Alison Butler

Alison Butler is an economist at the Federal Reserve Bank ofSt Louis. David H. Keh~’provided research assistance.The author would also like to thank Richard Blackhurst for hisassistance.

I

Environmental Protectionand Free Trade: Are TheyMutually Exclusive?

Having to compete in the United States in a totally free market atmosphere withcompanies and countries who have yet to develop such environmental standards isinherently unfair. It puts us into a game where the unevenness of the rules almostassure that we cannot win or even hold our own.

James E. Hermesdorf, Testimony to Senate Finance Committeeon Trade and the Environment, October 25, 1991.

‘-I3LAOMMENTS LIKE THE ONE cited above arebeing heard with increasing frequency. In fact,protecting the environment has always had im-plications for international trade. In 1906, forexample, the United States barred the importa-tion of insects that could harm crops or forests.Similarly, the Alaska Fisheries Act of 1926 estab-lished federal regulation of nets and other fish-ing gear and made it illegal to import salmonfrom waters outside U.S. jurisdiction that violat-ed these regulations- More recently, a U.S. lawrestricting the method of harvesting tuna toprotect dolphins has been the subject of a tradedispute between the United States and Mexico.

In recent years, as global warming and otherenvironmental concerns have multiplied, en-vironmental issues have played an increasing

role in trade negotiations, further complicatingwhat are generally difficult negotiations.Negotiating environmental regulations mul-tilaterally is especially problematic because ofdifferences in preferences and income levelsacross countries. What’s more, scientific evi-dence is not always conclusive on the effects ofcertain types of environmental degradation.Finally, environmental considerations can beused to disguise protectionist policies.

This paper examines the different waysenvironmental policy can have internationalramifications and their implications for interna-tional trade and international trade agreements.A general introduction to environmental eco-nomics is given, followed by an analysis of therelationship between environmental policy and

international trade. The paper concludes with adiscussion of the status of environmental con-siderations in multilateral trade agreements.

.~ i~.1Hliif~iL1U

The environment is used primarily in threeways: as a consumption good, a supplier ofresources and a receptacle of wastes.’ Thesethree uses may conflict with one another. Forexample, using a river as a receptacle of wastescan conflict with its use as a supplier of re-sources and as a consumption good.2 Wheneither the production or consumption of a goodcauses a cost that is not reflected in a marketprice, market failures that are termed “external-ities” may exist.’ Such market failures frequent-ly involve the environment.

A. C. Pigou, in The Economics of Welfare(originally published in 1920), presented one ofthe classic examples of an externality. In theearly 1900s, many towns in Great Britain wereheavily polluted by smoke coming from factorychimneys. Laundered clothes hung outside todry were dirtied by the smoke. A study done inthe heavily polluted city of Manchester in 1918compared the cost of household washing in thatcity with that of the relatively cleaner city ofHarrogate. According to the Manchester Air Pol-lution Advisory Board:

The total loss for the whole city, taking the ex-tra cost of fuel and washing materials alone,disregarding the extra labour involved, and as-suming no greater loss for middle-class than forworking-class households (a considerable under-statement), works out at over £290,000 a yearfor a population of three quarters of a million.~

Thus, a by-product of production—smoke—unintentionally had a negative effect on anothereconomic activity—clothes-washing.

Externalities exist when the social cost of anactivity differs from the private cost because ofthe absence of property rights. In the precedingexample, because no one “owns” the air, thefactory does not take into account the extrawashing costs it imposes on the citizens of thetown. As a result, more pollution than is sociallyoptimal will occur because the private cost ofthe smoke emissions to the firm (zero) is lowerthan the social cost (fl90,000 a year). In gener-al, if nothing is done about negative externali-ties, environmental damage will result asecologically harmful products are overproducedand the environment is overused.

To eliminate externalities, the divergencebetween the social and private costs must beeliminated, either by assigning private propertyrights (that is, ownership rights) or by directgovernment regulation. The approach taken oftendepends on whether property rights can beassigned.’ The advantage of assigning propertyrights to an externality is that it creates a marketfor that product and allows the price mechanismto reflect the value of the externality.

iIY,~i1flflIt~ ‘[H ./•.i.ss-[’[li[ii.g

Suppose a chemical factory locates upstreamfrom a small town and emits waste into theriver as part of its production process. Supposefurther that the town uses the river as itsprimary source of water. As a result of theseemissions, the town must process the waterbefore use. Clearly there is an externalityassociated with the firm’s use of the water—it isno longer usable to the town without cost. Ifproperty rights to the river could be assigned toeither the town or the firm, then the two par-ties could bargain for the most efficient level ofpollutants in the water.

‘There are many definitions of what constitutes theenvironment and therefore what is environmental damage.Production pollution results from the act of producing aproduct. Consumption pollution arises when the act of con-suming a product causes pollution. Deforestation reducesboth the capacity of the earth to naturally process carbondioxide and biological diversity. Elimination of a biologicalspecies also has environmental implications. Other thingsthat have environmental consequences include productsafety standards (such as limiting chemicals that can beused in agriculture) and soil erosion. This paper, unlessotherwise noted, focuses on production pollution, thesource of many trade-related disputes.

2For a more detailed discussion of these problems, see Sie-bert (1987).

‘This paper focuses only on negative externalities. Positiveenvironmental externalities occur when one use of the en-vironment costlessly enhances another. For example,cleaning a river for recreational use could also increase itsfunction as a supplier of fish.

4Pigou (1952), p.185, footnote 18.5Even if they can, social mores or standards may preventsuch an assignment. For example, people might be op-posed to selling timber companies the property rights to alltrees in national forests.

S

If property rights are assigned to the firm, thetown pays the firm to reduce its pollution. Thetown’s willingness to pay for reduced levels ofpollution depends on the benefits it receives fromcleaner water. Generally speaking, as the waterbecomes more pure, the additional (marginal)benefits to the town likely decrease. On theother hand, the firm’s willingness to reduce pol-lution depends on the costs it incurs to reducepollution by, for example, changing to a morecostly production or waste-disposal method.Generally speaking, as the firm pollutes less, theadditional (marginal) costs to the firm increase.The amount of pollution agreed upon will besuch that the added benefits to the town of afurther reduction in pollution are less than theadded costs to the firm of the further reduction.

If property rights are assigned to the town,on the other hand, the firm pays the town topollute. The firm’s willingness to pay for theright to pollute depends on the benefits itreceives from polluting. These benefits aredirectly related to the costs it incurs from usinga more costly production or waste-disposalmethod. Similarly, the town’s willingness to sellpollution rights depends on the costs it incursfrom additional pollution. The amount of pollu-tion agreed upon is where the additional benefitsto the firm of increasing pollution are less thanthe additional costs to the town of additionalpollution.

