Beckerman, TRADEABLE PERMITS for Carbon Emisions

INTERNATIONAL JUSTICE AND ENVIRONMENTAL POLICY: TRADEABLE PERMITS FOR

CARBON EMISSIONS

by

Joanna Pasek and

Wilfred Beckerman

GEC Working Paper 94-24

INTERNATIONAL JUSTICE AND ENVIRONMENTAL POLICY: TRADEABLE PERMITS FOR

CARBON EMMISSIONS

by

Joanna Pasek and

Wilfred Beckerman Acknowledgements The Centre for Social and Economic Research on the Global Environment (CSERGE) is a designated research centre of the U.K. Economic and Social Research Council (ESRC) Joanna Pasek is an Associate Professor of Philosophy at the University of Warsaw and a Research Associate at the Centre for Social and Economic Research on the Global Environment (CSERGE), University College London and the University of East Anglia. Wilfred Beckerman is an Emeritus Fellow of Balliol College, Oxford. The authors are grateful to Professor David Pearce for constructive comments on an earlier draft. ISSN 0967-8875

Abstract There has been much discussion of alternative international measures to restrict global carbon emissions in order to moderate global warming. There is widespread agreement that the most economically efficient mechanism would be some system of tradeable emission permits. However, although it is generally believed that acceptance of the 'equitable' character of any initial allocation of permits is essential to gain adequate international adherence to any such scheme, there is scope for vast differences of opinion as to what would constitute an 'equitable' allocation. Several proposals have been put forward as being justified in the interests of environmental justice or other ethical grounds. In the first part of this paper the more popular of these are discussed, and it is argued that their ethical foundations are, in fact, very flimsy. This is followed by an examination of the concept of "environmental justice", which is explicit or implicit in much of the literature on equity aspects of environmental policy, and it is concluded that such a concept, as distinct from justice in general, is not a useful one. The main central section of the paper goes on to discuss, more generally, what principles of international justice can be identified that might have practical application to problems such as those of sharing internationally the burdens of environmental protection. It is shown that there are major obstacles to the formulation of any coherent theory of international justice, including the question of whether such a theory should apply between nations as distinct from individuals. In particular, it is shown that such conclusions concerning international justice that can be extracted from existing theories provide little or no guidance to the equity of any particular international allocation of carbon emission permits. A final section of the paper puts forward some concrete proposals for the initial allocation of carbon emissions, and argues that these represent a reasonable compromise between alternative equitable criteria.

1

1. INTRODUCTION In June 1992 in Rio de Janeiro two major international environmental agreements dealing with the "global commons" were negotiated: the Framework Convention on Climate Change (FCC) and the Biological Convention on Diversity (CBD). In 1987 the Montreal Protocol set the framework for control of ozone layer depleting substances. Currently (1994) a Convention on "Desertification" is being negotiated in respect of land degradation that contributes to loss of biodiversity and to climate change. All these agreements involve issues of "inter-generational equity" - fairness between generations - since many of the costs of failing to agree will fall on future generations. They also raise issues of "international justice" - burden sharing between nations now - based on issues such as "responsibility" for global pollution and species loss. Although the prevention of global warming is not the only environmental issue that gives rise to the need for some form of internationally co-ordinated action this paper concentrates on this particular problem. This is partly because it is a widely discussed and relatively familiar issue and partly because, as a result of this, some of the contributions to the discussion have included suggestions concerning the ethical principles that ought to be taken into account in any international agreement to combat global warming. This applies particularly to what appears to be one of the most widely entertained instruments for international action, namely the system of "tradeable permits", which will be referred to below as TPs. The basic idea of this system is that some agreement is reached as to the total global amount of greenhouse gas emissions that is to be tolerated, and permits to emit this total amount will be allocated amongst countries according to some formula that will be agreed in the course of the negotiations. The focus of this paper is on the main greenhouse gas, namely carbon dioxide. Countries will be allowed to trade their allocated permits, and each country will presumably buy or sell permits in the light of its own costs of reducing carbon emissions. Some countries may not need to use all their permits and will prefer to sell some of them if they find it cheaper to reduce their emissions, up to a point, even further. Other countries will find it cheaper to buy permits than to reduce emissions to the amount covered by their initial allocation of permits. This trading will lead to the emergence of a world price for emission permits. Clearly, a central issue in any such negotiation will be the initial distribution of permits amongst countries. Any internationally negotiated method of sharing out amongst countries the burden of reducing CO2 emissions will, of course, be dominated by the self-interest of participating nations. In particular, there is a conflict of interest in this matter between the "South" and the "North" (terms used as rough classifications of poor and rich countries respectively) (Pearce et al., (1992)). Nevertheless, unlikely though it may appear, it is also usually maintained - sometimes with great conviction - that considerations of inter-national equity will also play some part, even if only as providing "focal variables", to use Barrett's term, around which the negotiations can be centred (Barrett (1992); Kverndokk (1992); Solomon and Ahuja (1991); Rose (1992)).

2

Indeed, explicit recognition of the role of "equity" considerations is contained in the Framework Convention on Climate Change agreed at the 1992 Rio de Janeiro World Environmental Conference (INC 1992). It is stated that "...the parties should protect the climate for the benefit of present and future generations of humankind, on the basis of equity (our italics) and in accordance with their common but differentiated responsibilities and respective capabilities" (INC (1991)).1 However, the existing discussions of the various "equitable" rules that could be drawn up for regulating international action to combat global warming invariably contain three major flaws:- (i) they tend to assume, without discussion, that there is a valid concept of

"environmental justice", as distinct from justice in general, or even of international justice in particular;

(ii)they also assume that there is a clear and valid concept of international justice that

takes the form only of justice as between nations, rather than justice between people who happen to be citizens of different nations. This assumption may not be as difficult to justify as the preceding assumption, but it has been the subject of explicit criticism in the philosophical literature concerning international justice and its validity cannot be taken for granted without discussion; and

(iii)some, if not all of the "equitable" rules that have been proposed seem to bear no

clear relationship to any of the main theories of justice or to any other major compelling moral imperatives that are widely adopted in civilized societies. And in the few instances where some specific theoretical justification is claimed for some particular rule the claim is usually based on a total misunderstanding of the particular theory in question.

In the next section of this paper we shall discuss some of the most popular rules proposed for allocating TPs on an equitable basis and will show that, in fact, there is little or no particular ethical force in any of them. We shall then examine two fundamental general weaknesses in the existing literature specified above. First, we shall argue that there is little basis for a theory of environmental justice in particular, as distinct from justice in general. Secondly, we shall argue that the main relevant features of some of the more important theories of justice, or related ethical theories, provide very little support for a theory of international justice in terms of relationships between sovereign states as distinct from their inhabitants. We then discuss what can be said, if anything, about the main features that any "fair" system for allocating TPs should exhibit. This will lead on to our own very tentative proposed equitable criteria. We regard these as a reasonable compromise between, on

1 INC (1991), Article 3.1

3

the one hand, the need to take account of so-called "equity" rules in the interests of providing generally acceptable "focal points" in negotiation and, on the other hand, what would be implied by more defensible equitable criteria. Our proposal may be summarised as consisting of dividing the total of tradeable permits into two portions, one of which should be allocated between nations in a manner that take account of the interests of the nations according to a formula the ethical justification of which is set out. The rest of the permits should be sold (or, preferably, leased) by some international agency, such as the World Bank. This agency would be charged with devoting the revenues to assisting governments to carry out projects that remedy the worst environmental problems (e.g. lack of clean drinking water or adequate sanitation) experienced by the poorest individuals of the world's population in developing countries. We believe that such a proposal kills three birds with one stone, namely:- (i) reducing the threat of global warming, the benefits of which will, on the whole, be

shared by all classes of the population irrespective of their initial wealth; (ii) satisfying some of the moral claims of poorer nations for special preference in the

conduct of international environmental policy, and (iii) providing resources for helping the poorest sections of the world's population in the

environmental field. This is one more bird than is claimed by Barrett for the use of the TP system (Barrett 1992, page 85).

2. A CRITIQUE OF SUCH "EQUITY" CRITERIA AS HAVE BEEN PROPOSED A perfectly adequate selection of the more plausible and popular criteria that have been proposed, on "equity" grounds, for allocating TPs between countries is given by Barrett (1992).2 Whilst some equitable rules may have been proposed that are not included in this selection it certainly encompasses all the more popular candidates. We shall not attempt here, however, to study most - let alone all - of the proposed rules. We restrict ourselves to commenting on a few that seem to be the more frequently considered (or already embodied in other international agreements) for purposes of illustrating the main thrust of our argument. One very commonly suggested rule is to allocate the total number of permits amongst countries on a per capita basis. In fact it was explicitly proposed in an early draft of the (International) Framework Convention a revised version of which was agreed at the 1992 Rio "Earth Summit". According to this convention emissions were to converge to a common per capita level over the course of the years (INC 1991). One of the experts on the international negotiation of conventions to reduce carbon emissions, Grubb, says of this rule that "The moral principle is simple, namely that every human being has an equal right to use the atmospheric resource" (1989). Others commentators, too, have

2 Barrett's list includes all those listed by M.Grubb and J.K.Sebenius (1992), or A.Rose, (1992), and goes well beyond the formula considered by A. Grübler and N. Naki_enovi_, (1992).

