THE WTO, UNFAIR TRADE AND DEVELOPMENT

Don Ross

1. Philosophical preliminaries

There may not be many points of consensus over what best promotes economic development, but here is one that has formed over the past decade: the institutional context matters a lot. This represents the single greatest shift in economic thinking about development since World War II, for there once was an almost equally clear consensus that institutions don’t matter, since capital was supposed to be able to find its way through and around them, whatever they were like, to the magnet of highest returns. I know of no significant group of economists who believe that anymore. To this extent, development theory is not just the cacophony of rival opinions it might appear to be when the grain of analysis is drawn more finely.

But of course we do require finer grains of analysis when debating policies, including policies for the design of the all-important institutions. In this essay I will focus on one of these institutions, the World Trade Organization, and on the ethical criticisms to which it is regularly subjected. My contention will be that most of that criticism has, so far, been of very limited use to the promotion of development, because it has been insensitive to the dynamics of bargaining, including ethically informed and guided bargaining.

The stalking horse for my argument will be the reflections of a leading moral philosopher, Peter Singer, who devotes about a quarter of his 2002 book on globalization, called One World, to the shortcomings of the wto. I choose Singer for the role of foil for two reasons. First, while his analysis is organized around the main themes of the populist criticisms heard against the wto, Singer channels them through careful reasoning and attentiveness to facts. Thus, for example, he scrutinizes the evidence for a proposition dear to the hearts of self-styled ‘anti-capitalist’ activists, that globalization in general, and the wto regime in particular, have made poor countries materially worse off, and rejects it. This is not to say that he supports the opposite proposition; rather, he argues that the facts we would need to justify either generalization are not to hand. Arguing with ideologically committed people who are interested only in facts that support their preconceptions is tiresome. Singer is worth arguing with because he is not such a person. Second, I share enough of Singer’s background moral assumptions that when I enter into this debate with him, our disagreements are really about the wto, and not about wider issues in moral philosophy that are logically prior.

Let me begin, then, by stating these background assumptions. Like Singer, I am a consequentialist. That is to say, I believe that the moral status of a political, social, legal or economic policy or institutional framework should be assessed by reference to its impact on actual human welfare, not against its conformity to abstract, deontological principles. I do not assume that material welfare is all that matters. However, I do assume that if an effect of implementing a suggested moral principle were reduction of, or stagnation in improvement of, average human material welfare, given the present actual and soon-to-be-actual global population level,[1] this would constitute overwhelming evidence against the acceptability of that principle. I also join Singer in assuming that the welfare of all people in the world should be weighted equally in choosing principles for designing policies and institutions. Specifically, I assume that nationalism, of a kind that allows people to prefer policies that benefit their co-nationals to the detriment of, or in indifference to, the welfare of foreigners, is prima facie immoral. Putting this in terms that will be familiar to philosophers, I reject ‘communitarianism’ as a fundamental principle. I of course accept that membership in communities is a very important part of most people’s well-being, and that for that reason we ought not to prefer policies or institutions that gratuitously disrupt communities. But I assume that communities have no moral standing in and of themselves, over and against the welfare of their actual members. Thus, if most members of some ethnic, national or regional community would be better off, by their own lights, were that community to disappear, than the extinction of the community in question would not be a morally relevant loss (all else being equal).[2]

In this essay, I will use the concepts of welfare’ and ‘well-being’ interchangeably. I do not share the view of Sen (1999 and elsewhere), let alone the more extreme version of the view expressed by Hamilton in this volume, that people can be made better off against their own preferences. This is mainly because I do not believe that the tyrannical effects of paternalism can realistically be controlled if it is allowed a conceptual foot through the door, given all our inductive evidence about human moral and factual reasoning. I therefore prefer to ban paternalism outright by requiring that actual, subjective preferences of all people, as expressed in their actual behavior, be fully respected in moral reasoning and bargaining. I would add, against Hamilton’s contention that ‘needs’ should be weighted heavily against mere ‘wants’, that I cannot imagine how any broadly democratic institutional mechanisms that tried to capture this distinction could avoid incentivizing people to promote their merest ‘wants’ into ‘needs’. I thus believe, as a factual matter, that insofar as a policy regime sought to implement Sen’s or Hamilton’s ethical framework, it would further encourage rent-seeking behavior; and that is about the last thing we should want to see. (Indeed, speaking as a philosopher with some considered metaphysical opinions, I don’t really understand how Hamilton’s distinction between ‘wants’ and ‘needs’ could be made consistent at all. Very competent philosophers, such as Braybrooke (1987), have attempted this and, in my view, failed.)

