This article was downloaded by: [Seton Hall University]On: 25 September 2014, At: 15:23Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

Project AppraisalPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tiap18

Economic valuation of environmental impactsEdward B Barbier aa International Institute for Environment and Development , 3 Endsleigh Street,London , WC1H 0DD , EnglandPublished online: 17 Feb 2012.

To cite this article: Edward B Barbier (1988) Economic valuation of environmental impacts, Project Appraisal, 3:3,143-150, DOI: 10.1080/02688867.1988.9726674

To link to this article: http://dx.doi.org/10.1080/02688867.1988.9726674

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”)contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensorsmake no representations or warranties whatsoever as to the accuracy, completeness, or suitabilityfor any purpose of the Content. Any opinions and views expressed in this publication are the opinionsand views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy ofthe Content should not be relied upon and should be independently verified with primary sources ofinformation. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands,costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly orindirectly in connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial orsystematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution inany form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Project Appruisul, volume 3, numbcr 3. Scpicmbcr 1988, pagcs 143 - 150, Uccch Tree Publishing, 10 Walford Closc, Guildford. Surrcy. GUI 2EP, England.

Resource management

Economic valuation of environmental impacts

Edward B Barbier discusses data and methodology requirements with an example from Indonesia

Using a study conducted for UNDP (United Nations Development Program) in Indonesia as a base, the data and methodology requirements for economic valuation of environmental impacts are investigated. Such valuation procedures are instrumental not only to project appraisal but also to other areas of policy analysis vital to designing environmentally sustain- able development approaches in developing countries.

In addition to adequate biophysical and resource stock assessment, the data and methodology require- ments include proper accounting and valuing of envi- ronmental functions, sirfficient flexibility of data to reflect changes in key resource systems and zones, arid high quality standards. Although specifVing the exact requirements of this data base is difficult unless choices are made concerning its ultimate purpose, the approach adopted serves as a useful guideline to the foundation of any such data base.

Kcywords: environmental impact; cconomctrics; policy; Indoncsia

Edward B Barbier is at thc International Institutc Tor Environnicnt and Dcvelopment, 3 Endsleigh Street. London WClH ODD, Eng- land.

This papcr is based on a rcport by the aulhor (Rarbicr 1988) for the study Environmental Sector Review of Indonesia, Phase II [or UNDP and the Ministry of State for Environment and Population, Govcrnment of Indoncsia. Thc vicws exprcssed in this papcr arc those of the aulhor and should not in any way be attributed to UNDP or thc Ministry of State for Environmcnt and Population.

HE GROWING RECOGNITION that environmental considcrations must be incorporated into develop T ment strategies is starting to have some influence on

policy-making and planning in dcvcloping countries. This has lcd to the reccnt interestin the economic valuation of environ- mental impacts, which has already started to influence projcct appraisal tcchniqucs and other areas of policy analysis.

For example, as pointed out by the authors of the classic UNIDO (United Nations Industrial Development Organisa- tion) Guidclincs, the main rationale for conducting social cost-bcncfit analysis is “to subject project choice to a consis- tent set of gcncral objectives of national policy” (UNIDO 1972). As perceptions of national policy objectives in devel- oping countries have changed, for instance emphasizing the need for scarce foreign exchange and cquitable income distri- bution, project appraisal and planning have been expandcd to rcflect the new objectives (Little and Mirrlees 1974; Squire and van dcr Tak 1975).

Consequcntly, thc rcccnt emphasis on the role ofenviron- mcntal quality and the long-run productivity of natural re- source systcms in sustaining economic development has Icd to furthcr extcnsions of social cost-benefit analysis to include environmental impacts (Dixon et al, 1986; Dixon and Hufschmidt 1986; Hufschmidteiall983). Thatis, incontrast to traditional project evaluation which considcrs only the dircct projcct bcncfits and cosLs, “the cxpandcd approach includcs the extcrnal and environmental improvement bcne- fits from cnvironmcntal protcction, as well as the costs of cxtcrnal and/or cnvironmcntal damages and of environmental control mcasurcs” (Dixon and Hufschmidt 1986, page 7).

Thc basic methodology is first to idcntify and measure h e cnvironnicnul cfl‘ccts, and thcn sccondly, to translate thcm into rnonctary tcrrns for inclusion in the formal project analy- sis.

Howcvcr, cxtcnding cost-bcncfit analysis to incorporate thc cnvironmcntal impacts ofprojccts encounters anumbcrof problcms: 0 physical cstimation of environmcntal cffccts is often diffi-

cult.

