. Protectionof a single species often results in significant gains from recreational viewing. Canada's "capistrano"(the Pembroke swallow) was protected in 1983. The mass flocking of these birds produces a spectaclemuch appreciated byrecreationists. Estimated net benefits, based on the travel cost approach (Annex2) were some $C 0.5 million p.a.(Clark, 1987).

Ecotourism

The travel cost method (see Annex) has been applied to the valuation that visitors place onthe Monteverde Cloud Forest Biological Reserve in Costa Rica (Tobias and Mendelsohn, 1991). Thereserve is ma.nly virgin rain forest with, despite difficult access, a growing tourist demand. Domesticvisitors were sampled to find their area of origin, and the distances they had travelled were calculated.Distance was converted to currency using an average cost per kilometer of $US 0.15. A demand

44

function was then estimated l-k4ig visits to cost of travel, population density and a measure ofliteracy in each of the areas of .igin. The expected links were: (a) the higher the cost, the lowerthe visit rate; (b) the higher the pupulation density, the higher the visit rate (low density populationswould being more likely to have their own forest areas to visit); and (c) the higher the literacy rate(and hence the higher permanent income), the higher will be the visit rate. This is indeed what wasfound; estimated visits were found to correspond to actual visits. From the demand function it waspossible to estimate consumer surplus, the excess of willingness-to-pay over the actual cost of travel.Expressed as a present value, the sum of these valuations was $US 2.4-2.9 million for this one site,or around $35 per visit. Value per year was some $100,000. Further, this figure exgludes foreignvisitors who outnumbered domestic visitors by 4:1 in 1988. Assuming a similar per capita valuation,this imply a mean present value of $2.5-10 million, or some $1250 per hectare. New land can bebought for $30-100 per hectare, suggesting that expanding the reserve to allow for more recreationwould be a worthwhile investment.

The Economic Value of Plant-Based Pharmaceuticals

No-one is sure just how many species there are. A probable number for higher' plant species,which are widely used as bases for pharmaceutical drugs, is some 500,000, counting both known andunknown species. Rates of extinction are positive but also unknown. Perhaps 10% or more of thesespecies will be extinct by the end of the century. Between 65% and 75% of all the higher plantspecies are indigenous to tropical moist forests. Hence loss of rainforest means losing potentialsources of future pharmaceuticals; existing sources are likely to be protected through replication andsynthesizing of materials.

What is the economic value of these plants? To date, valuation has been fairly speculativebut iliustrative of the orders of magnitude involved. There are several ways in which to approachvaluation:

- by looking at the actual market value of the plants when traded;* by looking at the market value of the drugs for which they are the source material;* by looking at the value of *he drugs in terms of their life-saving properties, and using avalue of a "statistical life".

Table 5 summarizes estimates. The method of valuation is important because it affects thesize of the estimate significantly. Valuation based on life-saving properties gives the highestestimates, assuming the value of a statistical life to be $1 million - a figure representative of actualvalues used in life-saving investments in a number of countries. Market values of plant-based drugsgive lower values, and actual traded prices for plant material give the lowest value of all. Of coursethe price of drugs reflects many factors other than the cost of the plant source material. In thatrespect, the drug price grossly overstates the value of the plant. Equally, market prices understatetrue willingness to pay for drugs: there will be individuals who are willing to pay more than themarket price for a given drug. Indeed, since the evidence suggests that such drugs tend to be priceinelastic, consumer surplus could be substantial. While there is no empirical basis for supposing thatconsumer surplus exactly offsets the overstatement in the price estimate, the two factors do work inopposite directions. The price of plant-based drugs is therefore a relevant indicator of the generalscale of the benefits of medicinal plants.

In the 1980s only about 40 plant species accounted for all plant-based prescribed drugs soldin the USA. Each species was therefore responsible on average for an expenditure of some $200

45

million. Clearly, some species were far more valuable than others, but, taking the average, it ispossible to get some idea of pharmaceutical value lost through species eradication. Extinction ratessuggest that, by the end of the century, as many as 50,000 additional species are likely to beunavailable for medical research. The probability that any given plant will produce a marketableprescription drug has been estimated at between 1O-3 and 104. This would imply that, by 2000, 25plant-based drugs will be lost from species reduction. The annual loss to the USA alone wouldtherefore be 25 x $200 million, or $5 billion, and to all OECD countries combined, perhaps $15billion. As a benchmark, it is worth noting that the GNP prou- ed in the whole of BrazilianAmazonia is some $18 billion per annum.

The results are speculative and a great deal more research is needed. But clearly thepotential value of pharmaceuticals in the developing world is very large. Thus far, nothing has beensaid about substitutes: if plant source material did not exist, then other substitutes would be available.Yet if pharmaceutical companies regard plant source material as so important why have they notpurchased large tracts of virgin forest? Option value considerations, however, argue in favor ofconserving biological diversity on at least this ground.

Table 6 The Value of Plant-Based Druas

$ billion 1985

USA OECD Countries

Market Value of Trade in Medicinal Plants 0.55 (1980) 3.91 (1981)

Market or Fixed Value of Plant-Based Drugs on Prescription -8.0 (1981) 19.0 (1981)11.0 (1985) 26.2 (1985)

Market Value of Prescrip.;on and Over-the-CounterPlant-Based Drugs 18.0 (1985) 43.0 (1985)

Value of Plant-Based Drugs Based on Avoided Deaths-anti-cancer only 30.0 (1985) 90.0 (1985)-plus non-cancer also 60.0 (1985) 120.0 (1985)

notes: Ratio of OECD to USA for prescribed drugs taken to be 2.38 based on market surveys in 1980. 'Value ofa statistical life' taken to be S1 million.