The Coase theorem proves that the equilibri-um level of pollution is the same in the preced-ing cases. Furthermore, such an outcome isefficient.°Thus, when property rights are clear-ly defined and there is an explicitly designatedpolluter and victim, the efficient outcome isindependent of how the property rights areassigned.

.tiiTi[H’[’[iT~flri.’[; 07 tHe Iheoren’[,

The key result of the Coase theorem, that theallocation of property rights does not affect theefficient amount of pollution, has limited appli-cation. If there are multiple polluters and/ormany parties affected by the pollution, the out-come can depend on how property rights areassigned. Similarly, if there are significant trans-

actions costs, such as measurement and enforce-ment costs, the Coase theorem may not hold!

Assume, for example, that two towns areaffected by the factory’s emissions, one furtherdownstream than the other. Suppose that thetown further away from the chemical plant haslower costs associated with cleaning the water.In this case, the amount of compensation thetowns would he willing to pay to reduce emis-sions by any given amount would differ. Thus,the allocation of property rights among the firmand the two towns would affect the outcome oftheir bargaining.

Suppose, instead, that more than one firm ispolluting. Determining how much pollution iscoming from each firm, along with ensuringthat each firm lives up to any agreement, maybe difficult and costly. If monitoring costs arehigh, the Coase theorem may not hold and theallocation of property rights again affects thechoice of optimal emissions.

The lack of general applicability of the Coasetheorem is not an indictment of using market-oriented incentives (which usually requiresassigning property rights). Most economistsbelieve that market-oriented solutions will leadto the most efficient use of resources because,rather than having the government attempt toestimate preferences, it allows the marketmechanism to reveal them.

7T1C.H1 .fl’[Hi.’[i”[if’[Ctfl

Property rights are not always assigned becausemany uses of the environment are consideredpublic goods. A pure public good is one that hastwo qualities: First, it is impossible or extremelycostly to exclude people from the benefits orcosts of the good (non-excludability). For exam-ple, even if a person does not contribute tocleaning the air, she still cannot be excludedfrom breathing the cleaner air. Second, the con-sumption of the good by one person does notdiminish the amount of that good available tosomeone else (non-rivalry). For example, the factthat one person is breathing clean air does notreduce the amount of clean air others breathe.In this case, property rights cannot be assignedbecause rationing is impossible.

6See Coase (1960). An (Pareto) efficient outcome is one inwhich no one can be made better off without makingsomeone else worse off. This type of economic efficiency,however, provides no information or guidance regardingequity issues. For a graphical analysis, see Nicholson(1985).

‘For a discussion of the limitations of the Coase theorem,see Baumol and Oates (1988).

While few uses of the environment are purepublic goods like air, many have enough fea-tures of non-excludability and non-rivalry tomake assigning property rights virtually impossi-ble. The functions of the environment that arepublic goods, such as breathable air and cleanwater, are summarized by the term environmen-

tal quality.

Regulating environmental quality is difficultbecause the government first needs to deter-mine the public’s demand for environmentalquality before deciding the efficient level of pol-lution. The free-rider problem that occurs withpublic goods makes this determination especiallydifficult. When people cannot be excluded fromuse, they have an incentive to understate theirwillingness to pay for environmental qualitybecause they can gamble that others will bewilling to pay. Similarly, if they are asked theirpreferences and know they will not have topay, people have an incentive to overstate theirdesire for a given public good. The degree towhich free-riding is a problem depends on thesize of the non-rival group affected. The largerthe group, the greater the free-rider problem.~

For the purposes of this paper, we willassume that to determine the “true” value ofpublic goods, the government measures thecosts of pollution reduction and the benefits ofpollution abatement accurately.° Using a cost-benefit approach, the optimal outcome is wherethe marginal cost of pollution reduction equalsthe marginal benefit of pollution abatement.’°

It is important to recognize that the sociallyoptimal level of pollution is generally not zero.Achieving zero pollution would require anextremely low level of production or an extreme-ly high cost of pollution control. In determiningthe optimal amount of pollution, both the coststo individuals and industry need to be takeninto account.”

1:1 XH1001e or ~

L’[ITH Oil’ I

Recall the previous example of a firm emittingpollutants into a river. Suppose the governmentdecides to regulate the industry because thereare too many polluting firms on the river todefine property rights adequately.” After deter-mining the socially optimal level of pollution,the government imposes a per-unit tax on emis-sions to reduce pollution to the optimal level.”

What happens to production? Figure 1 showsthe supply and demand curves for the indus-try’s output. The effect of the tax is to shift thesupply curve the distance AB (the additionalper-unit cost of output given the new tax).” Theprice rises from P, to P,, and the quantity ofoutput falls from Q, to Q, which is the outputlevel associated %vith the efficient emission lev-el” Emissions are reduced and environmentalquality improves.

8How to avoid this problem is the source of vast literaturein economics and is not discussed in detail here. For adiscussion of the free-rider problem in valuing publicgoods, see Browning and Browning (1983), or any otherpublic finance textbook.

°Significantproblems face governmental agencies trying todetermine the optimal amount of environmental quality.For a discussion of these issues, see Baumol and Oates(1988), Siebert (1987), and Anderson and Blackhurst(1992). For an evaluation of how successful currentmethods are in the United States, see OECD (1991).

“For a more detailed study of cost-benefit analysis, see Mi-shan (1971).

‘1While it is difficult for many people to think of placing amonetary value on health and life, in reality it is done allthe time. For example, though many lives are lost in carseach year, people are not willing to pay the - ‘costs” ofoutlawing cars to save those lives. For an excellent discus-sion of this issue, see Blackhurst (1977), footnote 18.

‘2For simplicity, we assume all firms on the river producethe same product and constitute the entire industry. Thisanalysis can be generalized, but it greatly complicates thegraphical analysis.

“This analysis assumes that the cost of reducing pollutionper unit is the same across firms in this industry. Oneproblem with imposing a per-unit tax, however, is that thecost of reducing pollution can vary significantly acrossfirms. One innovative approach to finding the most eff i-cient way to reduce pollution to a given level is the tradingof emission permits. In this case, the government decidesthe maximum amount of each type of pollutant that can beemitted overall and distributes permits to firms, allowingthem each a certain level of polluting emissions. The per-mits can be traded among firms, which allows firms to usefirm-specific information to set their own level of pollution.This enables firms for which installing pollution controls isrelatively inexpensive to sell emission permits to firms thatfind it more expensive to install pollution-reductiondevices. For a discussion of the theory of emission trad-ing, see Tietenberg (1990) and Nicolaisen, Dean andHoeller (1991). For a discussion of the effectiveness ofemission trading in the United States, see OECD (1991).