Project Appraisal September 1988 0268-8867/88/030143-8 US $03.00 @ Beech Tree Publishing 1988 143

Dow

nloa

ded

by [

Seto

n H

all U

nive

rsity

] at

15:

23 2

5 Se

ptem

ber

2014

Economic valuation of environmental impacts

0 as most environmcntal resources are non-markcted com- mon-property ‘goods’. economic valuation of their scrv- ices is not straightforward,

0 little consensus exists regarding mcthods for monetary valuation of ‘intangible’ environmental goods, such as the need to preserve unknown species and ‘natural bcauty’,

0 since this expanded approach inevitably raises issucs of inter-temporal choice, the interest rate chosen to discount the future may determine whcthcr environmental degrada- tion is ‘optimal’.

It is often stressed that the appropriate discount rate should emerge from the project appraisal process (UNIDO 1972). In practice, imperfect capital markets, inconsistent data on thc productivity of capital and large variances in domestic bor- rowing for investment make it difficult to cstablish an eco- nomic accounting rate of interest for developing countries (Phillips 1986).

Introducing environmental considerations furthcr compli- cates the picture. As Markandya and Pcarcc (1987) observe, natural resources are more likely to be over-exploited at high discount rates than at low ones, whcreas low discount rates discriminate against projects with an environmental dimcn- sion that have a long gestation period. Givcn the additional problems posed by environmental risk and inter-tcmporal impacts, these authors conclude that it is generally prcfcrable to adjust the project costs and bcnefit values and adopt additional sustainability criteria as constraints on these values than to adjust the discount rate.

All these considerations point to the need for propcr economic valuation of the environmental impacts of devclop- mcnt, which in turn requires developing an effectivc environ- mental and resource data base for economic indicators of sustainability. This papcr therefore focuses on the establish- ment of a feasible set of environmental and natural rcsource data consistent with this objective.

Some of the relevant environmental and resource data may already exist in developing countries, but not collcctcd or aggregated in a form proper for economic valuation of en- vironmental impacts. Moredata collection may be requircd to assist policy and planning choices concerning sustainable resource use; however, thcrc are limits as to the desirability of increased collection, monitoring and evaluation of environ- mental impacts if the costs of the extra information receivcd exceed the benefits to policy-makers in terms of a more informed choice.

Although such a data base is expcctcd to be particularly useful in project appraisal, this papcr assumes that propcr economic valuation of environmental impacts is also impor- tant for a variety of policy approaches that range from ‘statc of the environment’ indicators to propcr resourcc accounting methods to natural resource management policy analysis. Obviously, the data and mcthodological requircmcnts oT es-

There is a need for proper economic valuation of the environmental impacts of development, which requires developing an effective environmental and resource database for economic indicators of sustainability

tablishing a sct of cnvironmenbl indicators would be signifi- cantly difkrent from that for a system of rcsource accounts, for projcct appraisals, for simulation modcls or for policy analysis.

Nevcrlhcless, as the least common denominator for all thcsc approachcs is some form of economic valuation of the environmcnt, they share certain fundamcntal data needs. Thcsc include adcquate biophysical and resource stock as- sessmcnt, propcr accounting and valuation of environmcntal functions, sufficient data flcxibility to reflect changes in key resource systems and zones (such as watersheds and coastal areas) and high quality standards. These basic data and mcthodological rcquirements are examined in this paper, using examples from Indonesia where appropriate.

Environmental degradation

The evaluation of environmental impacts must take into account the multifunctionality of rcsources, the irreversibility of environmcntal dcgradation and the nccd for data on these impacts to fulfill many purposes.

Multifunctionality of resources

A kcy feature of any natural resource is its multifunctionality; that is each resource serves morc than one function - water resources can be uscd for municipal and industrial supplies, irrigation, waste disposal, and so on. Proper economic valu- ation of eitvironmcntal impacts on a whole range of rcsource functions is thcrefore crucial to understanding and analyzing the trade-offs among the various functions of the cnviron- ment, trade-offs bctwecn development and one or more of the environmental functions, and any spatial and inter-temporal dimcnsions of these possible trade-offs.

It is also important for emphasizing the value of the environmcnt as an ‘intcrmediatc good’ for instancc, as the rcsource base for production) as opposed to its value solcly as a ‘final good’ (for instance, as a source of amcnity value to be ‘prcscrvcd’). Such a valuation procedure therefore implies: 0 Understanding the functions of the natural resource base/

stock and basic ecological functions in a given economic- environmental systcm, and how thcsc functions interact with each othcr.

0 Propcr valuation of each of thcse functions, as far as eco- nomic and ccological data allow, at least to derive some broad ordcrs of magnitudes of value for these functions.

0 Using thcsc valuations to indicateanddetcrmine the trade- offs that may cmcrgc from natural rcsourcc degradation ovcr timc (Pcarcc et a1 1987).

Propcr cvaluation of environmental impacts in this manner is csscntial for a widc rangc of issues ccntral to the analysis of natural rcsource rnanagemcnt. Thcse include: 0 Undcrstanding the implications ofenvironmental dcgrada-

tion for crucial resourcc systcms and ecosystem functions (such as watcrshcds, drylands, maintenance of soil fcrtility and cohesion, coaslal zones, wastc assimilation).

