Chapter 10

1. Chapter 10 Trade and the Environment

2. Key Questions • What are the classical theorems of international trade and the implications of extending them to environmental resources? • When does trade liberalisation damage the environment and if so does it matter? • Do countries competitively reduce environmental standards and if so for what purpose? • Does banning trade in products made from endangered species necessarily augment stock size? • Does adherence to the General Agreement on Tariffs and Trade unduly constrain attempts to tackle environmental problems? • Does the empirical evidence suggest that differences in the stringency of environmental regulations are a major determinant of trade flows?

3. Viewpoints Concerns expressed by environmentalists about international trade: • Will gains from trade liberalisation be outweighed by the damage to the environment? • Is there an incentive for countries to engage in ‘ecological dumping’ where one country lowers its emissions standards to gain some competitive advantage? • Will footloose industries migrate to ‘pollution havens’? • Will countries competitively lower environmental standards to generate trade benefits resulting in a ‘race to the bottom’? • What about the environmental impact of transporting goods internationally? Businessmen often argue that environmental regulations: • will result in the loss of jobs, and that • the production of pollution intensive goods will merely be pushed offshore. Politicians are concerned about the resulting unemployment. • Debate about the impact of international trade on the environment grew noisier in the early 1990s, mainly in response to fears about the possible impact of the proposed North American Free Trade Agreement (NAFTA) between Canada, the US and Mexico. Standing on a populist platform former US presidential candidate Ross Perot claimed that NAFTA would result in a giant sucking sound of jobs flowing to Mexico and recommended that Mexican goods be barred from entering the US unless they were produced under conditions meeting US standards for environmental protection. The exchange between Daly (1993) and Bhagwati (1993) is between two of the chief protagonists in the debate over NAFTA.

4. Debate • Debate about the impact of international trade on the environment grew noisier in the early 1990s ... • Mainly in response to fears about the possible impact of the proposed North American Free Trade Agreement (NAFTA) between Canada, the US and Mexico. • Former US presidential candidate Ross Perot claimed that NAFTA would result in large scale flow from USA to Mexico and recommended that Mexican goods be barred from entering the US unless they were produced under conditions meeting US standards for environmental protection. • The exchange between Daly (1993) and Bhagwati (1993) is between two of the chief protagonists in the debate over NAFTA.

5. Free Trade and Welfare • Many economists believe that free trade improves economic welfare. • Some welcome the fact that industries might wish to relocate in countries with low standards. • For them it just demonstrates that differing endowments are the basis of trade. • Economists do not generally accept the argument that countries ought to share the same environmental standards, given the existence of large differences in per capita incomes and population densities between countries. • Many economists therefore would see no reason for biophysical standards and environmental taxes to be harmonised. • Many would also contend that although changes in environmental regulation might result in temporary unemployment, changes in the exchange rate will ultimately restore the trade balance. • Calls for trade restrictions made on environmental grounds should be viewed with suspicion as they might be motivated by protectionist concerns.

6. Structure of this Chapter • Review of the traditional theory of trade • Extension of this theory to incorporate environmental resources. • Why might a country have an advantage in the production of environmentally intensive commodities. • Partial equilibrium analysis of the effects of moving from a position of autarky (no trade) to a system of free trade with and without the requisite controls on the production of environmental externalities. • We show that the problem is the absence of economically efficient environmental policy rather than trade liberalisation per se. • General equilibrium model analysis of the effect of trade liberalisation on welfare. • Consideration of situations in which governments might attempt to manipulate environmental standards in order to benefit national producers. • Development of a model in which jurisdictions compete in order to attract capital. • Is this likely to result in an erosion of environmental standards? • The role played by the World Trade Organisation (WTO) in policing the global trading system. • What, according to the WTO, is the difference between justifiable and unjustifiable restrictions on trade enacted in the name of environmental protection? • Brief overview and critique of attempts to test the proposition that environmental regulations explain trade flows and patterns of Foreign Direct Investment (FDI).