4

insisted on the ethical force of this rule (Grubb and Sebenius (1992); Rose (1992); Kverndokk (1992); Solomon and Ahuja (1991)). It seems to us, however, that this principle is based on a confusion between a moral right - if there is one - to an equal use of some resource, and a moral right - if there is one - to the equal welfare derived from its use. The reason for this confusion is that, in most cases, the two things are identical. When somebody uses, say, a given amount of the resource food he or she is actually "consuming" an identical amount of food. But one's use of the environment by emitting some CO2 bears no direct relationship at all to the benefits that one obtains from doing so. There is not even any direct relationship between how much more clean air the inhabitants of a country would breath and how far that country reduces its emissions of CO2. How much more clean air anybody breathes is determined not by how much he or she uses up but by how much everybody else is using up.3 It is true that the more food, say, other people eat the less food will be available for any given person. Nevertheless, it will still be true that the amount the given person eats is identically equal to the amount he is using up. Food is a private good; clean air is a public good. This common distinction in economics happens to have ethical implications. Thus although there may be a strong case, in terms of various ethical systems, for giving everybody an equal minimum right to satisfy some basic needs, such as minimum housing, food, and even clean air to breathe, and so on, it cannot be argued that basic needs include a basic need to pollute the environment! The ethical case, therefore, for allocating rights to pollute the environment on a per capita basis, is not as clear cut as might appear at first sight. The reason why the equal per capita rule seems to have some superficial appeal to our moral intuitions is presumably as follows. Consider a siege or famine situation in which there is a limited amount of food available. In order to ensure that everybody receives their basic food need, in accordance with the ethical injunction that everybody has an equal right to the basic wherewithal for survival, it would be necessary to lay down a maximum amount that anybody could eat. This would be achieved by rationing food. Some people might not want to use all their ration and could sell some of it, but, at the same time, everybody would be guaranteed a basic minimum. If, instead, the food had been rationed by the price mechanism, some people would have been unable to afford any, and would have starved. Rationing is required in order to provide everybody with their minimum right to some food. Does the same argument apply to emission limitations? At first sight it might appear that it does. It is true that emitting carbon is not the same as breathing clean air.

3 M.Grubb and J.K.Sebenius (1992) do add, parenthetically, that '(Note that equal emissions need not be the same as "equal benefits from emissions")', although this position does appear somewhat inconsistent with M.Grubb's earlier emphatic support, on ethical grounds, for this particular rule.

5

Nevertheless, if we take "breathing clean air" as code for "being spared the harmful consequences of global climate change", it is true that, in order to provide everybody with equal access to clean air, we must limit the total amount of carbon emissions. Why not put a maximum on everybody's emissions in the same way as is done in the case of the food shortage - i.e. by allocating emission permits on a per capita basis? The answer is that all that is necessary to provide everybody with equal minimum amounts of clean air is to limit the total global carbon emissions. Unlike the famine situation, it is not necessary to put a maximum on how much carbon individuals emit in order to protect everybody else's right to a minimum amount of clean air. This is protected by limiting the total level of carbon emissions. How this is allocated is irrelevant, for the very same reason that the amount of carbon one emits is not identically equal to the amount of clean air one breaths. If total global emissions of carbon are limited to the level required to prevent serious climate change then everybody is protected equally however the total amount is distributed amongst the population. For example, if all the permits were to be handled by, say, the World Bank and auctioned on the world market, some countries might find it cheaper not to buy any and others might prefer to buy very many. It would make no difference to the way individual nations are affected by the resulting degree of climate change. In other words, we return to the point made at the outset, namely that the fundamental flaw in the "per capita" allocation argument is the failure to recognize the "public bad" character of carbon emissions. Nor can it be argued that, notwithstanding the above, everybody should be given an equal amount of clean air simply because it is a scarce resource. Nobody would suggest that everybody should be given an equal amount of the world's supply of copper, for example. Insofar as we are motivated by egalitarian considerations we should aim at greater equality in the final flow of goods and services that are produced with the aid of the earth's resources. And an equal right to use some clean air (by polluting it with CO2, for example) does not necessarily promote this objective, and could even make the existing inequality of incomes worse insofar as some countries with large populations were richer than smaller countries. Another popular rule is that each country should reduce its emissions by a uniform percentage (e.g. Barrett (1992)). Strictly speaking, formulated in this way this rule ought not even be included in a list of rules for allocating TPs. For if the rule were to be applied emission permits would simply not be tradeable. This is because once trade takes place the degree to which countries will reduce their emissions will not be uniform and will vary from country to country. Presumably one is intended to interpret this rule as a rule for allocating TPs amongst countries proportionately to their existing, or recent, levels of carbon emissions, whilst remaining free to trade these permits. Countries would then trade their TPs and each country would finish up reducing emissions by whatever amount is optimal for it. Whether interpreted this way, or as a straightforward rule for uniform reductions in emissions without any scope for trading, this rule does seem to have wide appeal as a starting point in actual international negotiations. It

6

played a part, notably, in the Montreal Protocol for reducing CFCs, and in negotiations to reduce the sulphur dioxide emissions believed responsible for the acid rain problem. But the only principle that has been advanced in support of this rule is that it involves the minimal disruption to the status quo ante, and hence involves smaller transfers from some countries to other countries. There are three objections to this principle:- (i) as long as the permits are tradeable a uniform price will emerge on the market that

will be very similar whatever the initial allocation of permits.4 So there is no reason why this allocation should involve a different degree of disruption to the status quo than any other TP system;

(ii) in any case, leaving this aside, it is not clear what is so desirable from an equity point

of view in the minimal disruption of the existing position of countries; and (iii) minimal disruption of the status quo defined in terms of existing levels of emission

may not be the same as minimal disruption of the status quo defined in some other dimension, such as the loss of welfare imposed on countries. Indeed, an objection that has been raised to this system on equity grounds is that it rewards the countries that currently emit the most CO2. And, in turn, this would mean that most of the permits would be allocated to the developed countries of the world. Nevertheless, we believe that, suitably modified to take account of distributional considerations, this rule could form a defensible part of some equitable formula, for reasons set out in the final section of this paper.

The objection to the preceding rule to the effect that it rewards countries that are currently the heaviest emitters of CO2 has sometimes been embodied in a rule that is designed to have the opposite effect, namely to penalize such countries by allocating permits in inverse proportion to their per capita consumption of fossil fuels (Barrett (1992); Solomon and Ahuja (1991)). The ethical principle behind this rule would appear to be some principle of "retributive justice". Without wishing to go into the vast area of retributive justice, it should be noted that insofar as this would be designed to create a disincentive for people to behave in an anti-social manner - i.e. was not motivated by pure revenge - there would be little point in applying it here, since as long as emission permits are tradeable they provide the incentive to restrict one's emissions to the socially optimal level, irrespective of one's initial level of emissions. There seems even less case, in terms of any theory of retributive justice, for the rules that have been proposed to take account of how far countries are morally responsible for the amount of carbon emitted in the past in the country of which they happen to be

4 It need not be the exactly same under all allocations since different allocations of permits give rise to different income distributions and hence different demands and supplies for the permits and hence different ruling prices.

7

citizens (e.g. Solomon and Ahuja (1991); Kirk Smith (1991)). Apart from the practical difficulties of estimation, such "historical responsibility" rules and their related concept of "natural debt" ought to be rejected on ethical grounds. The notion of moral responsibility is closely linked to the notion of being a free agent, voluntarily carrying out any act, in knowledge of its consequences. The element of choice is an essential ingredient of moral responsibility. But it is difficult to see how the inhabitants of any country can have had any choice as regards the amount of fossil fuels burnt up by earlier inhabitants, even if they are not immigrants. It is difficult therefore to see why the existing citizens of any country should have to bear a heavier burden on account of their predecessors' consumption of fossil fuels. The counterpart to this problem is whether there is any ethical basis for the claim made by citizens of poorer countries today for compensation for practices carried out in the past by what are now the richer countries.5 One rule that is given some attention in the literature is what has been described as a "Kantian" allocation rule.6 This rule is that all countries are asked to say what uniform reduction in emissions they would prefer. Countries will give different answers, of course, depending on how they perceive the benefits they will obtain from general uniform abatement. There will be no agreed uniform reduction, but each country should then reduce its emissions by the amount of uniform reduction it had proposed. As explained in Section 4.6 below, however, this is a mis-representation of Kant. It would be like a rule emerging from a process in the course of which poor people opted for a rule under which all rich people were to transfer money to poor people and poor people were entitled to steal from richer people, and the rich people would opt for a rule under which rich people had no moral obligation to transfer anything to poor people (though they might do so out of charity). If, then, no agreement is reached the rich people would be entitled to transfer very little and the poor people would be entitled to steal from them. As explained in Section 4 below, this is not at all what Kant had in mind! One popular candidate for an equitable rule (e.g. Barrett (1992), rule 2) happens to coincide almost completely with the rule included in our own overall proposal. This is the rule that TPs should be allocated on the basis of carbon emissions per unit of GNP. This rule has been proposed by Grubb (1989) on grounds of economic efficiency, although, as we have already pointed out, as long as the permits are tradeable the outcome will be economically efficient whatever the initial allocation of permits. We

5 The difficulties surrounding such a claim are well set out in G.Sher (1981).

6 e.g. Barrett, 1992. Barrett pays considerable attention to this rule on account of the fact that - given the assumptions of his model some of which are highly debatable - it also appears to be the rule that maximises the total net benefits to all participating countries. Barrett adopted the term "Kantian" in this instance on account of its use in a somewhat similar case by J-J Laffont,(1975) who described a similar rule as Kantian. The error in Laffont's interpretation of the Kantian rule is explained in Section 4 below.