Among authors represented in this volume, my meta-ethical position is thus close to that of Rosenberg (though I don’t agree with him that any moral considerations count against discounting the utility of future generations at some argued and non-arbitrary rate based on preferences of actually living people). I agree with Hausman in this volume that we must be careful to take wealth effects into close account when performing cost-benefit analyses, but I think there is a widespread exaggeration of the extent to which economists are usually ethically insensitive in practice when they recommend policies based on such analyses. In my experience, most development economists are morally sophisticated people who are relatively good at empathizing with the poor because they know quite a bit about them.

Finally, by way of general preliminaries in stating my background moral assumptions, I do not join Singer in his famous (and thoroughly argued) view that a utilitarian evaluation should directly weigh the interests of all sentient creatures. I instead follow the framework of Binmore (1994, 1998) in believing that (cognitive) ability to bargain is crucial for establishing a claim to a moral stake, and that all who cannot bargain – animals, infants, soon-to-exist people, and the severely cognitively impaired – have morally relevant interests only to the extent that their well-being matters to the well-being of actual bargainers. Fortunately, I think there are good reasons (carefully examined by Binmore) for believing that where institutions for bargaining are morally and economically well designed, the interests of non-bargainers with capacities for suffering will generally be quite strongly represented.[3] Where my disagreement with Singer on this question matters to my arguments in this essay, this will be noted; and I will confine my case against him to questions on which it doesn’t.

2. The wto as a prima facie good

On, then, to consideration of the wto. At first glance, it would be expected that to a consequentialist as described above the establishment of the wto, with its actual institutional rules and procedures, would have constituted an improvement for the prospects of developing countries. The gatt (General Agreement on Tariffs and Trade) process that the wto replaced had been one that significantly affected developing countries and least-developed countries (ldcs), simply by significantly affecting the global economy in general, but in which most of them had no participation. These countries (henceforth, for brevity, to be referred to collectively as ‘non-oecd countries’[4]) were generally exempt from the constraints of gatt, and in this respect could be conceptualized as free riders on it (Schott 2000). However, what they got for ‘free’ was mainly a ‘right’ of dubious benefit: they could pursue import-substitution policies if they chose. I need not argue against Stiglitz (this volume) that this could, given the right mix of other policies and factors, like hard-working populations and high domestic rates of savings, have been useful in some cases. It is highly unlikely, however, that it provided a net benefit against the fact that, as non-participants in the gatt, non-oecd countries could profit from international trade only when and as oecd countries chose to allow them to, for self-interested economic or strategic reasons.[5] If some countries were permitted, for political reasons, to shelter their ‘infant industries’ behind protective walls when they had no institutionalized negotiating power, it is hard to see how more institutionalized negotiating power could have harmed them under otherwise unchanged political circumstances.

As Singer emphasizes, the most important way in which the wto plays an institutional role independent of the policy choices of its member governments is through the operation of its Dispute Settlement Body (dsb). dsb panels are essentially trade courts. As with other courts, their rulings will tend to be no fairer, against a given abstract ethical standard, than the rules they enforce. I will turn to issues around the fairness of current trade rules below. First, however, we should ask whether the dsb has been procedurally unfair. This question can itself be further analyzed into two subsidiary ones. The consequences of dsb procedures, as I will also discuss in more detail later, inherit lopsidedness directly from the global imbalances in economic power. However, there is no evidence that the dsb magnifies this imbalance by being biased towards the cases of oecd countries (Hudec, 2002). Evaluated as a court, the dsb seems to be in ethical circumstances quite similar to those of domestic courts in modern democracies. That is, it is legalistically fair in its procedures, but avoids confronting or dealing with, at least systematically, the fact that legalistic fairness is compatible with special exploitation of court procedures and outcomes by those disputing parties who have more resources.

Again, I will return to the details of all this later. For the moment, I want just to make the following point. To the extent that poor people within societies, or poor nations in the world, face a choice between a legalistically fair court system, however far short of a higher standard of fairness the consequences of its operations might fall in practice, and no court at all, the first is their better option. Prior to the establishment of formal dispute settlement procedures (which began under the gatt but which required voluntary participation by both sides prior to the wto regime), poor countries couldn’t even have trade `disputes’ with industrial ones. Richer countries determined trade rules and practices amongst themselves and poorer countries lived with the consequences. Some could, and did, exploit their strategic significance in the Cold War to extract reciprocal concessions, but this was entirely an accidental and passing historical basis of (limited) power. The fact that the dsb exists, and that an oecd country can anticipate losing a case before it to a non-oecd country if the former simply acts in its self-interest without regard to treaty obligations, must constitute some minimal gain in influence to poorer countries. It is for this reason, along with the fact that under the wto non-oecd countries are at least party to the making of trade treaties (even if they are distinctly weaker parties), that I said above that the transformation of the gatt system excluding most poor countries into the inclusive wto constituted a prima facie good for poor nations.