Source: adapted with modifications from P.Principe, 'The Economic Significance of Plants and their Constituentsas Drugs', in H.Wagner, H.Hikino and N.Farnsworth, Economic and Medicinal Plant Research, Vol.3, London:Academic Press, 1989, pp.1-17.

3. Comparative Economics of Environmental Conservation

Demonstrating the benefits of conservation is an essential part of the overall purpose ofeconomic valuation. Biological diversity is unlikely to be successfully conserved unless its economicvalue can be shown to be greater than alternative land uses. Some evidence is available to suggestthat conservation, in the sense of sustainable use of natural resources, is in many circumstances tobe preferred over conventional land uses (see Box 18). This evidence notwithstanding, the reasonswhy evidently superior market benefits for conservation uses are not realized in practice, remaincomplex.

46

gi2 2^, Qp, ^,^MRU^T USE AU §XRagus

Tha tablte howa estimates of th tconozti vatue of alternativo land uses for developing countries. Although the evidence is limited,it contradicts the ptwanWion that "doveloputent' I -always batter than "consenrvton. If ntural tesources sm nanaged wiely andpea Be anlowed ti ¢extise fMd Cho4es, eo tvation ftquently pays in tarm of otveaUotaslt namnial analysis,

gkgt ''YaUse of Habitat - iue (oer h^ Alternative Use VOlVO

Zlnbabe4 - Wildtife production z$ 4.2 cattte ranchin 2$ 3.6

altaysia Forest production $ 2455 Intensive asriculture $ 217

PerU: Forest production $ 6820 Ctear-feltlng timber $ .1000

Sou*rces T.S= Onso, "The Economics of Katural Habitat UtitIsation: a Suvey of -the Literature andissues", London Enviromentat Economics Centre, London, 1991, -inga.

-II.Valuation and Global Environmental Problems

Valuing damage from global warming extends economic techniques into controversial anduncertain areas. In the first place, the sheer scale of global environmental issues is likely to makethe credibility of damage estimates suspect. Second, some global problems - such as global warming -may produce "non-marginal" changes in wellbeing; valuation techniques, in contrast, have been

developed for comparatively small or "marginal"effects. However, there have been several attemptsto value global warming damage; these were intended to assist in setting global warming "targets" forreduced emissions of greenhouse gases.

Global warming damage is likely to show up as foregone GNP and as "non-GNP' costs.Existing studies have focused exclusively on the USA only: extending these to a wider domainobviously requires some assumptions about the transferability of US results to other economies. Theresults of one study is shown in Table 6. Estimated damage amounts to 1.1 % of GNP: expressed asa "price" of a tonne of CO 2 , damage amounts to some $9 per tonne.

The estimates shown relate to the damage done by a doubling of the concentration of CC?in the atmosphere, an outcome that, on current trends, might occur around 2030-2050. The table alsoshows guesstimates for 2250; these are substantially above those for 2050. This assumed doubling ismerely a benchmark; if nothing is done by way of prevention, warming will continue. The 2250estimates in fact correspond to a warming of 10°C.

The estimates in Table 6 are part of ongoing work on the valuation of climate change effects.They are therefore provisional. But if the damage done from a doubling of CO2 concentrations amountsto around 1% of gross world product, then it is not as dramatic as some forecasters suggest; globalwarming then becomes an appropriate area for the application of economic valuation techniques.However, the estimates make no allowance for dramatic change in the form of climatic catastrophes.Finally, the estimates shown in table 6 are present values: they have already been discounted, using adiscount rate of 1 %. As noted previously, justifying such low discount rates on conventional efficiencygrounds is probably not possible. Using 1% as a discount rate that reflects intergenerational equity isarbitrary, but reflects the state of play. At rates of discount above 1 %, the 2250 damage estimate wouldbe considerably reduced.

47

Table Sectoral Assessments of Damge from Global Warmins

(Present value, in bltlion U; 1990 $ p.a.)

2050 2250

Agriculture 17.5 95.0

Forest loss 3.3 7.0

Sea Level Rise 7.0 35.0

Electricity requirements 11.7 67.0

Non-electric space heating -1.3 -4.0

HNuman life 5.8 33.0

Hurricanes 0.8 6.4

Water supply 7.0 56.0

Urban infrastructure 0.1 0.6

Air pollution 3.5 19.8

Migration 0.5 2.8

Leisure activities 1.7 4.0

Species loss 4.0 16.0

Totals lb6 38.6

Source: W.Cline, Estimating the Benefits of Greenhouse Warmina Abatement, OECD, Paris, 1991.

VII. Concluions

Economic valuation is controversial in large part because its purpose has not been clearlyconveyed to non-economists. The purpose of valuation is to elicit measures of human preferencesfor, or against, environmental change. As a procedure, it consequently faces two immediatelimitations:

First, economic values are not the same as "intrinsic"values - values "in'things rather thanvalues "of"things. Economic valuation makes no claim to measure intrinsic values, although throughthe concept of "existence" value it may be capable of capturing human perceptions of intrinsic value.

Second, measuring preferences focuses on efficiency gains and losses from environmentalchange. It says little about the distribution of costs and benefits within a time period or between timeperiods. Within a time period, the use of efficiency gains and losses as a guide to policy or projectevaluation assumes that the prevailing distribution of income is socially acceptable, since it is thatdistribution which "weights" the measures of willingness to pay. Between time periods, the use of afurther efficiency concept - the discount rate - biases the outcomes of evaluation in favor of present,and against future, generations where costs and benefits in the future are both distant and significant.