‘~Thisassumes that the per-unit emissions tax increases thecost of production proportionately.

‘5Other means of reducing pollution, such as a tax credit forpollution reduction, may not result in lower output in theindustry.

Figure 1The Effect of an Emissions Tax on Industry Price andOutput

l’nce

B

Pollution can have international effects in twoways. First, it might be localized within nationalboundaries hut, through the impact of environ-mental policy, affect a country’s internationaltrade. On the other hand, pollution may hetransported across borders without the consentof the countries affected (so-called transfjontierpollution). These two types of environmentaldamage have different effects on internationaltrade and, therefore, are discussed separately.

~. .h. .7Th Cor.mlries

Countries trade because of differences incomparative advantage. The idea of comparativeadvantage suggests that, given demand, coun-tries should export products that they canproduce relatively cheaply and import productsfor which they have a relative cost disadvan-

tage. Traditional international trade models ig-nore externalities such as non-priced uses of theenvironment.

By not explicitly including the environment asa factor of production, the costs associated withusing the environment are ignored. More recenteconomic models have extended the definitionof factors to include assimilative capacity, that is,the capacity of the environment to reduce pollu-tants by natural processes. The degree to whichthe environment will he affected by its use orby the production of ecologically harmfulproducts depends on its assimilative capacity.The higher the assimilative capacity, the less theenvironmental damage caused by the emissionof a given amount of pollutants. Assimilativecapacity can differ across regions and countriesand thus is an important factor in determiningthe effects of environmental use on trade.

Traditional trade models also ignore the non-priced use of the environment as a consumptiongood. This underestimates the value consumersmay place on the environment and thereforethe cost of using the environment for otherfunctions. These two factors can be significantin determining a country’s comparative advan-tage.’°

77: H0 ‘[tori/ri. L0OH7r~’O.Oi (ii (‘[(‘[SO

~flt,Ptt.KrKK:KK’[’[. 1,K:’ K~KKi~2K’•lf ~ ironnre.rrH.ul

OoaUhr

Assimilative capacity is one of the principalfactors affecting a country’s choice of environ-mental quality. In general, assimilative capacityis lower in industrialized countries because ofthe effects of past pollution. Less-industrializedcountries often have greater assimilative capaci-ties and thus can tolerate a higher level of emis-sions without increasing pollution levels. Popula-tion density and geography also affect a coun-try’s assimilative capacity. For example, the in-troduction of a polluting industry in a sparselypopulated area, all else equal, will likely not af-fect the assimilative capacity of that area asmuch as it would in a densely populated area.

Other factors can also affect a country’s will-ingness to accept environmental degradation.

leflecently some have suggested that the United Nationschange its system of national accounts to take intoaccount environmental resources. This can be particularlyimportant for countries like Costa Rica that have large en-vironmental resources (see “Wealth of Nature,” January18, 1992). A different system of national accounting could

take into account the costs of irreversible environmentaldestruction, so that, for example, the costs (as well as thebenefits) of rapid deforestation are accurately reflected inmeasures of output growth and wealth.

p.

A

ft ft Quantity

~.:7iT;:[) ./%S.i.”[.~i~.1C1IS0 ‘[K :,

S

IIII

‘I II

For example, poor countries may put a higherpriority on the benefits of production (such ashigher employment and income) relative to thebenefits of environmental quality than wealthycountries. As income levels increase, however,demand for environmental quality also rSs1’Thus, countries with similar assimilative capaci-ties might choose different levels of environmen-tal quality. As the example below demonstrates,environmental policies that result from differ-ences in countries’ preferences and income levelscan have significant trade effects.

Environmental Policy When Pollu-tion Is Within National Boundaries

How does environmental policy affect trade?Recall that, in the emissions tax example, thehigher production costs that resulted from thetax caused the price of the industry’s output toincrease and the quantity produced to fall.Assume there is a chemical industry in anothercountry producing the san-rn product with thesame level of emissions. For simplicity, assumethat, prior to the implementation of environ-mental controls, each industry produced justenough to meet its home demand, and the pricewas the same in both countries. As a result,trade did not occur1t Suppose, because of

different preferences, income levels or assimila-live capacity, it is optimal to impose environ-mental controls in one country but not in theother. What happens to price, output and en-vironmental quality in the two countries?”

The answer depends in part on whether thetwo countries can trade. If trade does notoccur, the effect is the same as in the previousexample. As figure 1 shows, in the countrywhere pollution controls were imposed, theprice will rise to P3 and the quantity of outputwill fall to Q2, while in the other country noth-ing changes. Figure 2 shows the effect of anemissions tax on price and output in the twocountries when trade occurs.2’ The reduction insupply of the chemical in the taxed country(Tax) will reduce the world supply of thatproduct, causing the world supply curve to shiftupward to the left. At the new world equilibri-um D, the price, P3, is lower than the autarkic(no trade) equilibrium price in Tax (P3), buthigher than the autarkic equilibrium price inthe other country, Notax (P.). At P,. consumersin Notax demand Q, but firms are willing tosupply C),- The distance ~2 is exactly equal tothe distance X1, which measures the differencebetween what firms in Tax are willing to supplyat P3 (Q) and what consumers demand at that

‘t$ee, for example, Grossman and Krueger (1991)leThs example generalizes to the case where trade occurs

before the implementation of eSronmental controls.lamis section assumes that each country is large enough to

affect the wodd price If the country with the emissions tax

was a small country, it could not affect thewodd price. Asa result, the effect on output in that country would begreater than in the example above

~‘Thlsassumes prices are in the same currency, so all ex-change rate effects are ignored.

FIgure2The Effect of at EmissIon. Tax on Industry Price and Output In a Two-Country Wcdd

Cawtyl (Ta) CajabyZ4Notax) WSd sçplyad D.mad

PS. Pa.

F’

PS.

Ill

PSF’ :~tnnt~

Ill PIII

F,PS

‘4

V ~3Q, Qunlity

a’ ~, — - - ---- - - ——

a

price (0). As a result, Notax exports the quanti-ty X2 of the chemical to Tax.

What Is the effect on other economic varia-bles? Consumption of the chemical falls inNotax, even though output rises. In general,because of the increased production in Notax,there will be an increase in pollution emissionsin that country. How much the pollution levelactually increases in Notax (if at all) depends onthe assimilative capacity and the method ofproduction used in that country. Whether thepeople in Notax are better off at the potentiallyhigher level of pollution that resulted fromincreased production depends on that country’swillingness to accept higher pollution for higherincome.