0 Undcrslanding the implications of irreversible choices with rcgard to thc potcntial loss of environmental func- tions.

0 Incorporating analysis orcnvironmcntal impacts in projcct appraisals, as well as in sectoral programs and activities.

0 Dcvcloping a national systcm of resourcc/environmcntal accounts.

144 Project Appraisal September 1988

Dow

nloa

ded

by [

Seto

n H

all U

nive

rsity

] at

15:

23 2

5 Se

ptem

ber

2014

Economic valuation of environmental impacts

Table 1. Biophysical and resource stock assessment

Resource base

Non-renewables (eg. metals, minerals, land, fossils fuels) Renewables (eg, aquatic and terrestrial biomass, water)

Semi-renewables (eg, soils, genetic diversity)

Assimilative Capacities (eg, water, air and land quality) Resource systems (eg, coastal zones, drylands,waters heds)

Stock data

Total reserves Classes

Total reserves Classes

Total reserves Classes

Pollutant stocks Classes/ toxicity Total area

Physical assessment __ Depletions Additions

Extraction rates Losses

Extraction Degradation Damage

Extraction/losses Degradation Damage

Pollution/discharge PoinVnon-point sources

Extraction/losses Degradation Damage

New discoveries Extensions Revisions Growth Recharge New discoveries Extensions Revisions Growthhecharge New discoveries Extensions Revisions Recharge Dispersion

Growth/extensions

0 Incorporating analysis of environmcntal impacts into thc economic planing process through: - convincing policy-makers of the need to integrate sus-

tainablc natural resource managerncnt into this process; - facilitating the design of appropriate policies, inccntive

structures and investment stratcgies for environmen- tally sustainable dcvelopmcnt.

Thus the data requirements for the analysis of natural resource impacts must initially focus on the appropriate data set for proper economic valuation of environmental asscts, their multi-functional services, and the impacts of environmental degradation and subsequent loss of thcse services over time.

Multifunctionality of data

However, in order to determine the appropriate environmental and resource data requircments, the spccific functions of the data set need to be clarified. Economic valuation of environ- mental impacts in developing counlrics could perform many functions, such as early warning systcms of environmental degradation and impending natural (or man-made) disasters; ‘state of the environment’ indicators; policy analysis and design; long-term forecasting; model-building; and rcsource accounting.

As each function has its own data demands, thc priority in determining the appropriate environmcntal and rcsourcc data base to be compilcd must be to dccide what functions thc data are to scrve. For example, the dcgree and coverage of data

The priority in determining the appropriate environmental and resource data base to be compiled must be to decide what functions the data are to serve

rcquired for dcvcloping ‘state of the environment’ indicators is much less than that for policy analysis, forecasting,resource accounting and modcl building. Without clarifying these functions of the data base, it is difficult to indicate precisely its requirements; nevcrthcless, the following indicates the gen- eral guidclincs that any such data base needs to fulfill.

Data base requirements

An appropriate data basc for analyzing environmental im- pacts must satisfy a number of gencral rcquirements. The first is an accurate estimation of physical resource stocks and assimilative capacities, including rates of depletion and addi- tions, and the composition of important resource systems (such as watersheds and coastal systems).

Appraising the functions, or scrvices, of these natural rcsourccs and resourcc systcms is also important, and would include assessmen& of waste assimilative capacity, pollution stocks and dischargc ratcs; productive Cunctions; protective functions; and socio-cultural rclcvance. Mcthods of valuing thcse various functions (taking into account domcstic and evcn international pricc trcnds where appropriate, existcnce of rcnt, social versus private valucs and valuation procedures for non-marketcd scrvices) nccd to be dcveloped.

Finally, thcdatabasenecds to beflcxible; dataareroutinely collcctcd through administntivc/political units (for instance provinccs, districts, sub-districts, villagcs), yct the nature of cnvironmcnlal dcgradauon problems often requires data bascd on geographical/biophysical boundaries (such as soil crosion of uppcr watcrshcds, mangrove destruction in coastal zoncs).

Resource stock assessment

The starting point for compiling an cnvironmcntal and re- source data base is of coursc rcgularly assembling accurate in- formation on the biophysical characteristics and changes in thc natural rcsourcc basc. Onc possible method of classifica- tion is to distinguish among non-rcncwablc rcsources, rencw-

Project Appraisal September 1988 145

Dow

nloa

ded

by [

Seto

n H

all U

nive

rsity

] at

15:

23 2

5 Se

ptem

ber

2014

Economic valuation of environmental impacts

able resources, semi-renewable resources, waste assimilative capacities and important resource systems (see Table 1). In general, what is required is accurate information on the size, depletions and additions to this resource base.