7. Traditional trade theory • Traditional trade theory characterised by a number of theorems. These include: • Heckscher-Olin theorem • Stolper-Samuelson theorem • Rybczinski theorem • Factor price equalisation theorem. • These are covered by any standard textbook on international trade e.g. Bhagwati and Srinivasan (1983). • Like Rauscher (1999) and Steininger (1999) we wish to consider the implications of these theorems when environmental factors of production are brought into the picture alongside capital and labour.

8. The Heckscher-Ohlin theorem • Afundamental proposition regarding the pattern of trade between two economies. • The Heckscher-Ohlin theorem suggests that trade is determined by differences in factor endowments. • A country will export those goods relatively intensive in its abundant factor of production and import those goods relatively intensive in its scarce factor of production. • For example, a country having an abundant supply of capital will find it cheaper to manufacture goods whose production is capital intensive. • The country is said to have a ‘comparative advantage’ in the production of such goods. • Post free trade the consumption of the labour intensive goods will increase because of an increase in national income and a positive substitution effect whereas the consumption of capital intensive goods may or may not increase. • Taken as a whole, consumers unambiguously reach a higher level of welfare.

9. Heckscher-Ohlin theorem: assumptions The theorem rests on a number of assumptions: • Countries have identical constant returns to scale technologies. • Countries have identical tastes. • There are no impediments to trade and no transport costs. • There are two goods, two factors of production available in fixed quantities, and two countries one of which is small and the other one large representing the rest of the world. • Factors of production are immobile.

10. Environmental extension (1) • How can the HO theorem be extended to include environmental and natural resources as factors of production? • The exports of many countries reflect their exploitation of particular natural resource endowments. But the observation that countries with, say, deposits of bauxite should ‘specialise’ in the production of bauxite ore is not especially insightful.

11. Environmental extension (2) • But environmental resources include air, soil and water quality, and the capacity of the environment to assimilate the unwanted by products of economic activity. • From our perspective a country is therefore well endowed with environmental resources when it is sparsely populated and has a higher assimilative capacity. • HO theorem implies that countries well endowed with environmental resources should specialise on the production of environmentally damaging goods.

12. Endowment & Regulation • Ultimately, a country’s endowment of environmental resources is determined by environmental regulation. • The stringency of environmental regulation should reflect the population’s appetite for environmental quality. • The stringency of environmental policy is endogenous so the growth in income that accompanies trade liberalisation might increase the stringency of environmental policy and therefore reduce the endowment of environmental factors of production.

13. Harmonisation of standards? • Differences in the abundance of environmental resources and differences in the appetite of populations for environmental quality make it fairly clear that the harmonisation of environmental standards is inappropriate. • There are perfectly legitimate reasons why Governments should wish to set different environmental standards.

14. Impact of Regulation on Trade • HO theorem predicts that tighter environmental regulation at home leads to increased production of environmentally intensive products abroad. • These environmentally intensive products are then imported.

15. Topical Example • Asingle country or a bloc of countries imposes restrictions on CO2 emissions. • This should result in an increase in CO2emissions elsewhere, (‘carbon leakage’). • Economists have utilised CGE models to estimate the extent of carbon leakage associated with attempts to reduce carbon emissions by, say, OECD countries or the EU. • It is easy to understand why the threat of carbon leakage might prompt policy makers to: • (a) think about trade restrictions to prevent the import of carbon intensive commodities and • (b) reconsider the wisdom of unilateral measures to reduce carbon dioxide emissions.

16. Other bases for trade in waste and residuals Differences in: • population density • the waste assimilative capacity of countries • national preferences for environmental quality also provide a basis for trade in waste and residuals.