8

believe that this rule should not, therefore, be excluded on efficiency grounds and should be taken seriously on equity grounds. Our reasons for this are set out in the final section of this paper. 3. IS THERE A SEPARATE "ENVIRONMENTAL JUSTICE"? As indicated above, although the relevant literature contains references to the key role of equity considerations in international negotiations to combat the global warming threat they fail to recognize that they all make implicit assumptions that are philosophically highly contentious. One of these is the assumption that there is some separate "environmental justice", as distinct from justice in general. This assumption is implicit in the relevant literature for the following reason. The tradeable emission permit solution - on which we focus in this paper - is the one that has the greatest support in environmental circles. Permits to pollute are to be tradeable, in the interests of economic efficiency. The redistributive character of the scheme resides in the way the TPs are initially distributed amongst countries. If, however, these permits were regarded as just straight "purchasing power" - which is what they really are - there would be no question of how far their distribution is just or not in relation to environmental criteria. One would be concerned solely with the justice or otherwise of some pattern for allocating aid internationally. This is an interesting and important question, but it exists whether or not there was believed to be any problem of global warming. Indeed, a large literature on the theory and practice of international aid existed long before international environmental issues came to the fore. But, as shown above in the brief survey of the main "equity" criteria proposed, the literature on the equity aspects of allocating TPs closely relates the equity of alternative initial distributions of TPs to specific environmental considerations - such as how much CO2 different countries emitted now, or how much they had emitted in the past, or how much it would cost them to cut emissions, or how much they would lose by not doing so, and so on. By contrast, in discussions, for example, of how much social welfare payments ought to be made to the poor, the argument revolves around the amount of total purchasing power, in the form of money, that ought to be taken from some members of the community in order to be transferred to the poor. It is not in terms of what would be a "just" distribution of each component of their total consumption, such as food, clothing, shelter, entertainment, travel, etc. There is no analysis of whether a "just" distribution of clothing, for example, should be related to whether the wearers need to be smart at their place of work, or whether they work in warmer or colder environments, and so on. Thus, the fact that the literature on the equity of alternative allocations of TPs is in terms of specifically environmental criteria implies that "justice" in this field is believed to be related to specific environmental criteria - i.e. that there is a special valid concept of environmental justice. But no arguments in favour of this assumption are given, or even any recognition that any such arguments are necessary.

9

Is there any particular philosophical tradition that might be invoked in defence of the view that there is some special environmental justice? It would, of course, be far beyond the scope of this paper to attempt even a brief summary of the various theories of justice that might be relevant here. The most that can be done here, therefore, is to select a few illustrative examples - e.g. the widely accepted claims for an equal distribution of certain burdens or privileges, such as the obligation in time of war to do military service (subject to certain other conditions being satisfied), or the right to vote or to retain some control over one's labour other than that which one had voluntary relinquished as part of some contract. In these cases, it is believed that the distribution of certain goods or services should be equal and should lie outside the scope for redistribution through the market mechanism. But, as will be shown below, it is difficult, if not impossible, to see any connection between any of the types of theory mentioned and the need for any particular criteria for a "just" distribution of a tradeable permit to emit CO2. This does not mean that principles of distributive justice can have no bearing on the distribution of environmental assets. But it does mean that the existence of a valid special theory of environmental justice cannot simply be taken for granted without any discussion. We may approach the problem by asking whether there is any philosophical tradition according to which the criteria for a "just" distribution of some entity - whether it be overall income, or welfare, or some specific asset (including access to certain occupations) - would also imply a specifically just distribution of the particular asset, "the environment". During the course of the last two thousand years or so numerous criteria of distributive justice have been proposed, such as need, worth, and work. Relatively recently Vlastos has suggested that a new principle of distributive justice that has as good, if not so well-established, a claim, as the others would be "to each according to the agreement he has made"7 Do some of these alternative criteria imply that they might be more relevant for some goods than for others? In some cases the answer appears to be "Yes". For example, Bernard Williams's classic discussion of "The Idea of Equality" (1969) explains how the criterion of need, say, would apply to certain goods and services, notably medical care for sick people. Walzer, too, attaches considerable importance to the concept of "need" in his general defence of the notion that different criteria of just distributions apply to different goods (1983). Walzer's position is that the goods with which a theory of distributive justice is concerned are "social goods", and that "All distributions are just or unjust relative to the social meanings of the goods at stake".8 The idea that certain goods should be distributed in accordance with certain criteria such as need or merit has been

7 G.Vlastos (1984), page 44.

8 Walzer, M., 1983, page 9.

10

particularly persistent throughout time and in widely different cultures. For example, it has been very widely held that certain influential or highly paid offices should be held only by candidates holding relevant merits. Judges, according to this view, should be selected according to their knowledge of the law, their intelligence, impartiality and integrity - rather than irrelevant considerations, such as being rich or beautiful or having been educated at Oxford. It been suggested that Aristotle's well-known merit criterion is relevant in the distribution of other specific types of goods, such as access to certain forms of education (the merit criterion being one's intelligence or aptitude for learning, and so on). In other words, justice would include merit as a "relevant" reason for treating people unequally with respect to the distribution of certain goods rather than for treating them unequally with respect to their overall level of wealth or income. However, Williams points out - echoing Aristotle - there would be great difficulty in reaching agreement as to what constituted a "merit" in this context. At the same time many people, including many philosophers, would subscribe to the view that for certain "goods" - defined to include certain social positions that would carry with them greater wealth or power - merit is a perfectly valid criterion for a just distribution. So if we accept the view that a theory of justice that is politically relevant and potentially operational should respect and embody firmly held views as to what is just, one would certainly have to accord some status to the merit criterion as regards the distribution of certain goods. 9 Thus for one reason or another - some philosophical support or considerations of public acceptability - the notion that special criteria should determine a "just" distribution of certain particular assets cannot be ruled out. A similar picture emerges in connection with the criterion of "need", which would also be relevant in connection with certain kinds of goods in many approaches to justice, including Walzer's. The argument is that the first thing that members of a political community owe to each other is the communal provision of security and minimum welfare. This fundamental welfare commitment implies that some basic needs have to be satisfied by any members of the community. Indeed, the concept of "relative deprivation" introduced into the sociological analysis of poverty measurement in the 1960s (though set out by Adam Smith almost two centuries earlier) concentrated precisely on the question of how far lack of access by the very poor to certain goods and services led to their suffering a sense of being excluded from the community.10

9 There are survey data that confirm the widespread acceptance of the merit criterion as a basis for a just distribution of wealth in general. For example, D.Miller's extensive review of such data contains references to US data showing that, amongst the samples of people surveyed, there was virtually unanimous rejection of the proposition that everyone ought to receive the same income (only 5 percent in favour) and, furthermore, that 78% of the respondents believed that people with higher ability should earn more. D. Miller, (1992) pp.564/5.

10 See, in particular, Townsend, P. (1962) and Runciman, W.G. (1966).

11

Of course, since resources are scarce political decisions will constantly be made as to how far society should go in meeting "needs" and what constitutes a worthy need. Many societies, for example, think that deaf people should be supplied with free hearing aids if they are below certain minimum income levels. But it is not usually thought that they are also entitled to subsidized front row seats at the Opera. Walzer describes the way that some minimal provision of health care has been increasingly accepted over the ages, as "...a function of the community's sense of danger and the extent of its medical knowledge", and he gives the example of the ancient Athenians who provided public baths and gymnasiums but nothing remotely resembling social security.11 But even in connection with medical care he emphasizes the role of the different cultural concepts of what is needed, and what price the community is prepared to pay, which, in turn, depends partly on the strength of each community's social cohesion and bond. Thus there is a very respectable tradition in philosophy to the effect that special criteria determine what is a just distribution of certain goods (defined very widely). How far can it apply to environmental goods or services? The most likely relevant criterion would appear to be the need criterion. This would imply that people should have equal access to certain basic environmental goods, such as access to clean drinking water, sanitation, shelter and clean air, that are necessary to their attaining some socially minimum acceptable standard of living. But in the first place, this is not really an application of any sophisticated theory of distributive justice. It is much more the application of a basic humanitarian ethic that calls for peoples' basic needs to be met. It just happens that some environmental goods are recognized as being part of basic needs. Secondly, there is no connection between equal access to environmental goods of the kind enumerated above and equal allocations of TPs, which, as pointed out above, are simply forms of purchasing power. Countries receiving these TPs are free to convert them into command over whatever goods they like. They may keep them if they prefer to convert them into a right to use the environment. But they may equally choose to reduce their CO2 emissions enough to be able to sell some or all of their allocation of TPs in order to use the money to buy other goods. There may be good reasons in terms of distributive justice for giving some countries more money than others, but they do not seem to have much to do with countries' relative emissions of CO2. Another possible approach, therefore, could be to work out the implications of combining two propositions. One, the validity of which will be discussed in the next section, is that the existing international distribution of total wealth is inequitable in the light of some theory of distributive justice. The second is that the environment (or some components of it) should be treated as a "newly discovered scarce resource". It was always there, of course, but it has only recently been "discovered" to be scarce, because it has only recently become scarce. That is to say, now that the scale of