For all resources, the challenge is to determine the total reserves or area initially and then to account accurately for changes in these stocks over time. Determining the nct depletion of most exhaustible resources is fairly straightfor- ward and is routinely performed as part of commercial operations; what is required is information on extraction rates, new discoveries and extensions of existing rcscrves, as well as estimates of losses from inefficient operations and wastage.

Land area is a special category of non-renewable resource. Unless new frontiers or annexations are possible, land area is essentially a finite, non-reproducible asset. Although land is not exactly ‘extracted’, land capability and its utilization does change over time - sometimes with significant scarcity impacts. Thus it is important to classify, and kcep track of, changes in land utilization and types of land, for example, the conversion of forest land to agriculture, urban encroachment on arablc land and increases in irrigated land through public works schemes.

The extra dimension that needs accounting for with rcnew- able resources and ccrtain ‘mixed’ renewablehon-renewable resources is biological growth rates and natural recharge (recovery) rates. In addition, improper exploitation or man- agement of these resources may severely impair biological productivity or natural recharge/recovery rates: this is often termed resource ‘degradation’.

Physical damage from fire, man-made disasters and en- croachments and even deliberate destruction should also be noted. In the case of soils, natural restoration of soil fertility occurs over a long period of time; consequently, measures of soil losses and sedimentation from wind and water erosion, landslides and other geologically ‘natural’ sources are esscn- tial.

Whereas individual specics of flora and fauna may be renewable, genetically diverse ecosystems and habitats may not be. Thus assessment and classification of biologically diverse regions and resource systems, as well as individual species known and thought to exist, arc vital tasks.

It may be worthwhile to collect biophysical data specific to certain resource systems or ‘eco zones’, such as coastal systems, watershed/river basins and dryland areas. If so, collecting information on the resources particular to thcsc zones and assessing it as an integrated whole would be appropriate.

Accounting for multiple functions

As noted above, economic evaluation of environmental im- pacls requires taking into account that resources often have multiple, competing functions. Assessing these various func- tions is thereforeat lcast as important as analyzing the physical changes in the resource base if not more so. Ideally, sufli- cient information should be gathered on these multiple func- tions in order to assess the impacts of development on: 0 trade-offs among the various functions of ihe environment; 0 trade-offs bctween dcvelopment and one or more of the

cnvironmental functions; 0 development in one geographical area that affects environ-

mental functions in anoihcr area; and 0 development in one period of time that affects environ-

mental functions in another period of time (Pcarce et a1 1987).

A useful way of classifying these functions may be to distin- guish between major production, minor production, environ- mental protection and system maintenance and socio-cultural support functions. Table 2 indicates some of these functions for certain major renewable resources in Indonesia. Note that the relationship between these functions is not necessarily

Table 2. Multiple functions of Indonesia’s renewable resource system

Functions

Resource

Soil

Freshwater

Forestsltrees

Coastal zones

Bio-diversity

Major production

Staple crops Export crops Livestock Irrigation Power Domestic Industrial Transport Timber Fuelwood Ecosystem values

Fishlshrimps Rice Ports/industry Drugs/

Tourism Genetic

pharmaceuticals

information

Minor production

Minor crops Bricksltiles Minerals Fish ponds

Minor forest products

Mangroves

New crops

Environmental protection

Landslip prevention Flood prevention Water holding Saline intrusion

prevention Waste assimilation

Soilhnrater

Windbreaks Microclimate Waste assimilation

conservation

Pesffdisease control

socio- cultural support

Minority forest communities

Fishing communities

Hunter/

Education Science

gatherers

146

Source: Adapted from Ackermann eta/ (1987)

Project Appraisal September 1988

Dow

nloa

ded

by [

Seto

n H

all U

nive

rsity

] at

15:

23 2

5 Se

ptem

ber

2014

Economic valudion of environmental impacts

compctitivc. For example,clcaringorevcn sclcctivcly cutting a forest for timbcr production may affcct its provision of minor forcst products (such as rcsins, honey and rattan) and some of its cnvironmental protection functions, but exploitation of minor forest products may not necessarily reduce the cnviron- mcntal protection affordcd by forests and may actually bcncfit indigenous forest dwellers.

The importance of collecting sufficicnt data to analyze thcse trade-offs is illustrated by the casc of coastal zonc cxploitation in Indonesia. The current policy stratcgy is to expand fish and shrimp production through increasing coastal and tambak (shrimp ponds) fishing in thesc zoncs. Howevcr, as noted in Ackermann et a1 (1987), policics for sustainable managemcnt of tambak, off-shore and cstuarine fishcrics facc crucial tradc-offs in coastal land use:

Mangrove vs off-shore fisheries - Indonesia's mangrove forcst systcms scrve as a habitat for brceding spccics impor- tant for off-shore fishcries. At the samc time, mangrove forests are cxploitcd in 12-20 ycar harvcst cyclcs for wood chip production, which has significant cxportpotential. Thcse compcting uses occur mainly along thc south coast ofJava and in thc Outcr Islands.