Of course, being a merely prima facie good is the weakest possible sort of good. The wto as a whole, or the dsb in particular, could be prima facie goods for non-OECD countries and still be practically irrelevant, or even net bads, for poor countries if negative implications of their practices cancel out or overwhelm the good they promote through awarding increased voice in formal bargaining and disputes. That the wto and the dsb give non-oecd countries formal representation in trade bargaining and dispute resolution establishes only one point of importance: if we were persuaded that the current system ought to be scrapped on grounds of its unfairness, a return to the pre-Uruguay status quo could only constitute an improvement if poorer countries are better off pursuing autarchy, or at least radical import-substitution policies, than they are in some kind of integrated commercial relationship with the rest of the world. No persuasive economic analysis suggests that attempted autarchy is anything but disastrous for development, and no democratic or semi-democratic non-oecd government believes otherwise. I will therefore assume from here on that when we argue about the wto, we are arguing about reforming it rather than abolishing it.

3. Unfairness in world trade rules

So what are the main respects in which the current world trading system is ‘unfair’? And how is this unfairness directly manifest in the operations of the wto? Let me try to provide a summary descriptive catalogue. First, though, to avoid a lengthy philosophical digression on the analysis of ‘fairness’, I will just operationally stipulate a sufficient, but not necessary, condition for it weak enough to be non-controversial. I will call a practice, institutionalized norm, rule, or policy `unfair’ if it requires special, ad hoc rationalization to be defended even by reference to the prevailing ethical norms of those whose interests it best favors, and even if the people with these interests don’t regard highly unequal distributions of wealth per se as implying any unfairness. If something is unfair even against that standard, then it is unfair against any standard that allows anything to count as ‘unfair’.

And, yes, even by this standard the world trading system is unfair. Almost all advocates of freer trade defend it by reference to broadly Ricardian logic: welfare is best promoted if countries and regions can exploit their comparative advantages. The comparative advantages of non-oecd countries lie almost exclusively in lower labor costs, longer and wetter growing seasons (in most cases) for agricultural products, and direct access to some valuable natural resources. An ethical perspective according to which the wto improves on the gatt because it makes non-oecd countries party to trade bargaining and dispute resolution cannot consistently deny that this improvement substantially consists in bringing trade practice into increasingly better conformity with this minimal Ricardian logic. Yet every general source of poor-country comparative advantage listed above is systematically undermined by the policies of oecd countries and (ethically) defended by special pleas. When any non-oecd country successfully invades a new oecd market on the basis of a comparative advantage in labor costs, then unless it returns substantial profit shares to oecd-based investors – and even this provides no guarantee – it can expect its new products to be met with anti-dumping actions or other disingenuous ‘safeguards’. Non-oecd agricultural products are systematically shut out of oecd markets, while subsidized oecd agricultural products drive down profits from farming even in non-oecd countries. Whereas genuine dumping – below-cost pricing not resulting from efficient exploitation of comparative advantage – is almost non-existent in manufacturing or services, where non-oecd countries find themselves routinely accused of it, the flow of subsidized agricultural products into poor countries really is dumping. Trade in commodities most nearly approaches freedom under the rules, but in the context of widespread tariff escalation[6] this amounts to special defense of oecd comparative advantage. In general – as in the trade in services agreement (trims), and with respect to intellectual property protection rules (trips) – existing multilateral treaty provisions promote comparative advantages that oecd countries tend to have, and neglect or suppress comparative advantages that non-oecd countries tend to have. This is unfair by even the weak standard – as the overwhelming majority of economists agrees and emphasizes.

It is often pointed out that non-oecd countries interfere with one another’s exploitation of comparative advantages even more than oecd countries do. This is true, as even a well-informed source with no self-interested reason to stress the point, South African Finance Minister Trevor Manuel (2003), acknowledges. Furthermore, they often do so by means of the same disingenuous semantics over ‘dumping’ that rich countries resort to. However, from the perspective of an evaluation concerned with fairness, rather than efficiency, this is beside the point. Failure among a group of countries to achieve Ricardian improvements in general is unfortunate (for them), and may typically result from (unfair) influence of domestic rent-seekers within the countries in question; but it does not constitute unfairness in the network of relationships among these countries. What is unfair to non-oecd countries in the current international trading system is that it systematically promotes oecd countries’ comparative advantages while systematically failing to promote non-oecd countries’ comparative advantages. It is not – at least as envisaged in the multilateral processes of the wto – simply a consequence of a generalized failure of bargaining to crawl towards Pareto-efficient outcomes, as with the inefficiency of trade outside the oecd. It is a direct consequence of power imbalances which the logic of the process, by the norms that justify it according to all participants, is supposed to undermine or at least aim at discounting. To try to make this point as clear as possible: most multilateral trade relationships, outside of regional common markets, may be messes from the Ricardian perspective, but not all messes are made messes by being unfair. The rules enforced by the dsb are partly made problematic by being unfair.