Pollution declines in Tax. If the assimilativecapacity is higher in Notax, world pollution willlikely be lower after environmental controls areImplemented. The effect on world employmentis ambiguous and depends on certain country-specific variables. The terms of trade will de-teriorate for the country with the emissions tax.

If the new level of emissions in each countryis optimal given preferences and income, bothcountries are better off by trade- The taxedcountry is able to consume more at a lowerprice than in the autarkic case, while the valueof total output rises in Notax. If measures ofnational income or wealth accurately reflectedenvironmental damage, they would increase inboth countries.

Does Environmental ProtectionDistort Trade?

One concern is that environmental regulationunfairly discriminates against domestic firmswhen they compete with firms in a countrythat has lower environmental standards. In theexample discussed above, an externality existedin Tax but by assumption, not in Notax. As aresult, introducing environmental controls elimi-nated a distortion that previously existed. Thischanged the flow of trade, but caused all thecosts of using the environment, both as inputsin production and as consumption items, to be

reflected in market prices. Thus, assuming thatenvironmental quality was not socially optimalbefore protections were enacted, pollution-inten-sive sectors in Tax were actually receiving animplicit subsidy from those who had been in-curring the external costs of pollution.

The difficulties in trying to determine theoptimal amount of environmental quality withina country, as discussed above, are substantial.The optimal level of environmental quality inone country is unlikely to be optimal in another,particularly if the two countries have signifi-cantly different income levels. Attempting toImpose one country’s environmental standardson another by using import restrictions doesnot allow countries to capitalize fully on theircomparative advantage3’ As discussed later,it Is also illegal under current internationaltrading rules.

Environmental Policy WhenPollution Crosses NationalBoundaries

The previous section discussed the interna-tional effects of environmental policy whenenvironmental damage is contained withinnational borders. Many other uses of theenvironment cause environmental damageacross borders, such as acid rain, which resultsfrom sulphur dioxide emissions, or worldwide,such as ozone depletion, which results primarilyfrom chlorofluorocarbons (CPC5). Transfrontierpollution may occur in essentially four ways:~

1. A firm’s production takes place in onecountry, but pollutes only in another.

2. Both countries have firms whose produc-tion processes pollute, but each country’spollution is experienced only in the othercountry.

3. Pollution occurs as a result of production inone country but the effects are felt in bothcountries.

4. Both countries pollute, and the pollutiongenerated by each is felt in both countries.

21ft Tax puts trade restrictions on imports of chemicals fromNotax because of the lack of emission restrictions in No-tax, both countries would be worse off. It for example, atariff was levied against imports from Notax, the earningsIn Notax from exposling the chemicals would be tower.Consumers In Tax would pay a higher price and Import alower quantity as a result of the tariff. For a detailed dis-cussion of the effects of tariffs on trade, see Coughlin,

Chrystal and Wood (1988). For a discussion of the possi-ble application of trade and polIcy measures in relation toenvironmental problems, see Subramanlan (1992).

~SeeLloyd (1992).

If pollution is of form 1 or 2, in the absenceof an international agreement, the pollutingcountry has no incentive to curtail its pollutingactivities by implementing an environmentalpolicy. If, instead, pollution is of the form 3 or4, pollution may be regulated domestically.Without taking into account the pollution in theother country, however, these controls will notlikely be optimal internationally. In the absenceof a globally optimal international agreement,domestic policymakers have less incentive totake into account the costs imposed on a for-eign country than if the costs were bornedomestically. Thus, from a global perspectivethere will be excessive use of the environment.

.Iflhit ~itt .0)001 .I’tiljt.IIV • :.he••1.~oc1t~aof jJOfllfrOfl thEir

Politthion

Suppose, as in case 1, the river being pollutedby the chemical firm runs directly into anothercountry and all the towns affected are in theforeign country. How is an appropriate policydetermined? Previously, we assumed that acountry weighed the costs and benefits of pollu.tion, given its preferences for environmentalquality, its income level and its assimilativecapacity. Unfortunately, in the case of trans-frontier pollution, this is no longer sufficient. Inthis case, domestic policymakers will be lessconcerned with the costs imposed on a foreigncountry than those borne domestically. In addi-tion, the desired level of pollution could differsignificantly between the two countries becauseof their preferences and income levels. Other is-sues contribute to the difficulties in negotiatingan international agreement on pollution control.For example. should the polluter pay to reduceemissions or should the residents of the countryaffected by the pollution pay to induce the t’irmto reduce emissions?

In the early 1970s, countries belonging to theOrganization of Economic Co-Operation andDevelopment (OECD), the multilateral organiza-tion of the industrialized countries, adopted thePolluter Pays Principle (PPP) to deal with purelydomestic pollution.” This principle requires thatthe polluter bear the cost of pollution-reducingmeasures. This approach, hmvever, provides no

guidance about how to determine environmen-tal damage or what to make the polluterresponsible to pay for. For example, should apolluter be responsible for damage that hasalready occurred, or should it be required onlyto pay to reduce current emission levels? inaddition, PPP offers no instruction regardingtransfers between governments to resolveproblems of transfrontier pollution.

As a result of an OECD conference on trans-frontier pollution, it was suggested that theOECD adopt the so-called “mutual compensationprinciple.” This proposal requires the pollutingcountry to provide an estimate of the costs ofpollution abatement for various levels of pollu-tion, while the polluted country similarly pro-vides an estimate of the costs of treating thedamages. An independent agency determinesthe optimal level of pollution with these twocost functions. Given the level of pollution setby the agency and the cost functions providedby the two countries, the polluting (polluted)country pays a pollution (treatment) tax basedon the cost of clean-up (control) estimated bythe other country and is also required to payfor the cost of pollution abatement (clean-up)in their own country. The advantage of thisapproach is that it induces countries to revealtheir “true” value of the environment.2~Unfor-tunately, because of the problems inherent indetermining the optimal level of pollution aswell as negotiating and implementing such aproposal, the mutual compensation principlehas never been used.

There are other impediments to reachinginternational agreements on environmental use.For certain types of environmental degradation,there is debate about how much damage isactually being done to the environment. Anobvious example of this is global warming.25

Many environmentalists and governments areconcerned that excessive emissions of carbondioxide, nitrogen oxide and methane gas fromenergy use are irreversibly warming the planet.Many others, including the U.S. government,however, feel that the evidence is insufficient atthis point and are unwilling to significantly altertheir environmental policy. Scientific evidenceon global warming is inconclusive. An August

23These countries are the 24 main industrialized countries.248ee OECO (1976) for an analytic discussion of why this is

true. For more information on the mutual compensationprinciple, see the discussion therein.