11 M.Walzer, 1983, page 67.

12

anthropogenic pollution of the atmosphere has reached a point where climate change may have serious effects on world welfare, "clean air" has become a scarce resource. If these two propositions are accepted the way is opened for maintaining that the world now faces new opportunities to rectify any injustices in the international distribution of income that may be believed to exist. For example, although, as indicated above, redistributive policies have tended to operate in terms of the total monetary transfers made to needier sections of the community, it has often been recognized that political considerations sometimes make it easier to carry out redistribution if it is couched in terms of particular goods. For example, whilst many people may object to giving more money to poorer families, fewer of them seem to object to giving them free school meals or cheap milk for their children. Similarly, whilst the public may not easily accept higher taxes to pay for higher old-age pensions, there may be more support for providing subsidized train fares or TV licences to the "poor old dears". This "specific egalitarianism" may not be economically "efficient" (some poor children may be allergic to milk, some pensioners may never travel or watch TV, and so on), but is often politically more acceptable. As a matter of practical politics, therefore, the international community might want to take advantage of the opportunity provided by the discovery of a "new resource" - namely that the atmosphere is now a scarce resource the exploitation of which affects everybody in the world - to move towards a more equitable international distribution of income. In other words, one possibility is to ignore all of the proposed criteria in terms of climate or of current or historic emissions, and just concentrate on the redistributive aspects of any scheme - i.e. how far it can help move the international distribution of income in the direction believed to be ethically justifiable. Unfortunately, of the two propositions advanced above, the first - namely that the existing international distribution of wealth is inequitable - is probably the most contentious. In fact, it is far from clear that there is any existing coherent theory of international distributive justice at all. Whether there is, and what form it might take, is the question to which we address ourselves in the next section. 4. THEORIES OF JUSTICE AND INTERNATIONAL REDISTRIBUTION Historically, the idea that redistributive justice might apply between countries as well as between people within a country is relatively new. International justice was previously confined largely to the problem of peace and war between nations. But, it has been argued, the question of distributive justice between nations could not have arisen before institutional and technological progress had reached a certain stage. Consequently, as O'Neill puts it "traditional codes said little about economic justice to those who lived beyond the frontiers" (1991, page 276) Of course, amongst the important ethical systems from which one may choose, some may have implications for justice between nations. These implications may be negative

13

or positive. That is to say, they may imply that no valid concept of international justice is feasible. On the other hand they may imply - and according to some writers on the role of equity in the international allocation of TPs they do imply - that they indicate what would be "just" international allocations of the burdens of environmental protection. The reason why there is scope for such a wide difference between alternative theories of justice in their relevance to the problem in hand is that principles of international justice are not simple extensions of principles of intranational justice. For example, those - like "libertarians" - who argue against redistribution within a country will also presumably be against redistribution between countries. But those who believe that there is a moral case for redistribution within countries are not necessarily in favour of global redistribution. Whether there are any grounds for extending moral duties across national boundaries is a subject of intense controversy between those for and against it. In the following section the main theories of justice will be briefly examined from the point of view of their implications, if any, for the principles of international justice. Here, we shall begin by a very brief mention of some of the better-known systems that would appear to rule out any such redistribution on equity grounds. We shall then consider in slightly more detail two closely related concepts of justice that have been explicitly claimed, in the literature on equitable rules for allocating TPs, as being relevant. These are the Rawlsian principles of justice, and the Kantian approach to moral obligations. 4.1 Libertarianism The essence of libertarianism is that there is no compelling ethical basis for redistribution of wealth at a national level - i.e. as between individuals within a country. The main argument is that such redistribution would violate the individual's rights to liberty , notably the right to keep any goods that one has acquired legitimately. That right is fundamental and must not be violated even if its violation would lead to more equality. Nozick, for example, calls these liberty rights "side constraints" upon action. He derives them from one of the Kantian moral imperatives, which says that a person should never be treated solely as a means, but always at the same time as an end. Since individuals are to be treated as ends then, Nozick argues, they are inviolable. They may not be used or sacrificed without their consent for the achievement of some other ends (Nozick (1974), page 31). Any redistribution that is not purely voluntary implies using some individuals, namely those who are expected to contribute to the welfare of others, instrumentally, as means only. Of course, there is the old problem of whether the preservation of the liberty rights of the individual should be an over-riding absolute objective that "trumps" other possible objectives. These may include other rights, such as a right to some minimum level of welfare, or other values that may justify, or even demand, the violation of liberty rights on the national or international level. But this issue is of no concern to us here. All that

14

is relevant here is to indicate that such a libertarian ethical system is unlikely to provide support for the view that redistribution between nations is morally obligatory. 4.2 Communitarianism Communitarians, too, would not normally be sympathetic to the notion that there is some valid concept of international distributive justice. They believe that the idea of distributive justice presupposes a bounded world of a political community. It is a world of a common language, culture and history, which produce a collective consciousness that is essential for investing the concept of justice with common meaning (Walzer (1983), page 28). To make sense, international justice would require the whole world to be a single political community. Since it is not, communitarians argue that international justice is an illusion. This in turn leads to the conclusion that compatriots should have priority. As it is often pointed out however, in the modern world, the boundaries between ideologies or modes of thinking are not impervious and mutually impenetrable. Modern ideologies are not self-sufficient and mutually isolated. Furthermore, the degree of international interdependence has greatly increased on account of global issues that have forced countries to co-operate and behave as a "community" in a variety of fields. The relatively novel problems of the global environment, notably the climate change problem with which this paper is concerned, are striking instances of the new scope for the development of co-operative action which would itself define the nations of the world as being members of a community. These increasing relationships between nations and cultures may partly undermine a strongly communitarian belief in the incompatibility of different national traditions of understanding justice. 4.3 Utilitarianism As is well known Utilitarianism attaches no end-value to distributional considerations per se. But this does not rule out distributional considerations as means to the maximization of utility. Thus, a Utilitarian ethic could well have implications for the redistribution of wealth between countries, since such a redistribution might raise total world utility. And since the utilitarian principle of maximizing utility has a universal appeal it might well provide grounds for some international redistribution of wealth. As the utilitarian philosopher Smart puts it " the only reason for performing an action A rather than an alternative action B is that A will make mankind (or, perhaps, all sentient beings) happier than will doing B" (Smart and Williams (1973), page 30). If it is the well-being of the whole human race that we need to have in mind then we seem to have support for international redistribution. Sacrifices for the sake of the needy that would increase total utility are morally obligatory in the utilitarian system. (It is interesting to note that since any action that increases general utility is obligatory, there is no room for charity in strict utilitarianism). In economic terms the case for redistribution from rich to poor in order to maximize total world utility is based on the "law of the diminishing marginal utility" - i.e. that, as people

15

become richer they obtain less utility from further marginal increments in consumption. Economic redistribution should be carried out up to a point where the marginal utilities of those who contribute and those who receive economic assistance become equal. On an international scale this would, of course, require enormous sacrifices by the wealthier countries which would not be forthcoming by any stretch of the imagination. And support for such a redistribution on utilitarian grounds would require that no account be taken of the other objectives that would have to be sacrificed, such as the radical restriction of individual rights (liberty rights). One Utilitarian philosopher, Singer, defends a less extreme view in favour of an obligation to international assistance which he hopes will appeal to non-utilitarians as well. His view resembles Popper's "negative utilitarianism", according to which we should minimize suffering rather than maximize happiness. Singer believes that we should act upon the following principle: if it is in our power to prevent something very bad from happening, we ought to prevent it unless that meant sacrificing something of equal or greater moral significance (Singer (1993), page 229) And since "absolute" poverty is a bad thing, we have a moral obligation to prevent it when it involves no sacrifice of anything of comparable moral significance.12 However, as Singer himself admits "the precise amount of absolute poverty that can be prevented before anything of moral significance is sacrificed will vary according to the ethical view one accepts" (1993, page 231). Although the utilitarian position provides very strong grounds for international redistribution, it does not seem to fulfil the expectations it raises concerning the amount and the pattern of redistribution. As O'Neill puts it "consequentialism raises hopes with the prospect of replacing conflicts with calculation, but dashes them by providing overly pliant instruments for action"(1991, page 232). How pliant they are can be shown by the variety of actions that may be based on utilitarian recommendations. For example, in the context of international redistributive policy, these could include:- (i) transferring resources to the poor until further transfer would reduce total well-being,

account being taken of the law of diminishing utility; or (ii) abstaining from redistribution on the grounds that more transfers to the poor would

facilitate a rise in their populations, so that the addition to total misery might more than offset any addition to world utility; or

(iii) redistributing selectively to those who are likely to become self-sufficient, and not to those who are likely to remain dependent on aid.

12 "Absolute poverty" is a term that is intended to indicate some minimum standard of living, independent of the consumption levels of other people in the society in which one happens to live. It would correspond roughly to the provision of some needs that are essential for survival. It leaves open the length of the period over which one is expected to survive.