We may analyze the wto’s institutional role in this unfairness into two components. First, there are aspects internal to treaty-making processes and to the contents of existing agreements. Trade in agriculture has not yet been brought within the regime, and as of this writing all progress towards incorporating it has stalled. This is despite the fact that a crucial part of the basis on which the wto was created was a commitment by oecd countries that agricultural trade rules would be regularized. The standing definition of ‘dumping’ in the Uruguay Round Agreement is uselessly flexible and economically incoherent (Jackson ,1997, Messerlin, 2000), thus inviting its manifest abuse. The nuts and bolts of multilateral negotiation go on in so-called ‘Green Room’ sessions to which most poorer countries are not invited, lack the political clout to force themselves into, and would lack the resources for full participation on an ongoing basis even if a red carpet were rolled out for them (Tussie and Glover, 1993). At wto Ministerial meetings, where treaties must be ratified, time pressures and procedural rigidity hamper the evolution of fluid logrolling dynamics that might enable poor countries to profit from vote-selling. This makes it extremely difficult for them to change the status quo, even incrementally.

Second, dsb procedural features contribute to prevailing unfairness. Poorer countries lack resources to prepare or argue cases as thoroughly or professionally as oecd countries (Delich, 2002), something that is always a source of unfairness in any strongly legalistic process. Dispute resolution is widely regarded as slow (Messerlin, 2000). Since an ongoing dispute pending resolution is typically more damaging on the margin to a poorer disputant, this incentivizes non-oecd countries to avoid disputes by making concessions, rather than to stand on the law and try to win them. dsb panels may not recommend, in the first instance, specific remedies that would constitute compliance with their decisions; if an aggrieved party regards a response by its opponent as insufficient under treaty obligations, it must open a new procedure. This of course increases the capacity of stronger parties to drag issues out and thus reap the rewards of greater patience – the basic source of power in any bargaining relationship. In any case, poor countries can seldom find retaliatory measures that significantly damage, and thus effectively sanction, rich countries whose policies seriously harm them; and given Ricardian logic, retaliation by means of withdrawal of concessions – what victory at the dsb allows – at best hits the opposing party by means of hitting oneself.

An obvious way to inquire into the extent to which the wto could be improved as an institutional engine for fairness is to ask whether and how the structural problems just catalogued could be reduced by reform. I will return to this in the final section of the chapter. First, however, I want to turn to Singer’s ethical critique, and ask to what extent it encourages us to even ask these questions. My contention will be that it mainly doesn’t, that it instead distracts our attention towards issues of distinctly secondary importance, if not outright irrelevance. Since, as I indicated above, Singer is among the most careful and responsible ethical critics of the wto, this implies that most of the populist ethical skepticism he filters through better analysis is that much further beside the point of really helping non-oecd countries.

4. Singer’s charges

Singer makes three main ethical charges against the wto. These are:

(i) The wto’s mandate, and the structure of representation in its procedures, lets commercial values trump all others.

(ii) The wto undermines (ethically legitimate) national sovereignty.

(iii) The wto is undemocratic.

As noted earlier, he also considers a fourth charge, that wto agreements and dsb rulings have made poor countries and/or poorer people within countries, worse off, but finds that known facts are inadequate to decide for or against the accusation.

On the fourth issue, I am prepared to be a bit less cautious than Singer. Only one part of the world, sub-Saharan Africa, has actually slid backwards since being integrated within the wto system. Some other parts – central Asia and the Middle East – have slid backwards while not integrated within the system.[7] Eastern Europe experienced a period of increased impoverishment before integration, but has been doing better as it’s been integrated. Latin America, quite strongly integrated, has at worst been stagnant over the period since Uruguay, though this is likely too pessimistic an assessment for the region in general.[8] In the two countries where a massive proportion of the world’s poor are concentrated, China and India, increased integration has coincided with impressive growth. Inequality within these countries has probably increased, but I know of no persuasive evidence that the poorest have become worse off. Singer is certainly right that establishing causal relationships in global-level macroeconomics is extremely difficult, so any confident pronouncement on these matters must outrun available evidence.