Mangrove vs tambak - In coastal areas off Java, namely in South Sulawesi, Sumatra and South Kalimantan, mangrove forcst systcms arc also bcing threatcncd by tambak cxpansion. This in turn is rcducing the natural shrimp brecding grounds in mangrove swamps and coastal areas. In addition, thc dcstruction of mangrove forests for tambak incrcascs thc risk of shore erosion, which in turn thrcatcns thc tambaks thcm- sclves.

Tambak vs rice cultivation - On Java, rice cultivation and tambak arc competing for scarcc coashl lands. In addition, water reaching tambak from rice systems up thc rivcr basin needs to be rcgular in flow Lo control salinity and to bc rclativcly frcc from agrochcmicals and fcrtilizcrs.

Valuation of multiple functions

In determining the mdc-olfs among thc niultiplc functions of thc cnvironnicnt it is cssential to have propcr valuation of thesc functions, insofar as thc economic and biophysical data allow, at lcast to dcrive somc broad ordcrs of magnitudcs of value. Idcally, the data should bc suflicicnt to givc an indica- tion of the costs bornc by socicty of losscs in cnvironmcntal functions. Thcse costs would include (SCC Pcarcc 1986; Dewccs 1987): a User costs - the direct costs Lo the uscr of a resourcc for

a particular function (the privatc cost).

In deterinining the trade-offs among the multiple functions of the environment it is essential to have proper valuation of' these functions as far as economic and biophysical data allow

Intcr-tcmporal user costs - the benefits foregone by those who might usc thc resource in the future for the same function. Unlcss a user owns all future rights to the resource, thcse costs will be bomc by othcrs, including possibly future gencrations. Social costs -the incfficicncy and cxtcmal costs imposed on non-uscrs, both now and in hc future, from any loss of other functions due to exploitation of the resource.

Mcthods for valuing cnvironmcnlal functions in dcveloping countrics have bccn considcrably improvcd in recent years (scc, for cxample, Dixon et a1 1986, Dixon and Hufschmidt 1986 and Hufschmidt et a1 1983). Although it may not be feasible to implcment all of thcsc tcchniqucs. any valuation approach will requirc basic data on the values to be assigned to various cnvironmcntal functions. To determinc thc data nccdcd, it may be uscful to distinguish among tradeable, marketed and non-marketcd functions of the cnvironment.

Marketed funelions

In gcncral, whcre cnvironmcntal functions are marketed (for cxamplc thc majority of major and minor production func- tions in Table 2), thcir priccs should scrvc as the basis for cstimating thcir valuc -assuming that market prices closely reflcct social valucs. Thus amassing price and cost series data for thcsc functions would be a first priority.

This should includc estimatcs on exploration and extraction costs (for exhaustiblc rcsources); harvesting costs (for rcncwablcs); producer margins (thc proportion of markct pricc actually rcccivcd by rcsourcc uscrs); tax/subsidy mar- gins (the proportion of'markct pricc appropriatcd/subsidized by govcrnmcnt; and economic rent (uscrs' rcvenucs less costs). togcthcr, this data can provide an excellcnt basc for propcrly valuing thesc functions as wcll as providing indica- tors of thcir incrcasing scarcity.

Notc that markct prices of cnvironmcntal functions may not always be a rcasonablc approximation of their social valuc. For examplc, in Indoncsia, uscrs of irrigation watcr are currcntly paying only 13% of thc full costs of supply; consc- qucntly, uscr chaugcs would not bc an accuratc mcasure of the valuc of irrigation. In such cascs whcrc obscrvcd priccs and costs dcviatc significantly from social valucs, shadow pricing should bc cmploycd.

l'radeable functions

If the niarkctcd functions of thc cnvironmcnt are also vadcable (such as cxport crops, timbcr, fish and shrimps in Tablc 2) thcn intcrnational priccs should be uscd to indicate thcir valuc. Thus the loss to society of thesc functions would bc rcprcscntcd as thc loss in foreign exchange earnings to the economy.

In addition, i t is increasingly uscful to have sufficient data (for instancc on cxport/import taxcs and transport costs) to calculatc thc cllcctivc ratcs of protcction affordcd to these functions by govcrnmcnt policy, which in turn havc asignili- cant impact on pattcrns of resourcc usc. For examplc, high cll'cctivc protcction riitcs for sawn tiinbcr and plywood wcrc an important clcincnt in thc rapid expansion of the timbcr proccssing industry in Indoncsia, which has had a profound influcncc on thc ratc of timbcr cxtraction (Gillis 1987).