16

If each of these (and other) different actions can be defended on utilitarian grounds, utilitarianism does not really provide much guidance except perhaps for local and limited problems. And even then we need some method of comparing different options, including a capacity to predict the likely result of each option, and an adequate theory of value to rank them these options. In the light of these difficulties, O'Neill goes as far as to say that, as regards international justice, consequentialism is a non-starter (1991, page 283) For we are not sure which decisions would maximize utility. 4.4 Rawls's theory of justice Contractualist theories of justice have a long and distinguished pedigree, including, for example, Hobbes, Locke and Rousseau. But it is its latest formulation, by Rawls, that has been explicitly cited as providing some guidance in the international allocation of the burden of global environmental protection (Rawls, (1972)). But in fact Rawls provides far less guidance to the problem than claimed. For he does not apply his difference principle to inequalities between nations. As is well known, Rawls arrives at principles of justice as being those that would be chosen by people in some "original position" behind a "veil of ignorance", in which they would decide on just institutions and rules for a society without knowing in advance what particular position they would come to occupy in that society, or their particular possession of talents or even conception of the good life. In this position, they would agree on two basic principles of justice. It is the second of the two principles, called the "difference principle", that interests us here. According to this principle social and economic inequalities are allowed only when they are to the advantage of the worst-off members of the society. This principle is egalitarian in the sense that "The social order can be justified to everyone, and in particular to those who are least favoured" (1972, page 103). Rawls believes that initial differences in natural talents and abilities are "...neither just nor unjust; nor is it unjust that men are born into society at some particular position. These are simply natural facts. What is just and unjust is the way that institutions deal with these facts" (1972, page 102). And that is done by the adoption of the difference principle according to which "no one gains or loses from his arbitrary place in the distribution of natural assets or his initial position in society without giving or receiving compensating advantages in return" (1972,page 102). More specifically, the difference principle says that "Social and economic inequalities are to be arranged so that they are both (a) to

the greatest benefit of the least advantaged and (b) attached to offices and positions open to all under conditions of fair equality of opportunity" (Rawls (1972), page 83)

How far do these principles apply internationally? On the assumption that the two principles of justice have been adopted within societies, Rawls proposes to "extend the interpretation of the original position and think of the parties as representatives of different nations who must choose together the fundamental principles to adjudicate

17

conflicting claims among states" (1972, page 378). Just as persons in the original position choose behind the veil of ignorance, in the international original position representatives of states are deprived of various kinds of information. They do not know anything about the particular circum-stances of their own society, about its strength in comparison with other societies or about their own position in their society. "Once again the contracting parties, in this case representatives of states, are allowed only enough knowledge to make a rational choice to protect their interests but not so much that the more fortunate among them can take advantage of their special situation .....Justice between states is determined by the principles that would be chosen in the original position so interpreted. These principles are political principles, for they govern public policies toward other nations" (1972, page 378). Then Rawls gives examples of principles to be chosen, which, he admits are "familiar ones". The basic principle is a principle of equality (analogous to the equal rights of citizens) which according to Rawls implies such principles as self-determination of nations , their right to self-defence, the principle that treaties are to be kept, and principles defining a just cause in war. These are the principles that he believes will be selected in the original position to determine justice between states. There is no mention of distributive principles of international justice. Their absence is rather conspicuous. In fact, Rawls actually states that "The original position is fair between nations; it "nullifies the contingencies and biases of historical fate" (1972, page 378) The political principles which he says would be adopted in the original position do not nullify the contingent distribution of natural resources between countries, which is largely responsible for the enormous differences in economic welfare between different countries. Indeed, Rawls's exclusion of the difference principle from the principles of international justice is difficult to reconcile with his general position on distributive justice. One would have thought that representatives of different states in the original position acting behind the veil of ignorance as to where they will eventually manifest themselves, as it were, would make an attempt to establish some principles of international distributive justice similar to the difference principle. For presumably the initial distribution of natural resources - i.e. of natural environmental resources - should be regarded as a contingent fact which needs to be dealt with in the original position. If Rawls believes that contingencies such as being born with particular talents should work for the good of the least fortunate, he should make a similar claim about the international distribution of resources. In fact it can be argued that the natural distribution of resources is a purer case of something being "arbitrary from a moral point of view" than the distribution of talents. As Beitz puts it "Not only can one not be said to deserve the resources under his feet; the other grounds on which one might assert an initial claim to talents are absent in the case of resources, as well" (Luper-Foy (1988), page 33) 4.5 Extended Rawlsian principles The reason why Rawls does not apply his difference principle to the principles of justice between nations appears to be that he treats nation-states as "more or less" self-

18

sufficient. And since his entire scheme of establishing the principles of distributive justice applies to society as a "cooperative venture for mutual advantage" it cannot properly apply outside the boundaries of that cooperative venture. However, as Beitz rightly argues, the world is not made up of self-sufficient states, and international interdependence is an undeniable fact. Therefore, if the need for, and scope for, social co-operation provides a basis of domestic distributive justice, then international co-operation should provide a basis for global distributive justice. Thus, although it is clear that Rawls did not apply his difference principle to distributive justice between nations, there is no reason why one should not use his procedural device of the original position as a framework for investigating how, in fact, the participants in such a position would have wanted to nullify the contingencies of differences in the international distribution of natural assets. Beitz, for example, concludes along very Rawlsian lines that although each person has an equal claim to the world's natural resources, sharing them out unequally is justified if it is the best way of promoting the interests of the least advantaged (Luper-Foy (1988)). Another philosopher who has followed this line is Barry, who has been particularly critical of Rawls's failure to apply the difference principle between nations (Barry (1973)). He suggests that we could think of it by imagining that we were an embryo with a random chance of being born in any country in the world. Would we prefer to be born in the real world, with a high chance of being born in a country with widespread malnutrition and high infant mortality, or in a world in which the gap between the rich and the poor nations has been reduced? If rationality makes us opt for the second world then we ought to include some rules for international redistribution in the principles agreed in the original position. Indeed, according to Barry, participants in the "international" original position should take account of the fact that the situation of the worst-off people in the world will be improved best not through the application of the difference principle within countries but through its global application. "Whether we replace maximin with equality, maximizing the average level of well-being, or some "pluralistic" cocktail of principles, we still get the answer that from the standpoint of the "original position" the question of distribution between societies dwarfs into relative insignificance any question of distribution within societies. There is no conceivable internal redistribution of income that would make a noticeable improvement to the nutrition of the worst-fed in India or resourceless African states like Dahomey, Niger or Upper Volta" (Barry (1973), pp 130-131)). But a major problem with the Rawlsian framework is that it is not clear whether principles of international justice, such as those that might be drawn up in the original position, should apply to nations or individuals. This question poses a serious problem for any theory of international justice. Rawls himself, in the extracts quoted above, is clearly limiting himself to the principles governing just arrangements between sovereign states. Brian Barry's approach could equally well lead to the participants in the original position agreeing on some version of the difference principle that applied between sovereign states or between individuals across national boundaries.

19

But Beitz believes that the global difference principle applies in the first instance to persons in the sense that it is the situation of the worst-off people around the world that should be maximized (Luper-Foy (1988)), However, since it is sovereign nation states that are primary actors in international politics, the international difference principle should be applied to them (Luper-Foy (1988)). Perhaps in the ideal world in which countries were internally just Beitz proposal might indeed be a "second best". In the real world, however, with poor countries not being necessarily internally just, international transfers might not be used for the improvement of the lot of the worst-off people. Hence, individuals should be the legitimate subjects of international justice. As Hoffmann has put it " One has to keep in mind that states exist only as communities of people; states are not divinities , their rights are rooted in the presumption of a fit between them and their people; and this does put a kind of damper on the demands of the Third World governments for absolute sovereignty, for impermeable state rights. We may feel that we have a duty to share some of our wealth with them, but only if that wealth is used toward justice for those communities of people. This also means that equity claims presented by Third World states are in a sense conditional on their doing something for their people" (Hoffmann (1981), page 156). And later he adds that although "it may sound like blackmail, but in blackmail cases one of the actors is doing something illegitimate or selfish" (1981, page 181). If there is some uncertainty as to whether the principles of distributive justice should operate between nations or amongst individuals across national boundaries, why not "ask" the participants in the original position? In other words, how would agents in the original position prefer to specify the principles of international distributive justice? Would they prefer that the difference principle should apply separately, first, within countries, and then, secondly, as between whole countries? Or would they prefer some application of the principle among individuals without respect to national frontiers? If, following Rawls, the difference principle were not to apply between nations, its application within nations might be of little re-assurance to highly risk averse participants. They might fear that, as Brian Barry has indicated, no amount of redistribution within some poor countries would do much to raise the living standards of the poorer citizens. If, instead, the difference principle were to be applied between nations, the poor countries' living standards might be raised so much that safety net of redistribution among individuals, either within nations, or supernationally, might hardly be necessary. And if the difference principle were to be applied across individuals irrespective of the country in which they lived the distribution of goods between countries would - as Kverndokk has pointed out - be automatically determined (1992). Much would depend, therefore, on how badly off the worst off would be under alternative scenarios. What principles would be chosen by participants in the original position, there-fore, would then depend on what "knowledge" they are allowed to have about the how far international inequalities reduced or increased the position of the worst off countries, as well as the corresponding relationship within any country. All this might be an interesting subject for speculation, but would lie well outside the scope of this paper.