I suggest that, in general, we can and should side-step this issue in a way that doesn’t occur to Singer. Stiglitz (this volume) argues that we have little empirical evidence that `Washington consensus’ macroeconomic policies have consistently promoted growth. I here take no stand on this argument. What is clear, however, is that the imf and the World Bank are genuinely powerful institutions whose policies have surely made some significant difference, though probably of equivocal valence in different instances. Now, to the extent that trade liberalization as encouraged within the mandate of the wto is seen as part of a broader ‘Washington consensus’ agenda – which is the context in which Singer considers his fourth issue – the kind of assumption I just granted to Stiglitz in the cases of the imf and World Bank is much harder to justify with respect to the wto. The history of the wto to date, for reasons I will discuss below, has mainly been one of failure to achieve Washington-style objectives. I refer here not only to the fact that no real progress has been made in (global) multilateral trade negotiations since the conclusion of the Uruguay Round. It is also the case, as I will discuss in connection with Singer’s issues (i)-(iii), that implementation of the Uruguay Agreement has done at least as much to promote unintended non-Washington-style ends as those its oecd-based champions expected. In general, I will contend, the wto has accomplished so little of an intended nature, in an unequivocal and coherent policy direction – something for which the wto Secretariat is not to blame – that it would be surprising if we did or could find data implicating it strongly in global-level welfare changes in any consistent direction. I thus broadly concur with Singer’s reluctance to pronounce on his fourth issue, but not just, or even mainly, because of missing data. I think that Stiglitz is right, when trying to assess Washington-consensus macroeconomics, to focus his attention on the imf and the World Bank and say little about the wto. One can argue – though I am not prepared to endorse the conclusion without many qualifications – that the imf and the World Bank have often been hijacked by special oecd-based (especially American) interests. However, a vessel without a captain or a clear route to anywhere, which I’ll contend the wto mainly has been, isn’t open to hijacking in the first place.

5. ‘Democracy’, voting and logrolling

The charges numbered by Singer as (ii) and (iii) are often not distinguished by critics. As noted above, Singer is, like me, an ethical globalist. So are most `left-wing’ popular critics of the wto. In their frame of values, if the wto weakened national sovereignty this would not constitute an ethical objection in itself, so long as the transferred sovereignty were being ceded to democratically constituted and globally representative groups. (ii) constitutes an ethical concern just in case we suppose that national governments, where they are democratic, are the appropriate vehicles for representing people’s interests. In that case, (iii) then becomes the closely focused objection that wto procedures do not weight countries’ influence by reference to their relative populations: oecd countries have far more influence than their population shares warrant, and small countries have the same number of votes – one – as much larger ones.

Though some radical critics of the wto imagine that some sort of direct community representation would ideally by-pass national governments,[9] this is not what Singer advocates. He argues that ethical national policies, which should be pressed on international institutions by democratic member governments, would regard dictatorships or narrow oligarchies, except where these can demonstrate popular support arising from customary legitimacy,[10] as not entitled to trade national resources or to base claims for increased weight in international voting bodies on their population sizes. I agree with him on this, though the idea requires a careful theory of political legitimacy that is cognizant of the difficulties involved in applying it to cases of economically powerful governments – whose policies effect the rest of the world regardless of whether the rest of the world treats them as fully legitimate – and which may be sensitive to ranges of popular interest while violently blocking democratic reform. China of course furnishes the litmus test here at the moment.

Within this frame, we can take charge (iii) as referring to the decision-making structure of the wto itself. In practice, the wto operates by consensus: every member has a nominal veto. Singer says that this represents ‘a very strange view of democracy’ (p. 75). Relative to national and local governance rules, it of course does. But it is maintained for an important reason that Singer never considers. It is highly unlikely that most electorates in mature and economically powerful democracies would permit their own legislatures to be over-ruled by international votes on policies of importance to them. The wto would simply collapse if votes of its Ministerial Council imposed trade policies on the United States or the European Union. This fact would be of no ethical moment – though of overwhelming political moment! – if it simply resulted from the fact that most people are not ethical globalists (in the sense of weighting interests of foreigners and co-nationals equally in their ethical judgments). But ethical globalism does not necessarily imply international political majoritarianism. American and European economic and legal policies, by virtue of American and European wealth, are of far greater interest to non-Americans and non-Europeans than, say, Egyptian ones. If the rest of the world could dictate American or European trade policies it would have large incentives to actually do so, even where such dictation wasn’t mainly ethically motivated. While I do not think that a convincing argument can be made that it would be unethical for developing countries to try to determine American policy by votes if this were possible, it is equally far from obvious that Americans would be responding unethically if they resented this and opted to secede. They would surely be able to argue convincingly that such a scenario would be heavily biased in favor of redistribution of wealth independently of whether particular redistributive policies actually enhanced global welfare.