17. Brief comments on some other theorems

18. Stolper-Samuelson theorem • The HO theorem explains why a country is able to increase its income through trade. • But this does not mean that free trade will be advantageous for everyone. • The Stolper-Samuelson theorem demonstrates that those who supply the scarce factor of production can gain through protection that restricts imports of goods intensive in that factor. • Atariff will increase the income of the factor used intensively in the good that receives protection. • Intuition: the resulting increase in the price of the good increases the derived demand and therefore the price of the intensively-employed factor.

19. Applicability of Stolper-Samuelson • Many environmental resources are not privately owned. • But where environmental resources are subject to private property rights (perhaps because these have been created by the Government in the form of tradable permits) then the Stolper-Samuelson theorem may apply.

20. The Rybczynski theorem • An increase in the endowment of one factor will reduce the production of goods intensive in the other factor. • For example, if there is an increase in labour the production of capital intensive goods will decline. • A de facto increase in environmental resources caused for example by an increased allocation of environmental permits by an incoming Government that cares less for environmental quality should decrease the production of environmentally less-intensive commodities.

21. The factor price equalisation theorem • HO suggests that free trade would equalise factor prices since a country would export goods intensive in its abundant factor of production and import goods intensive in its scare factor of production. This would decrease the derived demand for the scarce factor of production and increase the demand for the abundant factor of production thereby reducing international differences in factor prices. • Despite the fact that there is much capital mobility, and quite a lot of labour mobility, observation suggests that factor price equalisation does not hold. Unskilled workers in the developed world continue to earn far more than equivalent workers in Less Developed Countries. This could be attributed to the rather rigid assumptions upon which the theorem, and indeed the Heckscher-Ohlin theorem, is based e.g. the existence of free trade and the assumption that production technologies are the same everywhere. • The factor price equalisation theorem would seem to suggest that international trade equalises the shadow price of environmental resources – a striking result. The shadow price of environmental resources means environmental tax rates or the price of environmental permits. Note that this does not require tax rate harmonisation or trade in permits. This of course does not mean that environmental quality will be equalised.

22. 10.1.1 North-South models of trade and the environment • So-called North-South models suggest that the source of the poor South’s advantage is not that they have abundant environmental resources but that the use of these resources is unregulated by the Government. • Such models are associated with the seminal paper of Chichilnisky (1994). This gives the South an illusory comparative advantage in the production of environmentally intensive goods. It is customary to refer to the region suffering from absent property rights governing access to the environmental resource as the ‘South’ and the region in which property rights both exist, and are enforced, as the ‘North’. • Due to the mismanagement of environmental resources trade may exacerbate the environmental problems of the South whilst the rich North benefits from trade.

23. 10.2 Does free trade harm the environment? A partial equilibrium analysis • The preceding section extended classical theories of trade to include environmental resources. • We will now address the twin questions of (a) whether trade is bad for the environment and (b) whether it matters in a partial equilibrium setting in which the production of a single good gives rise to external costs. • For the sake of simplicity these external costs are assumed to arise from the production of the good rather than the use of particular production processes. • The country is assumed to be small such that it cannot affect its own terms of trade. Factors of production are assumed to be immobile and there is no transboundary pollution.

24. Cases to consider • This analysis follows closely that of Anderson (1992). • Assume that the country switches from an initial position of autarky to one of free trade. • Four cases to consider • the country already imposes ,and retains, an (allocatively efficient) environmental tax or permit quantity of the correct amount and becomes a net importer • the country already imposes, and retains, an environmental tax and becomes a net exporter • the country does not already impose, nor introduces, an environmental tax and becomes net importer • the country does not already impose, nor introduces, an environmental tax and becomes net exporter. • Liberalising trade in the good whose production harms the environment can be welfare enhancing or not depending upon whether the country already imposes the efficient tax rate on domestic production.

25. Cases c) and d): efficient tax not imposed • the country does not impose an environmental tax and becomes net importer • the country does not impose an environmental tax and becomes net exporter. • Begin with cases where the country does not impose an efficient environmental tax, and trade is liberalised in the good whose production harms the environment. • If the good ends up being imported then welfare is definitely improved. This is because real resource costs fall whilst at the same time domestic pollution levels fall.