But economic valuation is useful in several contexts. Project and program appraisal cannotbe comprehensive or adequate without it. National environmental policy priorities will be betterinformed if economic values are known with some degree of certainty. The entire objective ofsustainable development almost certainly cannot be interpreted without some idea of the value 4environmental services and assets.

Empirical work on valuation remains limited, even in the developed world. It is fairly new

48

in the developing world, although many project evaluations have used some form of indirectvaluation. Its importance for the development process is that revealed economic values forenvironmental conservation and environmentally improving projects and policies have frequently beenfound to be large. Valuation demonstrates that there is an economic case for protecting theenvironment, in addition to any ethical case, and can assist the process of better decision-making.In so doing it offers the potential of more cost-efficient public choices, thus allowing limited publicincome to be optimally spent.

49

Annex 1

ENVIRONMENTAL POLICY AS A CONSTRAINT ON ECONOMIC GROWTH

The focus of this paper has been on the valuation of environmental impacts; one justificationfor this is that environmental degradation frequently involves losses in conventionally measured GNP.But policy-makers often have the opposite concern, namely that a stronger environmental stance willbe at the expense of jobs, trade and inflation. Environmental policy is seen as a drag on economicgrowth.

1. Current Spending on the Environment

Figure Al shows OECD data o.a developed economy spending on environmental protection.Despite the very imperfect database, it suggests that OECD nations spend around 1-1.5% of theirGNP on environmental protection. Future costs are likely to be higher as the pressure to strengthenenvironmental policy grows and as international agreements expand (ozone layer protection, globalwarming, global biodiversity, tropical forest protection, toxic waste trade etc.) To gain some idea offuture costs, it is worth noting that the Netherlands National Environmental Protection Plan, one ofthe strictest in Europe, envisages spending of up to 34% of GNP.

2. The Macroeconomic Impact of Environmental Protection

A number of studies have been carried out on the costs of environmental protection inmacroeconomic terms.

United Kingdom

Barker and Lewney have simulated the macroeconomic impacts of three hypotheticalenvironmental policies: a carbon tax designed to achieve the UK's conditional target of reducing CO2

emissions to 1990 levels by 2005; a fourfold rise in industrial pollution abatement expenditures by2000; and an intensified water clean-up policy'. Various scenarios are run through the CambridgeMultisectoral Dynamic model.

Figure A2 shows the results of combining all three policies. The carbon tax has virtually nonegative macroeconomic effects because it is offset by the use of VAT reductions. This fiscalneutrality assumption is important; the pervasiveness of carbon fuels means that any tax has thepotential to raise substantial government revenues. For a carbon tax regime to be fiscally neutral,revenues would have to be partly or wholly returned to the economy both to offset deadweight lossesfrom existing tax regimes and to compensate lower income groups who would be affected adverselyby energy taxes.

The overall result of this UK study is highly reassuring for the environmentalist. GDP in 2010is reduced from base levels by less than 1%, which translates to a reduction in annual growth ratesof only 0.05%. GDP actually increases before 2010. Unemployment falls by 2005 due to rapid

I T.Barker and R.Lewney, "A Green Scenario for the UK Economy", in T.Barker (ed),Green futures for Economic Growth: Britain in 2010, Cambridge Econometrics, Cambridge,1991.

50

expansion of the pollution abatement industry; however, this result is critically dependent onassumptions about full employment in the economy - a continuing debate among macroeconomists.In the Cambridge model, full employment is not achieved before 2005.

A separate simulation of a carbon tax using the Cambridge model has been run bySondheimer2. This supposes a hypothetical tax of some 30 pounds sterling per tonne of carbon,rising at the rate of inflation and offset by changes in direct and indirect taxes to secure approximatefiscal neutrality. Like the Barker-Lewney study, Sondheimer finds very little impact on GDP - a 0.5%reduction in baseline GDP - and a reduction in unemployment of some 70,000.

Ingham and Ulph have estimated the size of the carbon tax needed to secure 20% reductionson 1990 CO2 emission levels by 2005 (stricter than the UK's stated target) and its macroeconomiceffects3. While the future tax is large (120-280% on coal, in 2005, 60-130% on oil, and 16-71% ongas), they note that:

the effects could be rather different from what is often supposed; in particular, anyloss of competitiveness will be short-lived in terms of lost output, with a longer termboost to output due to enhanced productivity effects through scrapping of oldequipment.

As a result, manufacturing investment and employment increase rather dramatically. Assuming theeconomy grows at 2% p.a.,the carbon tax increases employment in 2005 by some 30% over base, andby 21 % for a lower growth rate of 1% p.a. Some runs of the model actually raise employment byover 100% compared to the base case. It needs to be noted that this result, while similar to the otherstudies, arises without tax revenues being recycled back to the economy in the form of othercompensatory tax changes.

While they vary in sophistication, these UK studies all suggest that environmental policy hasa strong potential for increasing employment, or at least making it no worse.

Norway

Norway has a strong tradition of using general equilibrium models to simulate policy measures.The Central Bureau of Statistics has since 1986 regularly simulated the effects of environmental taxes.