Singer will not find this objection persuasive, on grounds that a global utilitarian ethic would deny the Americans or Europeans a veto in judgment over what does and doesn’t enhance global welfare. But then I think the following consideration is relevant. We have plenty of experience with the actual workings of democracy in federal arrangements. Failures to respect power balances in assignment of regional vetoes have had baleful consequences for everyone’s welfare in federations, over and over again. Sometimes regions adopt morally unacceptable policies and should be over-ridden, as in the case of the American South before the Civil War and again after Reconstruction. However, many policy disputes are not mainly ethical disputes; many will arise simply from asymmetric information. No one knows how to make procedural rules that will allow vetoes in ‘mainly ethical’ disputes but not in others, because no one knows how to delimit ‘mainly ethical’ subjects of dispute in advance.

If we had an effective world government, so that the issue with respect to national vetoes was no more difficult than the issue of regional vetoes in federations, we would in any case be in a world far removed from the one in which we now contemplate reform of the wto. (That is, if an argument for reforming the wto turns into an argument for nothing less than world government, we have allowed the subject to change.) If, for the sake of still being able to talk about real policy issues, we confine ourselves to a world where trade rules are made by multilateral negotiations among sovereign states, then it amounts to diversion of useful critical energy, at best, or unethical institution-wrecking, at worst, to press for abolition of vetoes in the hands of the economic powers whose policies largely determine world welfare levels. And then it could hardly amount to democratic reform to urge that the economies no one can live without keep their vetoes while smaller economies lose theirs. I take it that nobody will be inclined to hold up the United Nations Security Council as a model of a democratic structure (even if large-power veto rights are a good idea there for other reasons).

As Singer of course recognizes, veto rights in the hands of less powerful countries, taken individually, are nominal in any case; Burkina Faso can’t really block a multilateral agreement all by itself. Now, this will seem at first glance to be the antithesis of democracy, as Singer suggests: it’s the un Security Council arrangement after all! But this is not in fact the case. As vividly demonstrated in the recent Cancun Ministerial meeting, a coalition of countries large enough to be economically significant to everyone else[11] can block agreement. In principle, it is the fact that everyone knows this, combined with the fact that trade treaties are ‘single undertakings’ across the whole range of issues on the agenda, that is supposed to incentivize all countries to negotiate. The consensus principle and the single undertaking principle structure wto negotiations as a logrolling process. It is an underappreciated fact that logrolling dynamics have at least one democratic virtue in their favor: they encourage revelation by negotiating parties of their preference intensities instead of just their preference orderings (see Stratmann, 1997). If Switzerland has little interest in policies that affect the world price of cotton, it is not a principle of democracy that it ought to have a vote of the same weight on this question as Benin. In a logrolling environment it effectively won’t: Switzerland might cede its marginal preference on this question to West African countries in return for their support on questions concerning the regulation of trade in banking services. Not only can logrolling dynamics be more democratic than majoritarian ones, they also, in any organization where members’ commitments to the organization’s rules are politically vulnerable, enhance stability by opening wider ranges of paths to equilibrium.

In practice, logrolling has not gone very well in the wto since the Uruguay Round. But far from objecting to the principle of logrolling, by opposing the consensus practice on ‘democratic’ grounds, my view is that we ought to be asking how we could make institutional adjustments that would allow logrolling to work. Singer is quite right in drawing attention to the green room process, for there is no doubt that a major barrier to effective logrolling has been the fact that non-oecd countries tend to get excluded from detailed bargaining until too late in the process. Since most of them do not take part in the details of agreement drafting, they rationally adopt the second-best approach of forming blocs around issues of principle, such as, at Cancun, resistance to the so-called ‘Singapore issues’ favored by the oecd, and then seeking to roll this very large log, as a single issue, against the even heavier matter of general reform in agriculture. The problem is not that these aren’t the appropriate sites for reciprocal concessions that need to be managed if the wto is to succeed. The problem is that logrolling, like much market haggling, requires tatonnement and antes of small change. Green room discussions are suitable for this; Ministerial Council meetings are not. But most non-oecd countries, as discussed above, are unrepresented in green rooms.

I have just defended the consensus principle on pragmatic grounds, not (directly) ethical ones. My point in doing so, in the context of an argument about ethics, is to emphasize that appeal to highly general ethical values, such as `democracy’, fails to speak to the considerations that actually matter to whether poor countries are able to benefit from the wto. The relevant sense of ‘democracy’ at the wto can be operationalized as follows: we will have good reason to think that we’ve made an advance along democratic lines if non-oecd countries, as the weaker parties, start to become less frustrated with the process and its outcomes. My contention is that if lively logrolling dynamics could be brought about, they would; whereas, if majority or weighted voting were used for decisions, there is no reason to expect this. Therefore, advocating a more standard democratic decision-rule for the wto for the sake of its fairness is mistaken.