“For a discussion of the effects of global warming see, forexample, Winters (1992) and Schelling (1992).

31, 1991, survey on energy and the environmentin The Economist pointed out one of the difficul-ties with transfrontier environmental damagesuch as global warming: the appropriate policymay need to be implemented before conclusiveproof that the damage appears, because of thecumulative effects of some types of environmen-tal damage over time.

Nevertheless, some international agreementshave been reached (see table 1) and, if thesignificant increase in articles, studies andconferences on transfrontier pollution are anyindication, there will be additional pressures tofind new ways to deal with the increasingproblem of transfrontier pollution.

N OIIT.H•S(INJIJH ISSUES

One of the main reasons environmental policyaffects trade is because countries are at differ-ent levels of industrialization and thus havedifferent income levels, which can cause theiroptimal levels of pollution to differ. Because theinterests between high- and low-income coun-tries may differ, it is important to look moreclosely at these so-called North-South issues.2~

Currently the industrialized countries, ingeneral, are greater polluters than less industri-alized countries and thus tend to put a relativelygreater demand on worldwide assimilative capac-ity. One concern heard in developing countriesis that industrial economies, rather than reduc-ing their own demand for assimilative services,could impose their environmental standards ondeveloping countries without any assistance inpaying for them, thereby reducing the opportu-nity for less-industrialized countries to grow. Asone news commentator suggests:

Developing nations are suspicious that born-again environmentalists in the North will saddlethem with commitments to regulate pollution,slow down deforestation, and control popula-tion growth, all in the name of sustainable de-velopment. yet won’t follow through witheconomic aid to improve their own productivityand employment. Meanwhile, developed nations

are reluctant to undertake radical domestic [en-vironmentall policy changes that threaten theirown economic growth.”

Other types of environmental issues have aparticular North-South nature. For example,many of the world’s nature preserves are indeveloping countries in Africa. Currently, tradein elephant hides and ivory, along with otherendangered species, are prohibited under theConvention on International Trade in Endan-gered Species (CITES). At a recent conferenceon CITES in Kyoto, Japan, several African coun-tries argued that their elephant herds are largeenough to be culled without endangering thespecies. In addition, they argued, revenuegenerated by the sale of ivory and other elephantproducts is needed to fund future preservation.

Here, the interest of the industrializedcountries, who do not have a native elephantpopulation, is to protect an endangered species.The African countries, however, face a tradeoffbetween the benefits of protecting the speciesand the loss of revenue associated with the pro-hibition of trade in elephant products.” As aresult, less-industrialized countries are puttingincreased pressure on industrialized countriesto help pay for the services they are providing(such as species diversity).

In March 1992, the General Agreement onTariffs and Trade (GATT), the main bodyregulating international trade, released a reportentitled “Trade and the Environment” that takesa non-traditional approach to North-Southproblems. One hotly debated issue concerns theprotection of the rainforests, most of which arelocated in Latin America.” Industrialized coun-tries have moved to bar wood imports fromBrazil and Thailand, for example, as a way toreduce deforestation in those countries. GATTargues that, rather than barring imports ofwood products (much of which is GAT’I’-illegal),the industrialized countries should compensaterainforest countries for providing “carbon ab-sorption services.”

Although this approach is novel, its advantageis that poorer countries are assisted with financ-

26For a more complete discussion of North-South issues inenvironmental economics, see Walter (1975).

“Walter Truett Anderson, in Walljasper (1992), p. 159.

29flainforests are valued for, among other things, theb abilityto reduce carbon dioxide in the air and for the biologicaldiversity they contain,

“At the close of the Kyoto conference, the calls for partiallyopening trade in elephants and rhinoceroses were ignored

Table 1Examples of Multilateral Environmental Agreementswith Trade ProvisionsConvention Relative to the Preservation of Fauna and Flora in their Natural State. 1933~

Objective to preserve the natural fauna and ‘bra of tno world, particularly of Africa. bymeans of national parks and reserves, and by regular’on of hunting and collection of species

Trade Frov’sion: prohb’ts the import ano export of trophies unless the exporter is given acerV’icate permitting export Parties shall take measures to control and regulate in eacn of itsterritories the ‘nternal import ano export of ttpnies acquirea in a manner not in accord withnat~onallaw (Art. IX~

Convention on Nature Protection and Wildlife Preservation in the Western Hemisphere. 19402

Objective: to preserve all spec!es and genera 0’ native American fauna and flora from extinc-tion and to preserve areas of extraordinary beauty. strik1’g geological formation or aesthot.c.histor~cor scentific value.

Trade P’ovis’on provdes ‘or the regulation of trado in protected species by the issuance ofexport permits (Art lx)

African convention on the Conservation of Nature and Natural Resources. 1968~

Objective: to conserve, utilize and develop tne soil, water, floral and faunal resources o~tneAfrican cont.nent.

Trace P’ov sion’ for all species. a Party shall regulate trade in and the transport of speci-mens or trophies, and shah do so in such a manner as to prevent the illegal capture or ki!lingof these Trade in lrophes and transport of specimens of protectec species shall be subjectto a standa”d author.zation U) additional to that requireo for the hunting. killing, capture orcoJlection: ~.flwhich ndcates the dest~nation.fiil which shall not be given unless they navebeen legally obtained. flv~which shall be examined pnor to exportation. Parties w’ll make theimport and transit of such specimens or trophies subject to tne presentation of the authoriza.tion requ’red uncer U) and confiscate illegally exported specimens or trophies (Art IX. NOTE:Parties ae all members of the Organization o4 African Unity.

iru4 eii~rr~iiii.iiciital rlr’tllrt’tiobl ,0 tililt it (lilt’’ 11111 £4i’01Ji).~iiiicl ‘.iilii ij.it’l,. It) till’ I”’tlilli4 £~0\(‘l’ilrlltllls

uphlti’ .it till’ i’\JJCIISC üI er)nOlnic cIe~ElOpllti’Ilt. in l’\t’llUllge liii’ in e.’,trnent il local ell\ iI’lliiIllPhl—

I ins .lpprnarll ako redtli’e~the free—rider tal project’ rllritldiIlg liii’ ptll’rith.~eol laud thatpi’Ohllt’ill thdl i’iiabit’’ intich ol the ~~‘orld to i~hen turned 111(0 en~ii’oninerital pI’esel’\ e~.

hen’’l it ‘ruin the t’ar’hon ahsiirptiun sen icesprot uhd h\ riimlorest, and the di~ersil~ ol5perie~.rn’~tIed In ronnlne’ that art not theJJI’iillill’\ users at the ent lrunrnellt In additionhit apprclarh dir’ecti~ pr’oterK the I’aintoreslsrallier’ Ili.iti hm’ring certain t\ lies ot wood

1n’tidict’ts iii tin’ !lript’~1

Ili,it doing sri n ill ‘wise .~lP eseill . iiitei’ii~itioii~iliif’l’CC nic’nts do clotthe (‘\j)Ol I lint i’onlllrle’. to l)i’OU’l’l them £lilt)t\ LI I otii’tti’~ to dlsrrlnlinau’ Lntaln.’l ii’odtit’t~.