28. Case (d): Figure 10.2 • The horizontal line P is now higher, indicating that opening up to trade will result in a higher price for consumers and increased domestic production leading to exports. • Production rises from Q to Qx whilst consumption falls to Cx. The change in economic surplus is given by eik although there is an increase in environmental costs of dekm. • The relative size of these areas and therefore the overall welfare impact is unclear. • Consequently if a country without environmental taxes ends up being a net exporter then the gains from trade may be more than outweighed by the losses from increased production and the external costs that this generates – a situation of some concern.

29. Cases a) and b): efficient environmental tax already imposed • If the pollution tax is already in place when trade is liberalised then benefits from trade are secured irrespective of whether the final position of the country is one of exporter or importer. • For the case in which the good ends up being imported, Figure 10.1 shows that the change in welfare is qcf. • For the case in which the commodity ends up being exported, Figure 10.2 shows that the change in welfare is cij. • So even if production is increased to supply the export market the benefits from trade more than outweigh the monetised losses from increased pollution.

30. Is a pollution tax superior to a tax on exports or a subsidy to imports as means of controlling environmental problems? • Suppose that rather than imposing an environmental tax on pollution, a tax is instead imposed on the exports of the polluting commodity. This reduces the price faced by domestic consumers of that commodity, meaning that they will now purchase too much of the good. More specifically, they value infra marginal units of the good less than do foreign consumers. Accordingly, whilst trade taxes and subsidies can influence domestic pollution levels they do so by imposing higher costs. • What happens to the small country when emissions taxes are placed upon the production of an environmentally damaging commodity in the rest of the world? Does this increase or decrease the benefits from free trade to the small open economy? • If the country is already an exporter of the commodity, Anderson shows that an increase in the international price of the commodity increases pollution levels in the small economy. However, the benefits to the economy are always positive if the correct tax on pollution is already in place since the gains from trade always outweigh the losses from more pollution. • If the appropriate tax is not already in place there could be either benefits or disbenefits since the quantity of pollution is already excessive. • If, on the other hand, the country is an importer and remains an importer then the gains from trade are reduced.

31. Summary of the partial equilibrium analysis • Free trade is better than no trade provided that the requisite efficient environmental taxes, or their equivalent, are in place – provided, that is, that the externality is efficiently internalised. • Taxes or subsidies on trade could alter domestic production levels and hence pollution but they are inferior to taxing environmental pollution. • Countries have no incentive to depart from the economically efficient level of tax placed on the production of environmentally damaging goods. • What about the assumptions underlying the analysis? • As developed countries raise their environmental standards the production of environmentally damaging goods will move to developing countries. Eventually developing countries will become exporters of environmentally damaging products. But any further increases in international standards will benefit developing countries. • The ability to import or export factors of production – assumed absent in the foregoing discussion – in response to changes in environmental targets actually enhances the gains from trade. • The assumption that pollution does not cross national boundaries is important since with transboundary pollution increases in pollution affect all countries. Such transboundary pollution would, however, occur even in the absence of trade.

32. Limitations of partial equilibrium models of trade and the environment • Partial equilibrium models are used for the purposes of analysing the effects of liberalising trade in one sector of the economy, when that sector is small in relation to the rest of the economy. • But there are important questions that a partial equilibrium approach cannot answer. • What happens when trade is liberalised across several sectors of the economy at once such that the economy grows? • What happens to the shadow price of pollution? • What happens to economy-wide emissions? • Answering these and other questions requires a general equilibrium approach in which all markets clear simultaneously.

33. 10.3 General equilibrium models of trade and the environment • We show the derivation of a general equilibrium level of pollution in an open economy as a function of the general equilibrium marginal abatement cost curve and marginal damage curves. • It goes on to consider the general equilibrium effect of trade liberalisation on economic welfare.