Glomsrod and colleagues impose a hypothetical carbon tax designed to stabilize NorwegianCO2 emissions at their 2000 level4 . The tax rises over time, and in 2010 is approximately 100%higher than the price of fuel oil on a reference "business as usual" scenario. GDP growth is reducedfrom 2.7% p.a. to 2.3% p.a.; imports and exports show modest declines (down by 4-7% from whatthey would otherwise have been), while investment falls slightly (by around 1%). The investment

2 J.Sondheimer, "Macroeconomic Effects of a Carbon Tax", in Barker op.cit.

I A.Ingham and A.Ulph, "Carbon Taxes and the UK Manufacturing Sector", Departmentof Economics, University of Southampton, mimeo, 1990.

4 S.Glomsrod, H.Vennemo and T.Johnson, "Stabilization of Emissions of C02: AComputable General Equilibrium Assessment", Central Bureau of Statistics, Discussion PaperNo.48, April 1990, Oslo.

51

result is in complete contrast to the Ingham-Ulph result, and is caused by a change in the capital mix- from shorter-life machinery to longer-lived assets such as houses. Employment effects are notcomputed in aggregate, but man-hours in the pulp and paper sector and petroleum refining both fallsignificantly by 12-15%; in housing, construction, and textiles, they rise by similar amounts. Overall,the effects on employment appear negligible.

This study is additionally notable for its attempt to estimate the benefits of such a policy. Itsuggests that the cost of some 27 billion krone (at 1986 prices) is largely recouped by some 19.1billion krone in health benefits and reduced congestion, noise, accidents and road damage.

Netherlands

The Netherlands has produced one of the most detailed environmental policy statements ofany country - the National Environmental Policy Plans. The policy involves a doubling ofenvironmental protection expenditures as a proportion of GDP, substantial increases in energyconservation, investment in public transport, constraints on the use of private vehicles, waste recyclingand reduced fertilizer use. Figure A3 shows the results of the NEPP in terms of macroeconomicimpacts. GNP rises by 95% compared to 98% by 2010 compared to 1985 (?), a slowdown of under0.1% p.a., but may actually increase if other countries pursue similar policies. Employment isunaffected, and could even rise if other countries respond with similar policies.

The European Community

OECD's INTERLINK country forecasting model has been used to simulate the effects of theEuropean Community's Large Combustion Plants Directive which requires significant reductions insulphur and nitrogen oxides from power stations and large industrial boilers6. The investments inpollution control equipment and plant modification necessary to achieve the targets have the effectof increasing GDP and employment in the five years following implementation of the Directive.Longer term adjustment invc!ves some slight reduction in GDP and employment. The results areshown in Figure A4. Again, the overwhelming impression is that restrictions on air pollutants havevery little impact on inc)me and employment. The investment in pollution abatement equipmenttends to expand the economy. As higher pollution control costs work their way through the economy,so prices do rise to some extent and deflate the economy slightly, an effect reinforced by the cessationof the extra investment once abatement measures are complete.

5 Netherlands Ministry of Housing, Physical Planning and the Environment, NationalEnvironmental Policy Plan of the Netherlands, Amsterdam, 1989. This policy was updated by"NEPP+" in 1990 which committed further funding for energy conservation and some othermeasures.

6 G.Klaassen, P.Kee, A.Nentjes, W.Hafkamp, A.Olsthoorn, The Macroeconomic Effectsof the Large Combustion Plants Directive Proposal: Economic Aspects of Controlling Acid Rainin Europe, Institute for Environmental Studies, Free University of Amsterdam, Amsterdam,1987.

52

The USA

US studies of the effects of environmental regulation are more plentiful. Hahn and Hirdassemble estimates of the costs of regulation, both economic (e.g. regulation of trade) and social (e.g.regulation of environmental damage)'. Economic regulations tend to have very limited economicbenefits. Hahn and Hird estimate that such measures cost the US economy some $46 billion eachyear. Social regulation costs between $78 billion and $107 billion, but yields benefits of some $42-181billion. Environmental regulation dominates social regulation, accounting for some 70% of the costand some 40-75% of benefits. Hahn and Hird suggest that environmental control costs are probablyslightly in excess of benefits.

Reference to the benefits of environmental policy serves as a caution to "straight"macroeconomic impact studies. For example, environmental policy benefits health, and hence(probably) contributes to productivity. These effects need to be seen as offsetting the direct "GDPcosts" of environmental regulation.

Probably the most detailed study of environmental regulation costs in the USA is that byJorgensen and Wilcoxen, which significantly makes no allowance for health benefits8. Through along-term growth model, the study simulates the past effects of US environmental policy on economicgrowth between 1973 and 1985. The annual GDP growth rate for the period is found to have fallenby 0.19 percentage points because of the "drag" effects of environmental regulation. This figureamounts to a reduction approximately twice those suggested by non-USA studies. The differencecould reflect either the nature of US environmental regulation, or the difference in the modellingprocedure. Converted to a long-run projection, the Jorgensen-Wilcoxen model suggests that GDPmight be some 2.6% lower as a result of regulation. The model results do not indicate the effectson employment, but clearly they would tend to be negative.

Nordhaus has looked at the broad impacts of "resource scarcity" on world economic growth9.He estimates that rising energy prices probably constitute a "drag" of about 0.15% p.a. and thatgreenhouse warming will add a further 0.03% to this. Allowing for other environmental and mineralscarcity costs, the total drag is tentatively put at 0.31% p.a. But on Nordhaus's estimates, the costsof environmental policy are not a major part of this cost. If the Jorgensen-Wilcoxen estimate isapplied to the world as a whole (a clear exaggeration given the relative strictness of USenvironmental policy), the 0.31% estimate might rise to 0.45%, which begins to look significant.