Siniilnrly, high rates of protection for fruits and vcgctablcs cxplain their incrcrisingly widcsprcad cultivation in thc up- lands of Java, with varying impacts on soil crosion (World Bank 1987).

Project Appraisal Sepfcmber 19S8 147

Dow

nloa

ded

by [

Seto

n H

all U

nive

rsity

] at

15:

23 2

5 Se

ptem

ber

2014

Economic valuarion of environmental impacts

Non-marketed functions

The non-marketed functions of the environment (such as environmental protection and socio-cultural support func- tions in Table 2) arc, of course, the most difficult to valuc. As suggested by Dewees 1987, there are thrce basic approachcs:

Valuing non-marketed functions through the value of associated market goods. Valuing non-marketed functions through ‘ willingncss-to- pay’ survzys; these are surveys and expcrimcnts in which individuals reveal eithcr the amount that they would pay to secure the bcnefits or the amount they would dcmand to give up the benefits. Valuing non-marketed functions through the cost of social programs and other investments that would bc neccssary to ‘restore’ the functions or substitutc for the loss of thcm.

The appropriatencss of any one of thesc approaches will dcpend on the nature of the environmental function to be valued. It may also depend on the inherent limitations of the approach in a developing country context. For cxample, willingness-to-pay surveys are thought to have limitcd appli- cations in Indoncsia, mainly in urban areas whcre there are large samples of income-earning households with high litcr- acy rates.

On the other hand, more than one approach may be cm- ployed to provide estimates of the value of an cnvironmcntal function. For example, the value of the environmental protec- tion function of forests, trees and othcr vegctativc covcr in providing soiVwatcr conservation and windbreaks in upper watersheds may be approximated by both estimates of the costs of soil erosion in terms of agricultural output losses (plus any off-site costs where estimatable) and the costs of rcforcs- tation programs in upper watersheds.

Thus a recent study of the on-site agricultural productivity costs of soil erosion on Java suggests an annual loss of US$324 million (Arens and McGrath 1987), plus an addi- tional US$25 million to US$91 million in off-sitc sedimcnta- tion costs. In conlrast, total regreening cxpenditures in Javan watersheds for one year, 1982-83, were USS23.35 million (World Bank 1987).

These figures would thcrefore suggest an upper and lower bound on the value of upper watershed protection afrordcd by forcst andother vegctativecover. Othcr important methods of valuation include travel cost method, propcrty valuc studies, simulation modeling and contingent valuation.

Flexibility of data

A final criterion for data on environmental functions is flexi- bility. Data arc currcntly collected and aggrcgatcd by adinin- istrative/political unit; that is, from villagc to sub-district to

Standard methods need to be developed for translating administrative data with geographical data, to avoid widely differing results and to make such procedures a routine part of data collection

district to provincial and finally to regional and national level. Many cnvironmcntal and rcsourcc problcms, howevcr, re- quire information based on geographical/ecological bound- aries (such as soil erosion in upper watersheds, mangrove dcstruction in coastal zoncs, pollution discharges in major river basins).

Thus the dam collectcd by adminislrativc/political units needs to be sufficiendy flexible and dctailcd to allow it to bc easily re-aggregated along geographical/ecological lines. Standard methods for translating data in this manner need to bc developed to avoid widely differing rcsults and to make such procedurcs a routine part of data collection.

Another chcck on the flexibility of data is usefulness for monitoring and evaluation. The same information collccted to provide indicators of environmental impacts should have a dual function as indicators of sustainability for any subse- quent activity or program in rcsponse to thcsc impacts.

Similarly, it is incfficicnt and misleading to have one set of indicators for environmental impacts and anothcr for sus- tainability. For example, studies examining thc extcnt of industrial pollution in waterways may choosc to look at ccrtain ineasurcs of pollution stocks and discharge rates. Studics to monitor and evaluate any clcan-up efforts should thcrefore agree to use the same indicators.

Thus, to bc consistcnt, data collcctcd for economic valu- ation of environmenul impacts should also be uscd for moni- toring and evaluation efforts, and thc indicators should be carefully chosen to cnsurc fulfillment of this dual function.

Policy requirements.

So far, thc papcr has emphasized mainly the rolc of economic valuation of environmental impacts in project appraisal. As stresscd throughout, howcver, this may be only one of many possiblc functions of an cnvironmental and rcsource data base for dcveloping countries. For example, thcrc is increasing intcrcst incnsuring that theappropriatcquality of thedata base is consistcnt with the policy rcquircments for overall effective natural rcsource managcmcnt in dcvcloping countries. This in turn nccessarily imposes limits on data accuracy and preci- sion.

A recent study (Barbicr 1987) idcntified two over-riding nccds for natural rcsourcc managemcnt policics to be effec- tivc in Indoncsia, which in turn can bc gcncralized for all dcvcloping countrics.