20

In any case, how feasible is any modified international difference principle along Rawlsian lines? Hoffmann calls it a "tall order" and a "maximalist view", that is too demanding to be applicable in the real world of politics (1981) Apart from the theoretical problems of judging how the participants in the original position would view alternative applications of the difference principle internationally, there are serious practical difficulties such as:- (a) it would require giving precedence to all the poor people abroad rather than using

resources for domestic self-improvement. It would ignore the moral importance of national borders. And it might often require a transfer of resources from a country which is already reasonably "just", in the Rawlsian sense, to a poorer country, in a manner that might make the donor country unjust not to mention the matter of how just is the recipient country (Hoffmann (1981), page 155)

(b) as O'Neill points out, given the enormous complexity of international interrelations, it

would be very difficult to evaluate which institutional changes would most improve the welfare of the worst-off in the world (O'Neill 1991). Just as in Utilitarianism we encountered the difficulty of evaluating which decisions would maximize utility, so in the Rawlsian view we would find it difficult to asses which reductions in international inequalities would maximize the welfare of the worst-off. It is not clear what "knowledge" there is about the effect of inequalities on the position of the worst off that is to be concealed from the participants in the original position.

Thus, at the end of it all, what bearing does Rawlsian theory have on the problem in hand here, namely the equitable distribution of TPs? The answer appears to be "not much, if at all". First, Rawls himself does not apply the difference principle between nations. Secondly, he does not even discuss whether it should apply to individuals globally across national boundaries, and, indeed, he seems to assume that it should not do so, on the grounds that nation states are self-sufficient entities, whose political integrity and independence have to be respected. Any international redistribution would appear to conflict with the political principles that he set out governing the just relationships between nations. The notion adopted by some environmentalists to the effect that Rawlsian principles provide one possible framework for allocating TPs amongst countries on an equitable basis thus seems to be very far-fetched. Nor does it seem to help much if one follows Beitz or Barry in order to apply the basic insights of the Rawlsian approach to the formulation of some sort of international difference principle. The theoretical and practical obstacles seem to be too great. Nevertheless, as Barry has pointed out in his discussion of Rawls's views on a rather similar problem, namely justice between generations, the difficulties encountered in a strict interpretation of Rawls's theory should not blind us to the crucial role in his theory of the notion of fairness and its relationship to the acceptance of rules as being "just" if they are rules that would have been accepted by everybody in ignorance of their own

21

particular position (Barry (1989)). This, of course, is inherent in the Kantian core of Rawlsian theory13. And since Kant, like Rawls, but even more implausibly, has been cited in the environmental literature as a guide to an equitable allocation of tradeable emission permits, it is to Kant that we must now turn. 4.6 The Kantian approach In the economic literature on global warming and on the proposals concerning how to deal with it in an "equitable way" one often comes across references to Kantian moral philosophy. Mention has been made above of a so-called Kantian rule for allocating TPs amongst countries. The so-called "Kantian" rule in the present context is one that "...requires each country to choose an abatement level that is at least as large as the uniform abatement level it would like all countries to undertake" (Barrett (1992), page 108). As explained in Section 2 above, under this rule, each country is asked to propose a uniform abatement level to which, if generally adopted, it would also conform. It would be understood that each country would choose a uniform abatement level in the light of its own costs of reducing emissions. Thus, China, say, may prefer that everybody be obliged to reduce emissions by 0.2 percent because that is the uniform reduction that would suit it best. For the same motive, Russia might want everybody to reduce carbon emissions by 2.0 percent, and the USA by, say, 4 percent. If then no agreement is then reached on a uniform reduction, each country agrees to reduce its emissions by the amount of the uniform reduction that it had itself proposed at the outset. Thus, China will reduce its emissions by only 0.2 percent, and so on. This has been described a "Kantian" rule because of its apparent affinity with the well-known Kantian "categorical imperative". But, as pointed out in Section 2 above, this rule is a mis-representation of Kant. For, even allowing for the scope for varying interpretations of Kantian rules, the essence of his categorical imperative is that, in arriving at a view as to how everybody should behave, we should all abstract ourselves completely from our own particular situation and see the operation of the selected rule from the point of anybody else that may be affected. That is to say, in considering what rule should govern assistance to needy people a rich person should take account of the situation of needy people even if he had every reason to assume that he would never be needy so that the rule is not in his interests at all. Similarly, a poor person would take account of the consequences of any such rule for rich people even if he had no hope of ever becoming rich himself. Thus, in the first place, for an emission reduction rule to qualify as an application of the Kantian categorical imperative all countries would propose the same uniform reduction,

13 Rawls states explicitly that "The original position may be viewed, then, as a procedural interpretation of Kant's conception of autonomy and the categorical imperative". Rawls, J. 1971, page 256.

22

for they would all be taking the same account of every other country's interests as well as of their own. Every country would opt for a uniform reduction not on the basis of how much uniform reduction would be in its own interests, but on the basis of which such reduction would also take due account of every other country's legitimate interests. The rule for carbon emissions proposed by Barrett, by contrast, would lead to each country proposing a different uniform reduction in emissions in the light of its own interests. In other words, the mere fact that each country proposes a uniform reduction in emissions has nothing to do with Kantian impartiality. Countries are not expected to be impartial in their choice of the uniform reduction in emissions that they would like every country to adopt. As pointed out in Section 2, Barrett's "Kantian" rule may be compared to a situation where rich people propose a uniform rule according to which they have no moral obligation to help the poor and poor people propose that wealth be shared out much more equally (a situation that, in fact, bears a striking resemblance to reality!). The famous Kantian categorical imperative which is meant to help us decide which actions are morally right, makes impartiality a central requirement. In deciding what is morally right I ought to abstract from what lies in my own interest. The categorical imperative (in the first of its three formulations) says "Act only on that maxim through which you can at the same time will (i.e. would like to) that it should become a universal law" (Kant (1964), page 88) In one of Kant's examples he considers whether it is morally right to borrow money if we know we will not be able to pay it back. If we did judge that we were morally entitled to borrow money in such a situation we would be following a maxim such as "It is morally right to borrow money if you are short of it although you know that you will never be able to pay it back". For Kant this rule or "maxim" has to be rejected as it cannot be wished that it should become a universal law, although it may work to my advantage and be "quite compatible with my own entire future welfare" (Kant 1964). The reason why it cannot be wished to become a universal law is that it would make "the very purpose of promising impossible, since no one would believe he was being promised anything, but would laugh at utterances of this kind as empty shams" (Kant (1964), page 90). Indeed, Kant could hardly be more emphatic in insisting that what is morally right cannot be determined by what lies in one's own interests. At some point he explicitly says that the categorical imperatives (in the three formulations of which we have presented one) "did by the mere fact that they were represented as categorical, exclude from their sovereign authority every admixture of interest as a motive" (Kant (1964), page 99). Thus, no negotiated agreement that is influenced in any way by the interests of individual countries can be said to be based on the Kantian categorical imperative. Although the different allocation rules cannot be treated as applications of the categorical imperative, they are applications of what Kant calls the "hypothetical imperative". He writes that "If the action would be good solely as a means to something else, the imperative is hypothetical" ((1964), page 82). So, if a country, acting in its own interests, proposes an abatement level that best promotes that end, then, in Kantian terms, the country is acting on the hypothetical imperative. The hypothetical imperative,

23

however, does not have anything to do with morality. We act morally only when we act on the categorical imperative, which is categorical in the sense that it says "I ought to do this" in contrast to the hypothetical imperative which simply says "I ought to do this IF I happen to want that". The former tells us what is our duty to do, the latter simply advises how to best choose means to ends, without inquiring into the validity of the ends themselves. How much guidance does the Kantian categorical imperative provide? In the Groundwork, Kant was confident that his categorical imperative provided moral guidance even for people who had little experience in assessing the morality of various decisions. "Inexperienced in the course of the world affairs and incapable of being prepared for all the chances that happen in it, I ask myself only 'Can you also will (i.e. wish) that your maxim should become a universal law?' Where you cannot, it is to be rejected, and that not because of a prospective loss to you or even to others, but because it cannot fit as a principle into a possible enactment of universal law" (1964, page 71). Unfortunately, things are not quite so easy that we only need to ask ourselves the right question in order to receive the right answer. The categorical imperative tells us to act only upon principles (maxims) that are universalizable. It does not tell us what these principles are. Obviously different situations will require that different moral principles will be considered as candidates for a "universal law". For example, we could ask - as does Thomas Nagel does in his essay "Kant's Test" - " how are we supposed to discover, about a particular maxim, whether we can or cannot will it to become a universal law?" (Nagel (1991), page 41). In our particular case, how are we to discover which allocation rule would be closest to what we could will to become a universal law? Clearly, this question presents us with an enormous difficulty. First of all, what does it really mean to say that some rule would qualify to become a universal law? Apparently, it would be a rule which everybody who managed to abstract from his individual needs and interests would be prepared to endorse. But how can we find a formula that would take into account all conceivable (and often conflicting) points of view and hence become the Kantian universal law? At this point one may doubt whether the Kantian categorical imperative lead us anywhere. However, although the categorical imperative does not allow us to identify a single allocation formula that could be said to embody it, and leaves us instead with a very wide range of formulae as plausible candidates, it does seem to eliminate some extreme formulae as inconsistent with it. In this sense the function fulfilled by the categorical imperative is mainly negative. We do not learn which solution would be morally right but do seem to learn which ones would be morally wrong. For example, Nagel asks how much assistance to the poor would the Kantian ethic expect us to provide? On the one hand there is a minimum level of assistance that a well-off person should expect to give to the poor and below which he should not fall "because he cannot will such tight-fistedness as a universal law" (Nagel (1991), page 49) On the other hand there is some maximum level of sacrifice above which a rich man's assistance to the poor cannot be expected to be morally obligatory. But in