OECD

The OECD has produced a six-country survey of environmental policy costs". Covering the

7 R.Hahn and J.Hird, "The Costs and benefits of Regulation: Review and Synthesis", YaleJournal of Regulation, Vol.8, No.1, Winter 1991.

8 See D.Jorgensen and P.Wilcoxen, "Environmental Regulation and US Economic Growth",RAND Joumal of Economics, Vol.21, No.2, Summer 1990.

9 W.Nordhaus, "Economic Growth: Limits and Perils", Paper presented to the InternationalCongress on Environment, Ethics. Economics and Institutions, Milan, March, 1991.

10 OECD, The Macroeconomic Impact of Environmental Expenditure, OECD, Paris, 1985.

53

USA, Norway, Netherlands, France, Finland and Austria, the study concluded that the effects ofpolicy on GDP were indeterminate, with long run rises of 1% over baseline in some cases, and fallsof 1% in others; that inflation might generally be worsened by as much as 0.3-0.5% p.a; thatemployment is stimulated by the growth of the pollution abatement sector and a slight depressingeffect on productivity; and that the beneficial effects of increased regulatory expenditures occur inthe short-term, with negative effects occurring in the longer-run. The OECD concludes that:

While these various results are of interest in their own right, the main conclusion whichemerges from them is that the macroeconomic effects of environmental policies is relativelysmall. Most of the figures reported - with the exception of some of the results for consumerprice inflation - are in the range of a few tenths of a percentage point per year. Furthermore,it is important to recall that these small effects were registered during a period (the 1970s)of peak pollution control activity. when efforts were directed not only at limiting on-goingpollution, but also at cleaning-up the ba^klog caused by neglect of the environment duringthe 1950s and 1960s.

3. Conclusions

The available economic studies do not bear out the worst fears about employment, price andincome effects of environmental policy. They tend to suggest that environmental policy can actuallyincrease employment and income, or at least make them no worse than would otherwise be the case.But there are several caveats.

First, the detailed study by Jorgensen and Wilcoxen for the IJSA suggests annual GDPsacrifices that would probably be regarded as politically significant. How far this result reflectsspecific US regulatory measures, and how far it reflects the sophistication of the niacroeconomicmodel, is impossible to say. Second, these sacrifices have to be compared to the benefits ofenvironmental policy, on which there is even less empirical evidence.

Third, the relevant studies are still limited in number. The fact that regulatory impacts takesome time to work through the economy also mean that the methodologies involved require longerrun economic growth models than those typically used for short-term forecasting. But we can be surethat the evidence available does nMt support the politically-received wisdom that more environmentalregulation will be harmful to economic growth - at least in Europe.

Finally, whatever the cost of environmental policy, there ought to be more emphasis on policyefficiency; and that tends to point towards a bolder policy of embracing green taxes andsupplementing traditional regulatory policies with tradeable permit systems.

54

Flbure Al Pubic and Private Expenditures on Pollution Control In OECD Countries 1972 - 1986

Public and Private Pollution Control Expenditure-1I8(t prices' and exchange rates

ltIIIan of US Biltlon of US SS14F - , Wo

S12UNlED STAJEL. -8T->-$40

S10 -eEGRMANY

- -~~---~~.~z- -- $31)

St _-__6~ FRIWCE ! $w' 20

S4 ~ ' -'

-..RETHERLMW-- $1

2 .-10FINLAND NORVAY

So(- . .I.* . I1$0

72 73 74 78 77 78 70 80 81 82 e 14 688 6e~br

U.S. fiyure refcr it, nght-ha.tndscale Private h usehtild expndilture excluded1. Deflated with GDP pnce index

Public and Private Pollution Contml ExpenditurePercentage of GDP

17 .

A JIT3STRTA 6

NETHERLANDS8

OA FRA!CNORY

72 73 74 76 76 77 76 70 60 81 82 68 64 6 8666wr

Private houusehold expemfiturc excludcxl

55

Fbaur. A2 Macroeconomic Impacts of Various Environmental Polces for the UK (Cambrdge Econometrics)

A G3REEN SCENARIO: THE MACROECONOMICEFF:EC'TS

(differences from base)1995 2000 2005 2010

ODP and Components of (DP(% difference from base)

Consumers'expenditure (.5 1.8 4.2 -1.7Fixed investment 1.4 2.1 2.7 -1.3Exports of goods mid services -0.1 --0.6 1.2 -1.7Imports of goods and services 0.4 0.5 1.1 -3.2ODP at factor cost 0.4 0.9 2.2 -0.9

Inflation(pp difference from base)

Consumers' prices 0.1 0.3 0.1 .0-2Average earnings 0.3 0.6 0.7 -.1.1

Employment(difference from basc iln '000)

Employment 191 427 682 0Unemployment --92 --207 --365 0

Source: Cambridge Econometrics

56

Flours A3 Macroeconomic Impact of A Strong EnvironmenteI PoUcy: the Notherands

Results in 20(10 (compared to 1985)Unchauigcd Policy Sustainable Growth

without' with'Volume GNP (%) +98 +95 +100Real wages (%) +61 +59 +61Consumption(%) -119 +118 +121Employment (%) +26 +26 +28Budget dcefcit(% point of NI) -2.4 + 1.0 -3.4

Interest rate (%point) -1.1 +0.2 -0.7

C02 emissions (%) +35 -20 to -30CFCs emissions (1) -I(H) 1 00S02 emissions (MY) -50 8() to -9(NOx emissions (%M --10 -70 to -80Hydrocarbons (%) - -20 -70 to -80Discharges intoRhine & North Sea (%) *-50 .. 75Waste dumping (%) 0 -. 70 to -80

Environmentalexpenditure (%NI) 2% 4%Total investmnent inpetiod 1985-2010 1(0 350

I)without equivalent policy ip foieig,n countries2)with equivalent policy inf tdrcigii coutitries3)compared to 1980

Source: NEPI'. 1989.