Natural resource iniplications

Thcre is a nccd for subslantivc and cxtcnsivc analysis on the natural rcsourcc implications of various macrocconornic, lradc and scctoral policics.

Firstly, givcn thc incrcasing population prcssurc and cco- nomic dcmands on dcvcloping countrics’ natural rcsourcc basc, problcins arising from environincntal and natural rc- source dcgradation will continue to act as a constraint on thc successful implcmcntation of cconomic dcvclopmcnt poli- tics. Altcrnativc policy options that cxplicitly take thcse rcsource constraints intoaccount, thcrcfore, nccd tobc formu- latcd and analyzcd.

Sccondly, bccausc rnacrocconomic, tradc and scctoral policics influcncc and constrain what is possible to accom- plish at the program or project levcl, i t is csscntial for success- ful investment activity that analyscs arc conductcd of the natural resourcc nianagcincnt implications of thesc policics.

148 Project Appraisal Septcmbcr 1988

Dow

nloa

ded

by [

Seto

n H

all U

nive

rsity

] at

15:

23 2

5 Se

ptem

ber

2014

Economic valuation of environmental impacts

Economic costs As a general rule, more information should only be obtained on a particular At the micro levcl, there is a need for more analysis of thc

economic costs of environmental impacts. Firstly, micro-level analysis of natural resource allocation

decisions at the village or farmer level is needed in order to design appropriate policies and investment programs for natural resources management.

Secondly, such micro-level analysis is also important for monitoring the impacts of policy decisions and invesmcnt programs at the village and household lcvcls, not just in agriculture but all investment projects/programs. Although some of this information is sometimes available from rcscarch stations, independent, project and provincial studics, it nccds to be co-ordinated and reviewed consistently at thc national level for national policy and investment decisions.

Thcsc policy requirements for effective natural resource management should form the broad guidelines for the type of environmental and resource data base necdcd in developing countries. Attempting to assess and value the functions of the environment will not be a useful endeavor unless it can contribute to these policy requirements. The quality of the data collected must therefore be regularly reviewed to cnsure that it is consistent with these policy objectives.

In addition. the above needs for effective natural resources management policy suggest the broad institutional arrange- ments for this data base. On the one hand, to be useful for policy analysis, data on environmental and resource impacts must be co-ordinated and reviewed at the national level, which implies some centralized location of a national data base.

On the othcr hand, the need for more micro-level land project evaluation of the economic costs of environmental impacts also implies the routine collection and analysis at the local and sectoral level of data on these impacts. Thus the existence of local data bases, perhaps at the provincial or cvcn district level, linked and co-ordinated with a more centralized national data base center seems the most appropriate institu- tional arrangement.

Limits on data accuracy

The importance of both the accuracy and precision of data at the laboratory levcl has been stresscd by Acrtgeerts (1987), where

“accuracy can bc expressed quantitativcly through calculations of the absolute and relative errors, and is a measure for the difference between an analytically obtained datum or means of data and thc truc value. Precision rclates to the spread of individual data around themean and can be quantified by mcans of thc standard deviation of the measurement.”’

Thc accuracy and precision of data arc thcrcforc considcrcd to be essential for successful laboratory and ficld analysis. As a result, it may be thought desirable LO achievc the same dcgrce of accuracy and precision in the data bases nccdcd for natural resources managemcnt policy dccisions

However, for policy-making purposes, thcrc arc limits to thc accuracy and precision of the data actually rquircd. The general rulc is that more information should only bc obtaincd on a particular environmcntal problcm if the bcnclits to decision-making exceed the costs of collecting thc cxtra data. Additional information, or ‘refinements’ in existing inrorma- tion, that docs not assist policy-makcrs in arrivingata dccision

environmental problem if the benefits to decision-making exceed the costs of collecting the extra data

should not be collected. In many inslances, broad ‘ordcrs of magnitudes’ of the

economic consequcnces of environmental impacts may be sufficient to reach a policy decision - any additional infor- mation collected to improve the accuracy and precision of the data base for policy-making purposes would give rise to unnecessary costs.

This is not LO imply that precision, accuracy and general data quality control are not important, especially at the l a b ralory, experimental testing station or collecting level. As pointcd out by Acrtgeerts (1987), the costs of any analytical quality control program at this level should be compared against the expenditure the laboratory, and ultimately society, would incur in the production of data of lcsser quality.

Morcovcr, errors in data quality will be magnified as the data are compilcd and aggrcgated for use in larger data bases. In fact, if proper quality control measurcs are implemented at the laboratory and collecting levcls, then the data base should bc sufficienlly accurate and prccisc for policy-making pur- poses.

Conclusions

This paper has broadly oullined the data and mcthodology requirements for economic valuation of environmental im- pacts in dcvcloping countries. Although these requirements appcar fonnidablc, the growing dcmand for such valuation techniqucs at all levels of the planning and policy-making proccss will surely lcad to rapid improvcmcnts in existing (or in many cases non-existent) environmental and resource data bases in dcvcloping countrics.