(lii tileil’ Jil OdlIt’lilIrl terhnRjtl1’~. I lidel’

Other’ approaches taken io irnprm e em iron— C \ I I har’ririg imports her’atlse tilt’ pn)dlIrloll

I1ll’iltLil sl.lild~1l’tis iii lo~~l’i’—lOt’oIili’. lc’s,.—iri,.Iiislr’i— riit’titod”, cv~&’tI(10 not InteL the .~lanciai(Is oi heali,.ed countries inrltide dt’hl—ior—nature Sn LIDs. iulpol’tihig t’oiiiitr~ i’ illegal. IN5 rule ha’. (‘nineI lere lnreign dc’ht i’ lllli’chZNed b~en~ir’ommrnenud under Ire reec’ ritl~ par’Lirui.ir’l~ in light iii the

‘°Fora discussion o# dect-~or.natureswaps and a partia1 Iot some o’ tiese arrargenonts see Jevl.r (1991)

Table 1 (continued)Examples of Multilateral Environmental Agreementswith Trade Provisions

Convention on International Trade in Endangered Species of Wild Fauna and Flora. 1973’

Oolect’ve: to protect endangerec species aqa.nst overexploitation through interrational trade

Trade Provisions: trade of species threatened wltn extinction (listed in Appendix U, ano t’adein species that may become endangered unless trade ;s strictly regulated Ilisted in AppendixII). is author-zed by export ano import permts approved by the Scientific Authorit,es of thePati~sconcerned fArticies III ano IV) Spec~esthat a Party identifies as being subject toregulation wtn,n ts own uj’isdiction and as requiring international cooperation to controitrace I hsted in Appendix 1W is suoject to an export pe”m.t authonzeo by the Sc-entif’cAuthority of the Party (Article V~.Art.cie XXIII permits a party to exempt itself from ther~qu~r~m~ntsof the convention with regard to a spec”ic sp’~ceslisted in Appendices I. II orIll. NOTE: CITES ouilds on a long nistory of controll:ng trace in endangered spec es throughthe issuC & export permits It adds tie twist of requinng an import permit for an export per-mit to be issued. in o’der to prevent circumvention to non-Parties

Montreal Protocol on Substances That Deplete the Ozone Layer. 1987~

Oblective’ to reduce ~nc eliminate nan-made emissions o ozone-depict ng substances.

Trade Provisions- trade prov!sions affect non-Parties onry. Parties are to ban the -mportationof controlled substances as o’ January 1. 1991. and bar the export of controlled substancesas of January 1. 1993. Part’es a’e atso to ban tne export of tne relevant tecnnology to non-Part.es Tie 1990 amendments. which are not in fo’ce, require Parties to ban the importat onof CF~-containng products as of January 1, 1993

SOURCES GATT (i992~and U S. Deparyrient of State (1991;.

‘The signatory countr’es are: Beig urn. Italy. Portugai. Soair. Sweden. United Kingdom. Egypt.South Africa Unizec Republic of Tanzania and lnd.a. Fm notes ‘egarcing certain counties.please consult onginal source.2The s’gnatory ccuntrres are. Un teo States. Argentina Brat I Ch:le. Costa A ca. Denlin:canRepuetic. Ecuado”. Ei Salvador. Guatemala. Ha t., Mexico. N caragua. Panama. ParaguayPeru. Surname. Trinidad ard Tobago. Uruguay and Venezjola For nores ega”dng certaincountries please consult origina1 sout~1Tne signatory counties are Alger a. Burkina Faso Cameroon. ~cnt”a A’nca’n Repuolic. Con-go. Djibouti. Egypt. Onana Ivory coast. Kenya. Liberia. Madagascar. Malaw’. Ma!i. Morocco.Mozamo:que. N’ger Nigera. Rwand~.Senegal Seycnebles. Sudan. Swazitand. Toqo. Tunis:a.Uganda. United Republic o’ Tanzan a. Za-re and Zamo.a. For notes egarding certa’n coun-tr.es. please consult original sou”ce4Tne signatory countries are’ all OECD countries except Greece. Iceland. lreiano and Turkey:Argentina. Braz-l. Costa Rica. Venezuela. congo. Egypt. United Republ.c o Tanzania. Za’ro.Z.mbabwe India. lndones,a. Malaysia Union o’ Soviet Socialist Republcs and lsrao, plus 73othe” develop’ng countr’es For notes regarding certain countr,es and br Appendices I. II andIll, please consult original source.

~Thesignatory countries are. All OECD countries except Turkey, Argentina, Braz.l. Mox’co,Egypt. Kenya, Soutn Af”-ca. Zambia. Maaysia. Tnaiiano. Uniun of Soviet Social st Repuol:csBulgaria. Czechoslovak.a and Hungary. plus 25 other deveioprrg countries Fo” notes regard-.ng certa n countries, please consult orig:nal source

controversial tuna-do)phin dispute between theUnited States and Mexico.”

The justification for prohibiting trade restric-tions based on the production method is to pre-vent countries from using such restrictions to

protect domestic industries. Unfortunately,GATT was not designed to address some of themore complicated issues of environmental pro-tection, particularly regarding productionmethods that could have transborder or global

31ln this case. the United States barred Mexican tunabecause the process by which it caught tuna killed moredolphins than is permitted by the United States. Accordingto GATT, however, the ban was illegal because the fishing

waters in question were not under U.S. jurisdiction. For adiscussion of the tuna-dolphin case, see GATT (1992) orthe original panel report.