34. The Copeland and Taylor Model • Deals with a small open economy facing prices p for two commodities x and y. • Production of commodity x results in pollution and is referred to as the dirty good. • Production of good y produces no pollution and is the clean good. • Economy has at its disposal L units of labour and K units of capital. • Emissions are denoted z and are proportionate to potential output of good x. • Emissions can be reduced by sacrificing some of the output of good x. • Reducing emissions shifts the economy’s production possibility frontier.

35. Model Specification (1) • The national income function G of the economy has as its arguments world prices, the level of both factor inputs, and permissible emissions • G returns the maximum value for national income which can be achieved given prevailing world prices, available factor inputs and production technologies, and maximum allowable emissions. • Differentiating the national income function with respect to emissions yields the general equilibrium marginal abatement cost function. It indicates by how much national income would increase if emissions were allowed to rise by one unit. • Associated with the general equilibrium marginal abatement cost function is the marginal tax rate τ necessary to ensure that emissions do not exceed z.

36. Model Specification (2) • We turn now to the consumer side of the economy in which there are N individuals. •  The indirect utility function V of each individual is given by where I=G/N is per capita income. • The indirect utility function gives the maximum amount of utility as a function of prevailing prices, available income and emissions (which impact negatively on utility).

37. Deriving the Model Solution (1) • The indirect utility function gives the maximum amount of utility as a function of prevailing prices, available income and emissions (which impact negatively on utility). • This function is maximised subject to the constraint linking national income to permitted emissions.

38. Deriving the Model Solution (2) • The first order conditions include • Given the assumption that world prices are wholly independent of national emissions rearranging equation 10.4 yields

39. Deriving the Model Solution (3) • The RHS of (10.5) is the Marginal Damage (MD) from emissions. This will be a function of prices, per capita income and emissions. • For optimality we already know aggregate marginal damage should be set equal to the tax on emissions so

40. Deriving the Model Solution (4) • The optimal level of emissions is therefore implicitly defined by • The optimal level of emissions and the emissions tax rate are jointly determined by the intersection of the ‘supply’ of emissions (i.e. the aggregate marginal damage curve) and the ‘demand’ for emissions (i.e. the marginal abatement cost curve).

41. The Copeland and Taylor model and the EKC • This framework can be used to trace out the relationship between national income and emissions – the environmental Kuznets curve. • There is no straightforward relationship between national income and emissions. • Both z and G are endogenous variables and the precise relationship between national income and emissions depends on what underlies the economic growth. •  If economic growth is caused by an accumulation of human capital then the economy will increasingly specialise in the labour intensive non polluting commodity (Rybczynski T). Hence demand for pollution falls whilst the growth in income causes the supply of pollution to fall. These effects are reinforcing => a negative relationship between economic growth and emissions. • If growth in income is caused by an increase in the capital stock then the demand for pollution increases. This may be more than sufficient to offset any income-induced reduction in the supply of emissions => positive relationship between economic growth and emissions. • Economic growth first characterised by increased capital and latterly by increased human capital generates the stylised inverse U-shaped relationship between income and emissions - the EKC. But that is only one of many possible outcomes.

42. The effects of trade liberalisation • What do C&T mean by trade liberalisation? • Trade liberalisation means either reducing the level of some tariff which has been placed on some imported commodity or alternatively, a reduction in trade frictions. • In order to export a unit of a commodity an amount (1+ρ) units has to be shipped, where ρ>0. This is known as the ‘iceberg’ model for obvious reasons. • This generates a real resource cost to international trade representing perhaps, the transportation costs of international trade.

43. Trade Frictions • Trade frictions create a difference between foreign and domestic prices but do not result in any Government revenue. • If the home country imports x the domestic price will be pd=p(1+ρ) where p is the world price. • If x is exported then pd=p/(1+ρ). • These frictions serve to reduce trade.