57

Flour. A4 Inoome and Employment Impacts of Acid Rain Pollution Control In the Europen Communkty

Annual X deviation from baseline LeveLs

SO, Control NOx Control

1988-93 1994-97 1988-93 1994-97Country:

France Y 0.04 0.03 n.e. n.a.N 0.02 0.02 n.s. n.a.

Germany Y 0.13 -0.06 0.06 0.00N 0.13 -0.11 0.05 -0.03

Italy Y 0.08 0.03 0.03 0.02N 0.02 0.01 0.01 0.01

UK Y 0.06 -0.05 0.01 -0.01N 0.03 -0.05 0.01 -0.01

Y = GDP, N = employment.

58

BIBLIOGRAPHY

General References

A.Kneese Measuring the Benefits of Clean Air and Water, Resources for theFuture, Washington DC, 1984

J-Ph.Barde and Valuing the Environment, Earthscan, London,D.W.Pearce 1991

J.Bentkover (ed) Benefits Assessment: the State of the Art, Reidel, Dordrecht, 1986

D.W.Pearce Environmental Policy Benefits: Monetaryand A.Markandya Valuation, OECD, Paris, 1989

A.M.Freeman The Benefits of Environmental improvement, Johns HopkinsUniversity Press, Baltimore, 1979.

R.Cummings et al. Valuing Environmental Goods: an Assessment of the ContingentValuation Method, Rowman and Allenheld, Totowa, NJ, 1986

R.Carson and IUsing Surveys to Value Public Goods. theR.Mitchell Contingent Valuation Method, Resources forthe Future, Washington

DC, 1989.

R.K.Turner aid A Critical Review of Monetarv AssessmentI Bateman Methods and Techniques, Environmental Appraisal Group, University

of East Anglia, 1990.

P-O Johansson The Economic Theory and Measurement of Environmental Benefits,Cambridge University Press, Cambridge, 1987

J.Braden and Measuring the Demand for EnvironmentalC.Kolstad Ouality, North Holland - Elsevier, Amsterdam, 1991.

For valuation in developing countries see:

D.W.Pearce and Environment and Economic Development:J.Warford Managing Natural Resources in the Developing World, forthcoming

1992.

J.Bojo, K-G Maler, Environment and Development: an EconomicL.Unemo Approach, Kluwer, Dordrecht, 1990.

59

J.Winpenny Values for the Environment, HMSO, London, 1991

References in the Text

N.Adger and M.Whitby [1991], "National Accounting for the Externalities of Agriculture andForestry", Countryside Change Unit, University of Newcastle-upon-Tyne, Working Paper 16, AprilD.Anderson [1987], The Economics of Afforestation: a Case Study in Africa, Johns HopkinsUniversity Press, Baltimore

D.Anderson [19891, "Economic Aspects of Afforestation and Soil Conservation Projects", inG.Schramm and J.Warford, Environmental Management and Economic Development, Johns HopkinsUniversity Press, Baltimore

E.Barbier et al. [1991], Economic valuation of Wetland Benefits: the Hadeia-Jama'are Floodplain.Nigeria, London Environmental Economics Centre, Paper 91-02, London.

J.Bennett [1982], "UsingDirect Questioning to Value Existence Benefits of Preserved Natural Areas",School of Business Studies, Darling Downs Institute of Education, Toowoomba

K.Boulding [1966], "The Economics of the Coming Spaceship Earth", in H.Jarrett (ed), EnvironmentalOuality in a Growing Economy, Johns Hopkins University Press, Baltimore.

J.Bergstrom et al [1990], "Economic Value of Wetlands-Based Recreation", Ecological Economics,Vol.2, No.2, June .

J.Bishop and J.Allen [19891, The On-Site Costs of Soil Erosion in Mali, Environment DepartmentWorking Paper No.2 1, November 1989, World Bank, Washington DC.

D.Brookshire, W.Schulze and M.Thayer 119851, "Some Unusual Aspects of Valuing a Unique NaturalResource", Department of Economics, University of Wyoming, February, mimeo.

G.Brown and W.Hall, The Viewing Value of Elephants, London Environmental Economics Centre,London, 1989; and E.Barbier, J.Burgess, T.Swanson and D.W.Pearce, Elephants. Economics andIvory, Earthscan, London, 1990.

K.Boyle and R.Bishop [1985], "The Total value of Wildlife Resources: Conceptual and EmpiricalIssues", Paper presented to Association of Enviromental and Resource Economists, Boulder, MayJ.Briscoe, P de Castro, C.Griffin, J.North and O.Olsen, "Toward Equitable and Sustainable RuralWater Supplies: a Contingent Valuation Study in Brazil",The World Bank Economic Review, Vol.4,No.2, 1990.

D.Brookshire et al.[1983], "Estimating Option Prices and Existence Values for Wildlife Resources",Land Economics, 59

J.Broome [1991], The Intergenerational Aspects of Climate Change, Report to the UK Economicand Social Research Council, Swindon.

60

W.Clark, "Economics and Marketing of Canada's Capistrano", in A.Diamond and F.Filion (eds), TheValue of Birds, International Council for Bird Preservation, Cambridge, 1987.

W.Cline [19911, Estimating the Benefits of Greenhouse Warming Abatement, OECD, Paris.