GATT and the EnvironmentCVI 1 isa rrr~rItil~iter’alti-ado agreement ble under GAll’s rules to make access to

I hat sets the t ules for international t rude, ones own market dc’pendcn t on the dornest itp rot ides a r nec han is Ut liv which to settle ent’ irorimen tal policies or’ practices of therude (1 ISpLI t05 a nrcing Louitt tic’s, and conducts cx pot’ ting cotintn. “~ As a r’est.ilI, tWO prImax-

multilateral tr,icle nigotiatLons trailed Roundsl screes of conflict hate arisen bE’t\\’et~nen-to red Lice trade barrie N. ‘till! a greemei it. ti rontnen tat policy a rid GA’i” I’. ‘[‘he first is iiwhott ever is silent concerning the c’orriplirta- violation of the national treat rnent prOVisicin,Lions stemniilig from environrru’nta I policies which states that ireign firms producing or’hat hat e trade effects. ‘I he out art ide that selling a pr’ocltiet ill a oun U y’ nitist be treat-

remotely deals with the erit ir’onmc’n L is Ar’tr- ccl the same as a clornesttc firm (i.e., a I oreigndc XX which lists gene rat e.tcep I ions LU iimn located in another coun try cannot heGAl t r’tiles. I ‘ncler ;\rticle XX: subject to mores tringent em tronrireutal stan-

sirhic-ct iii liii’ r’c’quirc’rneril that strclr (lards thin a doriiestic firm) the second isrflcasui’VsiIt’c’ not apjiliect in a tilLit)lic’i’ t-ic,latiotis of’ the tuin—disc-i’uniriatory status~ hicti tvotild c’oristttutc’ a mt’aniS of ill’’ which states that ari~’trade concession madehit ‘alt or’ trnjtrstitiabte disc’r’imiriatitnr Er,’- to ~ C;Vj’l’ rrtember’ must be made to all.’3

tiveen cc,unitr’ies u Iu’r’c’ the s,inric’ c’oricliti itis for’ example, the trade prot isions of the \fon-pre~nil, or’ a dis,guisc’cl restriction oni rrflc’r- trea] Protocol stipulate different trade nreas-national truck’. nothing ri ttrus,~i~r’n’c’,rn,’nt ur’es lot’ sigti~ttoric’sand non—signatories“hall Is’ c’ilnstr ut’d to pr’c’t cut the anon-tion lii’ eiriur’ct’nrrc’nit bt ann c’onitrac tirig lother’tt’ise known as most—favored nattc,nt)11r’til.s of tlrc’iqil’e~ ..lit’t’P.SScII’V to protect (MIX) statusl. which is a \ rolatton of lion—human. aninurl or plant tile or health.1 cliscr’iniination.

(‘rider- (I A II’. cc it mid Cs carl regulate ‘l’here are set era I \t’at’ s that ci ivi torimerita]polluting ii tins in their’ otvn ount rv as tong issties could he explicitly addressed in CVI’!as no distinct jar i is iii ade between domes tic ‘the ru es could be amencled , fcn’ cxiirnple. toand foreign-ott tied fir’irts. lii fact nothing iii set u p a penalty st-stern for count r’ies that areG~Vl1 r’es t rict s a governmer it’s nut onoir rv in ~ietow some ag;’eect-ti pan level of erivi ronmert-such things Ll’~ taxa lion . ritgtita tion and subsi- tat st andar’ds. ‘l’his requires acceptance by two-dies as long as fore go and riomest tc I trots thirds of the contracting pa flies I in ni ucliare t I ‘ca ted ec{ Lid 1k. con tit ties it thereby I ,ecomes effective) and is

n r ‘ega i’d t ci pci] lut itt g (‘onsti tti ption goods effec(ftc in the cit her’ countries as t lies’ acceptIc’ .g. . (‘a n’s t h~it p n’ochIfl’ all ~0ll Li tion), it is Idie amend men t . ‘f’h dii lieu I t\ wi tti t his a~ilso GA’l”t—lc’giil to place controls on these proacFi is that a sufficient number of rric’riiber

prod tic ts or r’equ Ir’c’ ROllu Rn in cont rol dcc icc’s ~ Ith~must agree to the ‘ accep tat) Ic’’’ s tart-Is tnc Ii :rs ca ta vt rc curt ye r’ter’sl, as long as ciard s. Even if ther-e ar-c differ’e nt p roy isior ishc’se I egu Ialions ar-c applied ccl ua1k tO for’ dcc eloping arid inch is tria flied c’ou nt lies.

clornestic~ilIt: produced and inrported agreenrent is still tikc’tv Icr be clitt’icLtlt.produt’ls. ‘l’hft’~c’regulations teticl to be —

a pplic’d eqiuilly . causing I en er trade-mt-ta I c’cI Artot her por’sihuliIv is that specif it’ waiverspr’obletns with coristrrrrption pollution. rout’ be granted to alt 01’ some of the signiato—

According lii the r’ec’c’itt GA rI report cnn iv countries. this waiver n-equti’es a tn 0-

the cut ironriient. ‘ lii priricipic-. it is riot possi- thir’d~major’itt ot those tciting. as tong as the

‘GATT (1966) p 37‘GATT (1992~.p ID.

~Tnereare, however, GATT-sanctioned exceptions tonondiscrimination, such as the Generalized System ofPreferences for develop’ng countrres

15

nlujor’it\ c’onirlnn’isc’s at Ic’a’~thaIl cit thc’ coo— issues in ‘i Im-anisparent. niuihil1tten LII u at- anti1I i’ac’l ing parties. \\‘aiyt’t’s lion c’’ er, ar’c’ mt Lint’ tint u’,ccl to mast~lir’otn’c’tiumus[ pcilit’In’s.assummriecl tin he Iong-tei’nt sottitic,nts, hint

- . - . I he c’cnriren’mn that mmncntnt atn’d tin’ m cm-cotc’\ct’lntlonis liii’ a Irntiiti’cl ttmc. — . -

I I s[tic._t rs that I ‘axle 1itnltr’ies ItIsl1trt~ctlit’CAt’ I could aI”u clan-itt- whelhc’n- \Il- \ arid a trading par’tniem’’s c’nivi’onniicntal policies ~it’c

national Ii’eatnttn’nit. tt hic’li m’edltnin’n’s ‘iItral tm-rat- cnftc-it sinirpi’. ariothn’i’ tt at to IiI’alec’t a clonic’s—nilt’nit trim’ ‘‘likc- In’odunrts.’ considc’m’ pr’oclcmrts tic inctiistrt’ al tirc’ tnpc’risr’ of its fcn’eignithat air’ pn’nicltuc’c-cJ nith ~rgnil’ic’anitIvcliflc’r’c-mnt c’ximttpc’tilol’s. ‘l’lrc’ mrc’xL set ol’ C \‘t ‘I’ talks tt illcnn n’onnm’nital standards a” ‘like pm’cndinc’ts. ‘ If’ tikn’lt uttcniipt tin mrmt~n’c’xplic’it ‘tiles alirnintnot tlrt’_t ran urn sulijcr’t lo cli f’f’c’r’c’nI n’rgula c’nvnm’onmmental uiolc’ to In n’c’vcrtt diftcrtnrrc’stionts ‘l’hc aclt aotagc- of ill UI’ tIie~c’up- mt tht’sc’ pcniic’it’s ft-our ht’inig tnsi’cI a~a nonpn’ouc’hn’~is thai thc-~~ncIdr’n’-,’,c’ot in’onttrc’ntal It pe oI non-Rim-Hf han’n’icn -

effects.32 For a discussion of GATT regulationsand environmental protection, see the shadedinsert at left.