44. Trade liberalisation: key results due to Copeland and Taylor • If trade is distorted by the presence of trade frictions, then trade liberalisation will necessarily increase welfare providing that the tax is set optimally. • C & T demonstrate that the same analysis can also be carried out when the trade distortion is in the form of a tariff. • Once more, we see that if the pollution tax is set optimally trade liberalisation would be necessarily beneficial. • C & T enquire further under what circumstances the change in emissions will be positive or negative as a result of trade liberalisation. But irrespective of whether or not emissions increase or decrease following trade liberalisation provided that the tax on emissions has been set optimally welfare is guaranteed to increase. • And if emissions are controlled via a system of tradable permits then welfare necessarily improves following trade liberalisation. • Trade liberalisation can therefore have quite different consequences depending on whether the country has a comparative advantage in the production or the clean good, and depending on whether environmental policy is optimally adjusted, and if not what policy instruments are used to control pollution.

45. Scale, composition and technique effects: Grossman and Krueger (1993) • G & K give a useful decomposition for thinking about the reasons underlying changes in emissions. • Emissions are by definition equal to the overall scale of activity S, multiplied by the share of dirty goods in total output σ, multiplied by the emissions per unit of the dirty good, e. • Taking logarithms and then totally differentiating this expression gives • This says that the percentage change in emissions is equal to the percentage change in the scale of output plus the percentage change in the share of the dirty commodity plus the percentage change in emissions intensity of the dirty commodity.

46. Reasoning which follows from this • Trade liberalisation boosts market access which will generate economic growth, so likely to lead to an increase in the scale of economic activity. • Other things being equal the scale effect will be environmentally damaging. • Trade liberalisation is also likely to alter the composition of output. • If a country enjoys a relative abundance of the environmental factor of production then that country will increasingly specialise in the production of environmentally intensive ‘dirty’ commodities. • Similarly, if a country possesses a relative abundance of capital then that country will specialise in the production of capital intensive commodities. In fact many goods are both capital intensive and environmentally intensive. • The overall effect of the composition effect on the environment is therefore ambiguous. • Insofar as trade liberalisation increases per capita income levels the public will call for greater environmental quality. • Assuming that the Government is responsive to such demands it will tighten environmental regulations. • Put another way, less of the environmental resource will be allocated to production thereby compelling producers to adopt different production techniques. • This is the technique effect of trade liberalisation. • The detrimental impact of trade liberalisation on the scale effect, the ambiguous impact of trade liberalisation on the composition of output effect, and the likely beneficial impact of trade liberalisation on the technique effect means that the overall impact of trade liberalisation is an empirical question.

47. Box 10.1 The effect of trade liberalisation on emissions: An empirical test • Cole and Elliott (2003) empirically analyse (using panel data regression techniques ) the effect of an increase in trade intensity (the ratio of imports plus exports to GDP) on the per capita emissions of four different pollutants: SO2, NOx, CO2 and Biological Oxygen Demand (BOD). • An increase in trade intensity is taken as synonymous with trade liberalisation. • With regards to the situation under autarky, Cole and Elliott find that increases in the capital-labour ratio increase per capita emissions of SO2, NOx and CO2 although for SO2 each additional increase in the capital labour ratio has a diminishing impact. For BOD, they find no statistically significant relationship between emissions and the capital-labour ratio. The combined impact of the scale and technique effects appears to be negative for SO2, positive for NOx and CO2, and curvilinear for BOD. • The effects of an increase in trade intensity depend on the relative value of the capital-labour ratio. A high income country with a high level of environmental regulations will find that an increase in trade intensity increases SO2 emissions. Nevertheless, a high income country will experience falls in NOx and BOD emissions in response to an increase in trade intensity. • Cole and Elliott’s findings for NOx and BOD are quite different to those for SO2 and CO2. For BOD, the trade intensity coefficient is negative and significant, whilst for NOx it is positive but insignificant. • With regards to SO2 emissions, an increase in income of 1 percent will generate a reduction in per capita emissions of 1.7 percent whereas the trade intensity elasticity is 0.3 percent. The results for BOD suggest that increased trade intensity will reduce per capita emissions; for NOx and CO2 it is likely to increase emissions. • In an attempt to isolate the technique effect, Cole and Elliott also estimate the impact of changes in income and trade intensity on emissions per unit of output. For all four pollutants, growth in per capita income appears to reduce the emissions intensity of output. The trade intensity elasticity of emissions per unit of output is either negative or statistically insignificant.