Currcntly, thc most progress seems to be made in incorpo- rating cconomic valuation techniqucs for the environment in project appraisal analysis. This papcr has stressed, however, that this may not necessarily be thc only purpose that these valuation tcchniqucs should scrve. On the othcr hand, if the environmental and rcsourcc data base is also directed to a diffcrent purposc, say, for improving the effectivencss of natural rcsource management policics or devising a system of natural resourcc accounts, this has implications for the data and methodology required.

Finally, to be truly uscful for cconomic policy-making and planning, any environmental and resource data base for a dcveloping country must go beyond being just a set of ‘envi- ronmental indicators’. It must also tackle the more difficult task of valuing thc various economic and social functions of the cnvironmcnt. This is nevcrthclcss a most intcresting challcnge for those intcrcstcd in improving project appraisal and policy analysis of rcsource and environmcntal problcms in dcvcloping countrics.

Note

1. These definitions of ‘accuracy’ and ‘precision’ are respectively analogous to the definitions of ‘unbiased’ and ‘efficient‘ estimators in econometrics.

Project Appraisal September I988 149

Dow

nloa

ded

by [

Seto

n H

all U

nive

rsity

] at

15:

23 2

5 Se

ptem

ber

2014

Economic valuation of environmental impacts

References

R Ackermann, E B Barbier, G R Conway and D W Pearce. 1987, Environment and Sustainable Economic Development in lndone- sia-An Overview Report, draft report(World Bank, Washington, DC, December).

R Aertgeerts, 1987. "Data quality and robustness study", Environ- mental Sector Review Phase /I (UNDPIMinistry of State for Environment and Population, Jakarta, Indonesia).

P Arens, and W B McGrath, 1987, The Costs of Soil Erosion on Java - A Natural Resource Accounting Approach (World Resources Institute, Washington, DC, November).

E B Barbier, 1987, "Natural resources policy and economic frame- work, Annex 1 in J Tarrant eta/, Natural Resources and Environ- mental Management Review (USAID. Jakarta, Indonesia).

E B Barbier, 1988, "Data and analysis for implementing sustainable economic development in Indonesia", Environmental Sector re- view of Indonesia. Phase / / (UNDP/KLH, Jakarta, Indonesia).

D N Dewees, 1987, "Towards conceptual frameworks for economic evaluation of natural resources and the Environment for sustain- able development", working paper, Environmental Sector Review Phase /I (UNDPI Ministry of State for Environment and Popula- tion, Jakarta, Indonesia, 27 May).

J A Dixon and M M Hufschmidt (editor), 1986, Economic Valuation Techniques for the Environment (Johns Hopkins University Press, Baltimore).

J A Dixon, R A Carpenter, L A Fallon, P B Sherman and S Mano- pimoke, 1986, Economic Analysis of the Environmental lmpacts of Development Projects (Asian Development Bank, Manila).

M Gillis, 1987, 'Indonesia: public policies, resource management and the tropical forest". draft paper (World Bank, Washington DC,

M M Hufschmidt, D E James, A D Meister. B T Bower and J A Dixon, 1983, Environment, Natural Systems and Development: an Eco- nomic Valuation Guide (Johns Hopkins University Press, Balti- more).

I M D Little and J A Mirlees, 1974, Project Appraisal and Planning for Developing Countries (Heinemann. London).

A Markandya and D W Pearce, 1987, Environmental Considerations and the Choice of the Discount Rate in Developing Countries, Report for the Overseas Development Administration, London, forthcoming July 1988.

D W Pearce. 1986, "The sustainable use of natural resources in developing countries", Discussion Paper 86-1 5, Department of Economics, University College London, London, England, 5 December.

D W Pearce, 1987, The Meaning and Implications of Sustainable Development, Report of a Workshop on Sustainable Develop- ment, Economics and the Environment Secretariat, OECD, Paris, France, 12-13 November. Members of the Workshop were D W Pearce, R K Turner, E B Barbier, P Nijkamp and the Secretariat.

D A Phillips, 1986, "Pitfalls in estimating social discount rates: a case study", Project Appraisal, 1 (1 ), pages 15-20

L Squire and H G van de Tak, 1975, Economic Analysis of Projects (Johns Hopkins University Press, Baltimore).

UNIDO, 1972, Guidelines for Project Evaluation (United Nations, New York).

World Bank, 1987, lndonesia - Java watersheds: Java uplands and watershed management, draft report (World Bank, Washington, DC November).

May).

150 Project Appraisal September I988

Dow

nloa

ded

by [

Seto

n H

all U

nive

rsity

] at

15:

23 2

5 Se

ptem

ber

2014