24

between these extremes there is a region "where we cannot will as a universal principle either that one must or that one need not help the needy" 14. It is in that region that we are left without guidance and various solutions that fall in that category are neither morally forbidden nor morally compelling. The guidance provided in this interpretation of Kant for present purposes would be very much in line with the Rawlsian injunction to be concerned with the position of the worst off in society. As Rawls says of his own principles of justice "The principles of justice are also categorical imperatives in Kant's sense" (1972, page 253). For it is clear that a Kantian approach would imply that there is a moral obligation to ensure that priority was given to assisting the worst-off sections of the world's population. Kant would say that no one who applied the categorical imperative would wish to adopt a universal law that denied help to those who "struggle with great hardships" (Kant (1964), pp.90-91). Thus, it appears that Kant, like Rawls, provides an ethical framework that can justify concern for the worst off members of society, but does not go further than that. One might argue that this conclusion is more or less identical with simple humanitarianism, the only difference - and perhaps this is an extremely important difference - being that it is grounded in some ethical system that has a fully articulated basis. But leaving aside the extent to which this conclusion has some sophisticated defensible basis or is just a basic intuitive ethical position, we do take account of it in our own proposed allocation rules that are set out in the next section. 4.7 The ethical contribution: a summary It should be of no surprise that the outcome of the above discussion is that none of the highly abstract philosophical discussions of theories of justice appears to have much bearing on the practical problem of the international allocation of the burden of global environmental protection, including the two specific ethical systems that have been claimed, in the environmental literature, as having such a bearing. Nevertheless, it is hoped that the preceding discussion does justify our extracting two conclusions that can have a practical bearing on the problem under discussion here, namely the equitable international allocation of tradeable emission permits. First, some respect has to be given to national sovereignty, along the lines, for example, of Rawls's first principle of justice, and taking into account the importance of formulating principles of justice as principles that are to be operational within some form of co-operative community. Secondly, since this clearly will not suffice to remedy the arbitrary inequalities amongst individuals that are partly the result of arbitrary differences in natural environmental resources between countries, there is room also for some rules of justice, appealing to Kantian maxims of impartiality or Rawls's similar emphasis on

14 Nagel, T. (1991), page 50. O'Neill has something similar in mind when she says that Kantian deliberation "is a decision procedure for detecting forbidden and obligatory action and does not aim to rank all possible actions" (1986), page 136.

25

justice as fairness. In the next section we propose an approach to the practical problem of the allocation of tradeable emission permits which is designed to embody some combination of these two conclusions 5. An Almost Equitable Allocation of Tradeable Carbon Emission Permits: When a TP (tradeable permit) system is introduced and the permits are allocated according to some formula or other, two things are done:- (i) something that had hitherto been a free good - namely the right to emit CO2 - now has to be paid for; and (ii) a new form of money is issued; the permits are equivalent to money. For example, if the permits are allocated to countries, no country is obliged to use all its permits or to use only the amount issued. It can sell some of them or buy some more. TPs are like tradeable ration coupons. Consider ration coupons for soap. Some people will prefer not to wash much and to sell their coupons in order to buy other goods (perfume, perhaps?). Others will prefer to give up some other goods in order to buy more soap. Giving people the coupons is equivalent to giving them money which they can spend on soap or other goods as they prefer. Giving countries TPs is equivalent to giving them money to buy the right to pollute or to sell some of these rights to buy other things. The problem is "by what principles or criteria should the money be allocated?" This looks like another way of putting the old question of "by what criteria should one distribute money to different countries?". Although this is not a new problem, there has been relatively little discussion in the aid literature of the ethical basis for aid. And, as we have shown in the previous section, such limited discussion as there has been in the philosophical literature of the principles of international distributive justice suggest that there is no firm basis - and perhaps not even a fragile basis - for a theory of distributive justice between countries, although there may well be a stronger case for a theory of distributive justice between people in whatever country they may happen to reside. Nevertheless, we have here an extra dimension arising out of the change in the status quo brought about by converting a hitherto free good into something that has to be paid for. Thus, even if it is believed that there is no substance to any critique of the existing international distribution of wealth, there might be a stronger case for believing that the costs of adapting to the status quo should be shared out in accordance with some principles of equity. One starting point for an equitable principle for dealing with this change in the status quo would be to compensate countries for it in proportion to the costs they will incur as a result of their agreeing to the solution (the TPs) that is in the interests of the world as a whole. A very rough approximation to their costs would be the extent to which they already emit CO2. Big emitters will have to either use more permits or incur more costs to cut their emissions. One way or another they will bear greater costs in reducing emissions by any given amount.

26

However, as pointed out in Section 2 above, it will be objected that this should not be a reason to allocate permits to countries in proportion to their emissions, since it would mean allocating them mainly to developed countries. This is regarded as objectionable, either because it is rewarding sin, or because the big emitters may be richer (and usually are). In the latter case, and assuming a sort of international diminishing marginal utility of income, the proportionate loss of welfare they will suffer from any given cut in emissions or payment for permits to emit will be less than for a poor country. Hence, an allowance could be made for this in the initial allocation of permits by taking account of the countries' income levels. The argument here is parallel to the views put forward - and almost universally accepted - in connection with special measures adopted in Britain to compensate old age pensioners for extra fuel bills in the course of a very cold winter spell a few years ago or as a result of the extension of value added tax to fuel bills in the November 1993 budget. In both cases rich peoples' expenditure on fuel would have risen most simply because they consume more fuel. But it was old age pensioners who were thought to need compensation, since it was thought that they were generally much poorer and hence would suffer disproportionately from any extra financial burden imposed on them. The application of this principle to the allocation of CO2 emission permits would then be that permits should be allocated according to some formula that also takes account of countries' ability to pay. This is also the principle that has long been accepted as equitable in distributing the burden of taxes amongst people. At the same time it has been suggested that account should also be taken of the extent to which the new regime (having to pay for a hitherto free good) will affect countries, notably that it will impose a bigger burden on countries that emit the largest amounts of CO2. A formula that takes account of both these criteria, therefore, is one that gives most permits to countries with the highest ratio of current emissions (measured in any physical units) to GNP (measured in money at some agreed exchange rates). It may appear strange to divide some physical unit, such as million tons of CO2 emitted, by a monetary unit, such as GNP, but the former is just a rough proxy for the costs that each country will incur to either buy permits (perhaps "from itself" by keeping its own permits, or from other countries) or to cut pollution. The scale used is irrelevant. For example, suppose there are only two countries, X and Y, and X produces 60 units (whatever they are) of CO2 per unit of GNP (measured in any numeraire), and Y produces 40 units per unit of GNP. And suppose it is decided that the total number of units of CO2 emissions that is permitted in the world is 1000. Then country X would get 60 percent of them and country Y would get 40 percent of them. More complicated formulae could easily be devised in order to allow for diminishing marginal utility of income in a more sophisticated manner without changing the principle involved or complicating the problem of the unit of measurement.15

15 Dividing emissions by GNP in order to correct for differences in countries' marginal utilities of income in accordance with differences in their income levels involves the implicit

27

As indicated in Section 2 above, the proposed rule happens to be rule 2 on the Barrett list. It does not, as is sometimes alleged, discriminate against poor countries. For, cet. par., a higher GNP means a lower ratio of emissions to GNP and hence a lower initial allocation of TPs. Thus, other things being equal, the above rule tends to impose the smallest burden on poor countries, who have the lowest denominators in the ratio and hence, cet. par. will get the most permits. It is true that if they happen to be relatively low emitters they will also get fewer permits. But there would only be a bias against poorer countries if, on the whole, the poorer countries had disproportionately smaller levels of carbon emissions. However, this does not appear to be the case overall, though there is no strong correlation one way or the other. For example, if this rule were applied on the basis of 1987 levels of GNP and carbon emissions, China would be entitled to six times as many permits as the USA and India to twice as many.16 However, we would not propose that the whole of the total global amount of TPs agree upon be allocated according to the above formula. For, as argued in Section 4, it is difficult to justify a theory of distributive justice between nations as distinct from between individuals. And as a matter of practical politics, it will be difficult enough to obtain the participation of rich countries, such as the USA, that will almost certainly have to be net purchasers of TPs, in a scheme that may transfer funds to countries who may use them to bolster their military systems or build monuments to their rulers or deposits in Swiss bank accounts. Thus the total of TPs should be split into two groups - not necessarily equal. In the interests of reaching agreement and out of respect for national sovereignty (a quasi-Rawlsian principle of international justice) one group of TPs should be allocated according to the formula specified above. The other TPs should be allocated to some international agency, such as the World Bank, that would use the revenues from the sale of the TPs for projects that directly alleviate the most acute environmental needs of poor people in developing countries. Such projects might be entirely financed by the agency in question or carried out by the agency in conjunction with the authorities of the countries concerned (to ensure that the funds are not simply substitutes for expenditures that the governments concerned would otherwise have carried out anyway). Of course, there is still enormous scope for negotiation as to what proportion of the total permits should be allocated by countries under such a formula as distinct from sold by some international agency that would be charged with using the revenues for direct

assumption of a logarithmic utility function, so that the ratio of the marginal utilities obtained by countries at two different income levels is inversely proportional to their income levels. A more egalitarian assumption could be adopted making them inversely proportional to, say, the square of their income levels.