R.Costanza et al. [1989], S.Farber and J.Maxwell, "Valuation and Management of WetlandEcosystems", Ecological Economics, Vol.1, No.4, December

L.Dahle et al.[1987], "Attitudes Towards and Willingness to pay For Brown Bear, Wolverine and Wolfin Norway",Department of Forest Economics, Agricultural university of Norway, Report 5/1987, (inNorwegian)

P.Dasgupta, A.Sen, S.Marglin [19721, Guidelines for Project Evaluation, UNIDO, Vienna.

H.Gittinger [1982], Economic Analysis of Agricultural Proiects, Johns Hopkins University Press,Baltimore.

R.Hageman 119851, "Valuing Marine Mammal Populations: Benefit Valuations in a Multi-SpeciesEcosystem", National Marine Fisheries Service, Southwest Fisheries Center, Report LJ-85-22, LaJolla, California

A.Hervik et al.[1986], "Implicit Costs and Willingness to pay for Development of Water Resources",in A.Carlsen (ed), Proceedings of UNESCO Symposium on Decision_Making in Water ResourcesPlanning, May, Oslo

D.Imber et al.[1991), A Contingent Valuation Survey of the Kakadu Conservation Zone, ResourceAssessment Commission, Research Paper No.3, Canberra, February

A.Jacquemot and F.Filion, "The Economic Significance of Birds in Canada", in A.Diamond andF.Filion (eds), The Value of Birds, International Council for Bird Preservation, Cambridge, 1987.

I.M.D. Little and J.Mirrlees [1974], Project Appraisal and Planning for Developing Countries,Heinemann, London

W.McGrath and P.Arens [1989], The Costs of Soil Erosion on Java: a Natural Resource AccountingApproach, World Bank, Environment Department Working Paper 15, Washington, August.

X.Mu, D.Whittington, J.Briscoe, "Modeling Village Water Demand Behavior: a Discrete ChoiceApproach", Water Resources Research, Vol.26, No.4, 1989.

S.Navrud [1991], "Norway", Ch.5 of J.Ph Barde and D.W.Pearce, Valuing the Environment,Earthscan, 1991.

J.Nelson 11980], "Airports and Property Values: a Survey of Recent Evidence", Journal of TransportEconomics and Policy, XIV

J.Nelson [1982], "Highway Noise and Property Values: a Survey of Recent Evidence", Journal ofTransport Economics and Policy, XVI

61

K.Newcombe, "An Economic justification for Rural Afforestation: the Case of Ethiopia", inG.Schramm and J.Warford (eds), Environmental Management and Economic Development, JohnsHopkins University Press, Baltimore, 1989.

W.Nordhaus [19911, "A Sketch of the Economics of the Greenhouse Effect", American EconomicReview, Vol.81, No.2.

OECD [19891, Environmental Policy Benefits: Monetary Valuation, OECD, Paris

J.Opschoor [19861, "AReview of Monetary Estimates of Benefits of Environmental Improvement inthe Netherlands", OECD, Paris, October

R.Ottinger et al [19901, Environmental Costs of Electricity, PACE University Center forEnvironmental Legal Studies, White Plains, New York, September.

D.W.Pearce and J.Warford [19921, Environment and Economic Development: Managing NaturalResources in the Developing World, forthcoming

P.Portney (ed)[1990], Public Policies for Environmental Protection, Resources for the Future,Washington DC.

P.Principe [19891, "The Economic Significance of Plants and their Constituents as Drugs", inH.Wagner, H.Hikino and N.Farnsworth, Economic and Medicinal Plant Research, Vol.3, London:Academic Press, 1989,pp.1-17.

R.Repetto [19891, Wasting Assets - Natural Resources in the National Income Accounts, WorldResources Institute, Washington

J.Ruitenbeek 1990al, "The Rainforest Supply Price: a Step Towards Estimating a Cost Curve forRainforest Conservation", Suntory-Toyota International Centre for Economics and RelatedDisciplines, Paper 29, London School of Economics, September.

J.Ruitenbeek 1990b], Evaluating Economic Policies for Promoting Rainforest Conservation inDeveloping Countries, Ph.D thesis, London School of Economics

J.Ruitenbeek [1990c], Economic Analysis of Tropical Forest Conservation Initiatives: Examples fromWest Africa, World Wide Fund for Nature, Godalming

K.Samples et al.[19861,"Information Disclosure and Endangered Species Valuation", Land Economics,62, No.3

W.Schulz [1986], "A Survey of the Status of Research Concerning the Benefits of EnvironmentalPolicy in the Federal Republic of Germany", OECD, Paris

W.Schulze et al.[19831, "The Economic Benefits of Preserving Visibility in the National Parklands",Natural Resources Journal, Vol.23, January.

W.Schulze et al.[19831, "Economic benefits of Preserving Visibility in the National Parklands of the

62

Southwest", Natural Resources Journal, 23

L. Squire and H.van der Tak [1975], Economic Analysis of Projects, Johns Hopkins University Press,Baltimore.

R.Stoll and L.Johnson [19841, "Concepts of Value, Non-market Valuation, and the Case of theWhooping Crane", Department of Agricultural Economics, Texas A&M University

F.rhibodeau and B.Ostro [17R], "An Economic Analysis of Wetland Protection", Journal ofEnvironmental Management, 12 (1), January

D.Thomas et al. [1990], "Use Values and Non-Use Values in the Cponservation of Ichkeul NationalParl, Tunisia", Department of Geography, University College London, L.ndon, mimeo.