GATT’s recently released report on theenvironment attempts to address some of theseissues, Some have suggested, in addition, thatGATT focus the next round of talks on environ-mental issues (assuming the current “UruguayRound” of talks is successfully completed),~~TheUnited Nations-sponsored “Earth Summit” in RioDc Janeiro scheduled for this spring is also anattempt to increase international cooperation onprotecting the environment, particularly inregard to North-South issues,

This article examines the role of environmen-tal policy on international trade. Environmentalpolicy is justified because of the nature of exter-nalities associated with using the environment.When the divergence between the social andprivate costs of using the environment isignored, polluting activities receive an implicitsubsidy. Environmental regulations may changeinternational trade, but enhance social welfareby removing this subsidy. The optimal amountof environmental protection, however, candiffer significantly across countries because ofdifferences in preferences, income and assimila-tive capacities.

One important concern is that countrieswill use environmental policies as an excuse toestablish protectionist policies. As environmental

protection and environmental use take on amore transnational nature and the assimilativecapacity is reduced worldwide, new agreementswill have to be designed to both protect scarceresources and protect countries from being dis-criminated against because of how they chooseto use their environmental endowments domes-tically. As the recent GATT report suggested,however, it is possible to protect the environ-ment without distorting trade flows, Thus, freetrade and environmental policy are not mutuallyexclusive but can work together to encourageboth economic growth and environmental quaff-ty worldwide,

S2For a comprehensive discussion of the application of en-vironmental exceptions under GATT. see Charnovitz(1991), and Sorsa (1992).

33For an additional discussion of why GAIT should nookmore closely at environmental regulation, see Petersmann(1991)

7 7<

Anderson, Kym, and Richard Blackhurst, eds,, The Greeningof World Trade issues (University of Michigan Press, 1992).

Baumol, William J, and Wallace E, Oates. The Theory ofEnvironmental Policy 2nd ed. (Cambridge University Press,l988).

Blackhurst, Richard. “International Trade and DomesticEnvironmental Policies in a Growing World Economy,” inRichard Blackhurst et al, eds,, international Relations in aChanging World (Siithoff-Leiden, 1977), pp. 341-64

Browning, Edgar K., and Jacqueline M. Browning. PublicFinance and the Price System, 2nd ed. (MacMillan Publish-ing Co., Inc., 1983).

Charnovitz, Steve- “Exploring the Environmental Exceptionsin GAIT Article XX? Journal of World Trade (September1991), pp. 37-55.

Coase, R, H. “The Problem of Social Cost,” The Journal ofLaw and Economics (October 1960), pp. 1-44,

Coughlin, Cletus C., K. Alec Chrystal, and Geoffrey E. Wood.“Protectionist Trade Policies: A Survey of Theory, Evidenceand Rationale,” this Review (JanuarylFebruary 1988),pp. 12-26,

Devlin, Robert. “Debt-for-Nature Swaps: A New Agenda,”International Economic Insights (SeptemberlOctober 1991),pp. 33-37.

GATT. The Text of the General Agreement on Tariffs andTrade (Geneva, July 1986),

Press Release #1529, February 3, 1992.

Grossman, Gene M., and Alan B. Krueger. “EnvironmentalImpacts of a North American Free Trade Agreement?’ pre-pared for the Conference on the U.S7Mexico Free TradeAgreement, October 3, 1991.

Lloyd, Peter J. “The Problem of Optional EnvironmentalPolicy Choice?’ in Kym Anderson and Richard Blackhurst,eds., The Greening of World Trade Issues (University ofMichigan Press, 1992), pp. 49-72.

Mishan, E. J. Cost-Benefit Analysis: An Introduction (PraegerPublishers, 1971).

Nicholson, Walter. Microeconomic Theory: Basic Principlesand Extensions, 3rd ed. (Dryden Press, 1985).

Nicolaisen, Jon, Andrew Dean, and Peter Hoeller.“Economics and the Environment: A Survey of Issues andPolicy Options?’ OECD Economic Studies (Spring 1991),pp. 8-38.

OECD. Economics of Transfrontier Pollution (1976).______ Economic Surveys: United States. Protecting the En-

vironment (November 1991).Petersmann, Ernst-Ulrich. “Trade Policy, Environmental

Policy and the GATT Why Trade Rules and EnvironmentalRules Should be Mutually Consistent,” Aussenwirtschaft,46, Jahrgang (1991), pp. 197-221.

Schelling, Thomas C. “Some Economics of GlobalWarming?’ American Economic Review (March 1992),pp. 1-14,

Siebert, Horst. Economics of the Environment: Theory andPolicy, 2nd ed. (Springer-Verlag, 1987).

Sorsa, Piritta. “GATT and Environment?’ in David Greenawayand John Whalley, eds,, The World Economy (January1992), pp. 115-31.

Subramanian, Arvind. “Trade Measures for Environment:A Nearly Empty Box?” in David Greenaway and JohnWhalley, eds., The World Economy (January 1992),pp. 135-52.

Tietenberg, T. H. “Economic Instruments for EnvironmentalRegulation?’ Oxford Review of Economic Policy (Spring1990), pp. 17-33.

US. Department of State. lieaties in Force: A List of Treatiesand other International Agreements of the United States inForce on January 1, 1991 (September 1991).

Walljasper Jay. “The Anti-Green Backlash,” Utne Reader(MarchlApril 1992), pp. 158-59,

Walter, Ingo. internationai Economics of Pollution (HalstedPress, 1975).

“Wealth of Nature,” The Economist (January 18, 1992),p. 67.

Winters, Alan L, “The Trade and Welfare Effects ofGreenhouse Gas Abatement: A Survey of Empirical Esti-mates?’ in Kym Anderson and Richard Blackhurst, eds.,The Greening ot World ~ade Issues (University of MichiganPress, 1992), pp. 95-114.

Pigou, Arthur C. Economics of Weifare, 4th ed. (MacMillanand Co., 1952).