I take a similar stance on the general issue of green room reform. If green rooms were official wto institutional fora, then the limited participation in them would obviously be objectionable on democratic grounds – unless, like the consensus principle, they facilitated greater weight being given to poor countries’ interests indirectly. However, the very fact that green rooms are not official fora, thereby lacking any objective institutional basis for participation in them, is just what makes them `old boys’ clubs’ – a standing image for democratic resentment. There are a number of serious proposals on the table within the wto for creating official drafting committees that would indeed have a kind of direct democratic legitimacy that green rooms cannot. However, it must be recognized that no rules can prevent the governments of the oecd countries from talking with one another in private whenever they like (which is continuously). Nor can anything prevent their joint view from carrying immense weight whenever they have one. This simply results from their economic power. If an ethical objection to the wto turns out to just be a disguised objection to power imbalance itself, then we will again have changed the actual subject of discussion from institutional reform to moral metaphysics. Remaining on topic, then: note that any new drafting mechanism intended to partly supplant the green rooms had better not insist on a representativeness principle pushed so far that oecd countries will treat its deliberations far less seriously than they do their private conversations, for then the new mechanism will come to be perceived as irrelevant and will serve to further emphasize poor-country exclusion. ‘They have a place to talk out their ideas’, it will be said, perhaps sometimes with a smirk.

Morality does demand that we not make a sham of multilateralism by failing to provide resources allowing poor countries to maintain proper, and properly prepared, representation in Geneva. (Perhaps we might also move wto Headquarters to Nairobi or Mumbai; as Singer notes, Geneva is among the most expensive cities on the planet. And it would surely not hurt if negotiators passed hovels, broken streets and hungry street children on their way to green rooms.) I will return to this issue at the end of the paper, when I describe some reforms to the wto that certainly should be advocated – and, in every case, are being vigorously advocated now, by the staff and using the resources of the wto and its sister institutions, especially the World Bank.

6. The wto and national sovereignty

Let us now turn to Singer’s charge (ii), that the wto undermines national sovereignty. For reasons given at the beginning of the previous section, we need only consider the sovereignty of democratic governments here, since, for globalists like Singer and me, sovereignty is of no value unless it promotes welfare, and dictatorships are hardly ever net promoters of welfare when evaluated against democratic alternatives.[12]

Singer argues that the wto in fact does undermine national sovereignty. He endorses the popular contention that if wto rules oblige national governments to reduce tariffs and domestic subsidies and harmonize their regulations on foreign ownership of businesses and assets, this reduces their discretionary power and ‘in the eyes of the left, makes it too easy for global corporations to do as they please’ (p. 74). Singer provides no arguments for this, other than to cite an unfortunately chosen example. The example in question is the South Africa Government’s 1998-2001 legal conflict with the Pharmaceutical Manufacturer’s Association over the former’s attempt to abridge some patent protections on drugs. (I use the vague phrase ‘abridge some protections’ because exactly what the Government wished and intended to do was subject to several interpretations, including different interpretations at different times by the Government; see Cleary and Ross, 2002.) Singer says that the pma based its case on appeal to trips provisions, and rightly notes that the us Government supported the pma’s interpretation of trips until American aids activists led it to abandon this stance for political reasons. However, Singer mistakenly claims that the pma dropped its court challenge to South Africa’s patent-abridging measures because of ‘public outrage’. As Cleary and Ross (2002) show at length, there are several things important things wrong with Singer’s version of events here (which, in fairness to him, was also the popular media’s version, both in South Africa and elsewhere). First, it was clear from the outset that the measures referred to in the South African legislation challenged by the pma were legal under trips; the core of the pma’s contention was that the legislation gave a vague domain of administrative authority to the Minister of Health, thereby creating unfair business uncertainty. More importantly, the pma withdrew its case when the Government conceded the only point under contention by agreeing that the Minister would not interpret the Act so as to allow compulsory licensing (but only parallel importation) of patented drugs. The pma allowed the Government to claim victory because, had the case reached court, aids activists who had been appointed amicus curiae were in a position to expose highly deflated profit claims on anti-retrovirals by the drug companies – and neither the pma nor the Government wished to argue in public over aids drugs. In light of all these complications, the case implies nothing one way or the other about the impact of trips on national sovereignty.

Of course, there are various other cases described in a large public literature in which multinational corporations are supposed to have put pressure on governments to comply with wto obligations. I am, however, unaware of any that bear close scrutiny. Certainly, oecd governments regularly exert pressure on each other, and on non-oecd governments, on behalf of their investors. wto provisions are sometimes cited, along with anything else that might be convenient, when political-economic strong arms are being dressed in legalistic rationales. Multinational corporations, however, generally go to considerable lengths to stay on good terms with governments, including non-oecd governments, because the former have sensitive and valuable public reputations to protect, and the latter hold strong cards with respect to influencing public perceptions. Many governments prefer dealing with multinationals, for reasons having nothing to do with the wto, because they generally pay higher wages than domestic corporations and because they are usually better placed to provide locational and other political offsets that are valuable to politicians. While scandals involving agents of multinationals, especially those tendering for government contracts, are recurrent episodes in developing countries, there is no reason to believe, based on reviewing actual cases, that wto rules have the slightest causal relation to this.