16 This can be seen from the last column of Table 2, page 96, in Barrett, S. 1992.

28

assistance to improve the environment of the world's poorest people. There would also be great scope for negotiation as to more detailed application of the allocation amongst countries of those TPs that are to be handled in this way. For example, allowance might be made for the extent to which some countries have already incurred heavy costs in order to reduce their CO2 emissions (e.g. France). Negotiators would also take account of the impact on their own countries of the overall limit on emissions and the precise allocation formula - e.g. over what time period "current emissions" should be measured; how to take account of future changes in emissions and GNP; what precise mathematical formula should be used for linking emissions/ GNP ratios to the permit allocations; and so on. Do these proposals bear any direct relationship to any major ethical principles of a just distribution? Not much, apparently. Utilitarianism might appear, at first sight, to provide some foundation for the above proposal, insofar as account is taken of the law of diminishing marginal utility and the consequent attempt to minimize the burden of emission reduction on poor countries. But the principle that richer people should transfer wealth to poorer people up to the point where their marginal utilities are equal is hardly likely to be accepted even as a starting point, or "focal point", in any international negotiations. Anyway, as already indicated in Section 4, Utilitarianism does not have any room for distributive justice as an end in itself. Rawlsian principles, which aim partly to fill this gap in Utilitarianism, seem to provide little guidance. Rawls himself did not provide any guidance as to distributive justice between countries as distinct from individuals. Secondly, even if we applied the Rawlsian basic original position device as a means of identifying what would be a "fair" system of justice between individuals wherever they happen to live (i.e. across national boundaries), we encounter apparently insuperable theoretical and practical obstacles. The most we can say is that Rawlsianism emphasizes that due concern should be paid to the position of the worst-off people in society on grounds of "fairness". The same applies to Kantian moral precepts, and in a similar way. 6. Conclusions In the literature on measures to reduce the threat of global warming it is generally accepted that much the most efficient system that could be adopted internationally would be a system of tradeable permits to emit CO2. But this raises the question of how these permits are to be distributed amongst countries to begin with. In the literature much lip service has been paid to the important role that has to be played by considerations of equity in drawing up rules for allocating the permits between countries. Some of these rules are also claimed to be supported by certain ethical systems, such as egalitarianism, Rawlsianism, or even Kantianism. We have argued that no theoretical justification has been provided for any of the more popular "equitable" rules that have been considered, and that the token gestures made in the direction of certain famous theories of distributive justice are based on misunderstandings of what these theories do actually claim. Furthermore, there is no

29

firm basis for the implicit assumptions in the literature to the effect that (i) there is some special theory of environmental justice, or (ii) that there is some special theory of international justice. None of this should be surprising since few, if any, philosophers claim that any ethical theory - as a theory of distributive justice must be - would have much practical application to a novel and detailed issue such as access to the environment. In the end, the most that theoretical considerations can do is to sharpen our awareness of the precise meaning of, and limitations on, various principles that may be considered, and hence to alert us to exaggerated claims for any particular principle. And picking among the debris of proposed allocation rules one can perhaps find one or two pointers towards an "equitable" system that can be justified in terms of rather basic notions of "fair" play in sharing out the burden of some change in the status quo (the environment has become a "scarce" good), whilst paying due respect to the role of sovereignty in international co-operative behaviour. In the light of these considerations we would propose the following system:- (i) A country's emissions of CO2 would be limited by the number of permits it holds.

Each country will receive an initial allocation of permits according to the procedure suggested below, and will then be free to buy or sell permits as it thinks fit;

(ii) An international negotiation would settle the total amount of such permits and the

duration of their validity. The permits would be divided into two groups. One group of permits would be allocated to some international agency, and the other permits would be allocated amongst countries;

(iii)The agency would be charged with using the revenues from the leasing of the

permits under its control to assist in the provision of better basic environmental facilities for the most deprived sections of the world's population. This would be justified on the grounds that the case for international justice, if there is one, would be in terms of individuals, not of nations, and that ethical principles or simple humanitarianism do require that consideration be given to the most basic needs of the worst off sections of the world community;

(iv)Those permits to be allocated to countries would be allocated between them

according to the ratio of their "current" emissions to GNP (which happens to be one of the rules often considered in the literature, and usually rejected for reasons that do not seem to us to be valid). No "grand" existing ethical system can be quoted in support of this part of our proposal, and it is vain to hope that one could find one. It is based on fairly rough justice, namely that it partially compensates countries for the burden of their adjustment to the new situation by taking account of both the size of this burden and their capacity to pay it. It also takes account of the political necessity and, some would say, the theoretical

30

desirability, of paying due respect to the role that national sovereignty should play in any such international co-operative venture.

31

References: Barrett, S. (1992), 'Acceptable' Allocations of Tradeable Carbon Emission Entitlements

in a Global Warming Treaty. In UNCTAD Combating Global Warming, UNO,New York.

Barry, B. (1973), The Liberal Theory of Justice, Chapter 12, Clarendon Press, Oxford. Barry, B. (1989), Theories of Justice, Vol 1 of A Treatise on Social Justice, Harvester-

Wheatsheaf, London. Beitz, C. (1988), International Distributive Justice, in S.Luper-Foy (ed.) Problems of

International Justice, Westview Press, USA. Grubb, M. (1989), The Greenhouse Effect: Negotiating Targets, Royal Institute of

International Affairs, London. Grubb, M. and Sebenius, J.K. (1992), Participation, Allocation and Adaptability in

International Tradeable Emission Permit Systems for Greenhouse Gas Control. In OECD Climate Change: Designing a Tradeable Permit System, Paris.

Grübler, A. and Naki_enovi_, N. (1992), International Burden Sharing in Greenhouse

Gas Reduction, World Bank Working Paper No.55. Hoffmann, S. (1981), Duties Beyond Borders. On Limits and Possibilities of Ethical

International Politics, Syracuse University Press. INC (Intergovernmental Negotiating Committee) (1992), Annex I to the Report of the

Intergovernmental Negotiating Committee for a Framework Convention on Climate Change: United Nations Framework Convention on Climate Change, document A\AC.237/18 (Part I)/Add.1, United Nations, New York.

Kant, I. (1964), Groundwork of the Metaphysics of Morals ,translation by H.J.Paton,

Harper and Row, N.Y. Kverndokk, S. (1992), Tradeable CO2 emission permits: Initial Distribution as a Justice

Problem, Centre for Social and Economic Research on the Global Environment, Working Paper 92-35, University College London and University of East Anglia.

Laffont, J-J. (1975), Macroeconomic constraints, economic efficiency and ethics: an

Introduction to Kantian economics, Economica, Vol.42. Miller, D. (1992), Distributive Justice: What People Think: A Survey Article, Ethics, Vol.

102, April 1992. Nagel, T. (1991), Equality and Impartiality, Oxford University Press.

32

Nozick, R. (1974), Anarchy, State and Utopia, Blackwell, Oxford. O'Neill, O. (1991), Transnational Justice, in Political Theory Today, ed.by Held, D.,

Polity Press, Oxford O'Neill, O. (1986), Faces of Hunger. An Essay on Poverty, Justice, and

Development, Allen and Unwin, London. Pearce, D. (1990), Greenhouse Gas Emissions: Part 1, Internationally tradeable

greenhouse gas permits. Mimeo, University College London and UK Department of the Environment.

Pearce, D., Fankhauser, S., Adger, N., and Swanson, T. (1992), World Economy,

World Environment. The World Economy, Vol.15, No. 3. Rawls, J. (1972), A Theory of Justice, Clarendon Press, Oxford. Rose, A. (1992), Equity Considerations of Tradeable Carbon Emission Entitlements,in

UNCTAD Combating Global Warming, UNO,New York. Runciman, W.G. (1966), Relative Deprivation and Social Justice, Routledge and Kegan

Paul, London. Sher, G. (1981), Ancient Wrongs and Modern Rights. Philosophy and Public Affairs,

Vol.10, No 1. Singer, P. (1993), Practical Ethics, second edition, Cambridge University Press. Smart, J.J.C. and Williams, B.(1973), Utilitarianism. For and against, Cambridge

University Press. Solomon, B.D. and Ahuja, D.R. (1991), International reductions of greenhouse-gas

emissions. In Global Environmental Change, Vol.1, No 5 Townsend, P. (1962), The Meaning of Poverty. In The British Journal of Sociology,

Vol.13. Vlastos, G. (1984), Justice and Equality, in Waldron, J.(ed.) Theory of Rights, Oxford

University Press. Walzer, M. (1983), Spheres of Justice, Blackwells, Oxford. Williams, B. (1969), The Idea of Equality, in Feinberg, J.(ed.) Moral Concepts, Oxford

University Press.