48. Do governments have an incentive to manipulate environmental standards for trade purposes? • Do governments distort their environmental policies for trade purposes and if so what can or should be done about it? • Environmentalists often express the concern that Governments have an incentive to engage in ‘ecological dumping’ i.e. lowering environmental standards in order to gain competitive advantage. • This claim can be examined in the context of a model of duopoly. • Analysis suggests that because international trade agreements severely restrict the use of more preferred policy instruments, policy makers might strategically manipulate environmental standards. • But, critically, policy makers sometimes have incentives to adopt stronger rather than weaker environmental standards. • In the context of a two-stage game Barrett (1994) examines the incentives for Governments to behave strategically whilst setting environmental standards. • Production is assumed to generate environmental emissions which cause localised environmental damage even though such emissions may be abated; there are no transboundary pollution flows; and for convenience consumers are located in a third country. • In the first stage of the game Governments choose permissible emissions. In the second stage of the game the firms choose their output. • Depending on the form of competition that ensues, Governments have an incentive to set an environmental policy that is either too weak or too strong. • Whilst Governments have an incentive strategically to manipulate environmental standards, such a measure is always inferior to conventional export subsidies or taxes. • Governments would always prefer using such measures if they were permitted.

49. Other results • Brander and Spencer (1985), employing a similar setup, find it optimal for Governments to subsidise their own firms. • However, Barratt demonstrates that although the Government is able to provide an implicit subsidy by increasing the limit on emissions this has a cost in the form domestic pollution problems. In this sense it differs from a pure subsidy. Consequently, a Government would prefer to use a pure subsidy for the purpose of inducing a higher level of output if only such a measure were available to it. •  The foreign Government would also respond strategically. This gives the Nash response and ironically both firms earn lower profits and experience higher levels of pollution. However neither country has the incentive to alter its behaviour. Barratt’s paper further shows that both countries have an interest in cooperating and imposing and stronger environmental standards than that corresponding to the environmentally optimal emissions standard. • Although there is policy competition between the two Governments this does not result in a race to the bottom since there is a cost to Governments engaged in an attempt to gain an increased share of the world market in the form of increased levels of domestic pollution. • Finally Barratt considers the case of Bertrand competition where firms compete over prices rather than output. The analytical model suggests that Government should strategically reduce permissible emissions. • Governments need to know the form of oligopolistic interaction being used in the market in which they want strategically to intervene.

50. Other results (2) • How important are the assumptions of no domestic consumers and no transboundary pollution flows to this particular model? • Ulph (2000) explains that relaxing these assumptions actually reinforces the conclusion that Governments have incentives to engage in ecological dumping. • Since imperfect competition implies less output than socially desirable, Governments would positively desire that their firms expand output. Similarly, where there are transboundary pollution problems Governments have an incentive to reduce foreign output and hence foreign pollution flows. • Do these arguments carry over to other kinds of markets? There is no strategic element in competitive markets since the world price cannot be altered. In the monopoly case there is no rival firm whose output can be manipulated. • There are however circumstances in which Governments have incentives to manipulate the terms of trade in their favour but, where prevented from using export taxes and subsidies for this purpose, manipulate environmental standards as a second best. • For example, a country that is a major importer of a polluting commodity will increase permissible emissions to encourage domestic production if unable to use an import tax to depress world prices. Likewise a country which is a major exporter of a dirty commodity will reduce permissible emissions if it cannot use an export tax to raise world prices.