D.Tobias and R.Mendelsohn [19911, "Valuing Ecotourism in a Tropical Rain-Forest Reserve", Ambio,Vol.20, No.2, April, 91-93

R.K.Turner [19901, Market and Intervention Failures in the Management of Wetlands, School ofEnvironmental sciences, University of East Anglia, Norwich, UK, mimeo.

US Department of Agriculture [19901, Resources and Technology Division, Economic PesearchService, Climate Change: Economic Implications for World Agriculture, Washington DC, November1990 (Draft).

US Environmental Protection Agency [1987], EPA's Use of Benefit-Cost Analysis 1981-1986, EPA-230-05-87-028, Washington DC, August

US Environmental Protection Agency [19851, Costs and benefits of Reducing Lead in Gasoline: FinalRegulatory Impact Analysis, EPA-230-05-85-006, Washington DC, February.

R.Walsh et al.[1984], "Valuing Option, Existence and Bequest Demands for Wilderness", LandEconomics, Vol.60, No.1

K.Willis and J.Benson [1988], "Valuation of Wildlife: A Case Study on the upper Teeside Site ofSpecial Scientific Interest and Comparison of Methods in Environmental Economics", in R.K.Turner(ed), Sustainable Environmental Management, Belhaven Press, London.

World Bank [1988], Madagascar - Environmental Action Plan, July

World Bank [1990a], Towards the Development of an Environmental Action Plan for Nigeria, WorldBank, Dec.18.

World Bank [1990cJ, Towards the Development of an Environmental Action Plan for Nigeria,Western Africa Department, December

63

Policy Research Working Paper Series

ContactTitle Author Date for paper

WPS963 Growth and Welfare Losses from Charles R. Blitzer August 1992 WDR Gff iceCarbon Emissions Restrictions: R. S. Eckaus 31393A General Equilibrium Analysis Supriya Lahirifor Egypt Alexander Meeraus

WPS964 Toxic Releases by Manufacturing: Robert E. B. Lucas August 1992 WDR OfficeWorld Patterns and Trade Policies 31393

WPS965 Coping with the Disappointing Rates William Ascher August 1992 WDR Officeof Return on Development Projects 31393That Affect the Environment

WPS966 Trade and the Environment: A Survey Judith M. Dean August 1992 WDR Officeof the Literature 31393

WPS967 Transition Problems in Economic Santiago Levy August 1992 M. StroudeReform: Agriculture in the Mexico- Sweder van Wijnbsrgen 38831U.S. Free Trade Agreement

WPS968 Biomass David 0. Hall August 1992 WDR Office31393

WPS969 Imports, Exports, and Industrial M. Ataman Aksoy August 1992 M. ShorePerformance in India, 1970-88 Helena Tang 31129

WPS970 Political Models of Macroeconomic Alberto Alesina August 1992 E. KhinePolicy and Fiscal Reform 39361

WPS971 The Effects of Democratic Milan Vodopivec August 1992 CECTMDetermination of Wages: Theory and 37188Evidence from Self-Managed Firms

WPS972 Commercial Energy Efficiency and Robin W. Bates September 1992 WDR Officethe Environment Edwin A. Moore 31393

WPS973 How Changes in the Former CMEA Refik Erzan September 1992 G. IlogonArea May Affect International Christopher Holmes 33732Trade in Manufactures

WPS974 How Minilateral Trading Arrangements C. A. Primo Braga September 1992 J. JacobsonMay Affect the Post-Uruguay Round Alexander Yeats 33710World

WPS975 Techniques for Improving Client George B. Simmons September 1992 0. NadoraRelations in Family Planning Sara Koerber 31091Programs Ruth Simmons

WPS976 Strategic Management of Family Cynthia P. Green September 1992 0. NadoraPlanning Programs 31091

Policy Research Working Paper Series

ContactTitle Author Date for paper

WPS977 Income Security for Old Age: Estelle James September 1992 D. EvansConceptual Background and Major 37489

WPS978 How Restricting Carbon Dioxide Charles R. Blitzer September 1992 WDR Officeand Methane Emissions Would Affect R. S. Eckaus 31393the Indian Economy Supriya Lahiri

Alexander Meeraus

WPS979 Economic Growth and the Dennis Anderson September 1992 WDR OfficeEnvironment 31393

WPS980 The Environment: A New Piritta Sorsa September 1992 WDR OfficeChallenge for GATT 31393

WPS981 After Socialism and Dirigisme: Andres Solimano September 1992 S. FalonWhich Way? 37947

WPS982 Microeconomics of Transformation Brian Pinto September 1992 S. Husainin Poland: A Survey of State Marek Belka 37139Enterprise Responses Stefan Krajewski

WPS983 Legal Reform for Hungary's Private Cheryl W. Gray October 1992 R. MarticSector Rebecca J. Hanson 39065

Michael Heller

WPS984 Barriers to Portfolio Investments Asli DemirgUq-Kunt October 1992 W. Patrawimolponin Emerging Stock Markets Harry Huizinga 37664

WPS985 Regional Integration, Old and Jaime de Melo October 1992 D. BallantineNew Claudio Montenegro 37947

Arvind Panagariya

WPS986 The Administration of Road User Roy Bahl October 1992 J. Francis-O'ConnorTaxes in Developing Countries 35205

WPS987 How the Epidemiological Transition Jose Luis Bobadilla October 1992 0. NadoraAffects Heafth Policy Issues in Cristina de A. Possas 31091Three Latin American Countries

WPS988 Economic Valuation and the David Pearce October 1992 WDR OfficeNatural World 31393