Attempts by governments to coerce each other over trade and investment are a feature of human history throughout its recorded length. What is novel about the gatt/wto regime, especially since its full multilateral flowering with the Uruguay Agreement, is that it facilitates coordination amongst governments in influencing investment patterns to a degree hitherto unknown on a global scale. As Steven Vogel (1996) has shown generally, deregulation has almost invariably involved complex bargains between governments and corporate interests that have allowed the former to promote industrial policies by new means, and without incurring expensive management liabilities that strain national accounts. (Those who fear for loss of government sovereignty often forget that indebtedness can restrict policy discretion more severely than any other single impediment. Public ownership of, or extensive subsidies to, utilities and large industries are almost invariably drains on the public purse.) Ronald Jones (2000) has applied this general point specifically to international trade dynamics in the post-Uruguay environment. A deep source of government power is capacity to control the location of economic activity. In the absence of multilateral trade coordination of the sort that the wto regime, along with regional trade agreements, makes possible, governments seeking foreign investment may find themselves in competitive inducement races, which lower their individual bargaining power. By grouping together as, in effect, a regulatory cartel, the world’s governments as a sector can substantially increase their power.

The specific nature of wto treaty agreements compounds this. In each gatt round up to and through Uruguay, average tariffs were sharply reduced, but governments retained discretion over which of their productive sectors would be the sites of the reductions. This has incentivized industries around the world to bid, using political and policy favors (and sometimes, of course, illegal bribes), for preferred slots in tariff reduction schedules. Governments collect the substantial political rents anted in such bidding. The fact that only governments have the power to launch anti-dumping actions, or invoke other wto-sanctioned safeguards, gives them yet another source of power for which industries rationally pay them.

When I alluded, earlier in this chapter, to the absence of a clear macroeconomic or political-economic policy direction steered by the wto since its inception, the dynamics described above are partly what I had in mind. They are distinctly not the sorts of outcomes that Ricardians or free-marketeers – or Mike Moore, the most visible public face of and source of energy in the wto during his tenure as Director-General (see his 2003) – had in mind when they celebrated the conclusion of the Uruguay Agreement. The other main respect, of course, in which the wto has not successfully driven a liberalizing agenda lies in the fact that it has had difficulty driving any agenda because the cartel members have been unable to agree on extensions of their original treaty, as evidenced by the breakdowns in Seattle and Cancun. The proliferation of regional and bilateral deals has contributed to the difficulties, since the value of such deals to signatory governments – especially non-oecd governments – lies precisely in gaining special market access from which competitors are frozen out. The better a given government’s bilateral and regional arrangements from its own point of view, the less enthusiastic it will become about multilateralism, an attitude that will be shared by domestic producer interests that undertake investments based around the country’s special access.

In evaluating the claim that the wto undermines national sovereignty, we must also ask what governments might actually wish, importantly, to do that the wto is supposed to interfere with. Every government on the planet presently gives preference, using thickets of loopholes and special exemptions clauses in the Uruguay Agreement, to domestic producers (especially of services) where internal bargaining over rents leads them to want to. Stiglitz, in this volume, worries about loss of freedom by non-oecd governments to limit short-term capital flows. However, given the highly incomplete status of the trims Agreement, the relevant part of the Uruguay package here, it is hard to see how this worry, to the extent that it is legitimate, has much to do with the wto. I think there is indeed a strong case to be made that short-term volatility in capital flows into and out of small non-oecd countries is a problem. However, it is far from obvious that pressure from oecd governments, let alone from the wto, is the main obstacle to tackling it. When Malaysia interfered with such flows in 1998, it was sharply criticized in the oecd but not stopped. Governments must live with crises caused by sudden outflows because, in most cases, they find periods of high inflows politically irresistible for domestic reasons. When the South African Rand rose dramatically during 2003 on the back of capital inflows attracted by high interest rates, it was the corporate sector that publicly fretted over damage to its export revenues, while labor unions celebrated the increasing spending power of their members. The South African Reserve Bank lowered interest rates much less, and much more slowly, than either corporate interests or economists urged. But governments and central banks cannot have it both ways, enjoying inflow surges without risk of outflow surges. Again, the relevance of the wto to this issue is, unless trims comes to be substantially broadened and strengthened by agreement, minimal.

Singer emphasizes two particular respects in which he fears interference with sovereignty that he can specifically link to international trade rules. As an environmentalist, he is anxious about the fact that these rules restrict countries’ rights to refuse imports of products harvested or extracted in ways that might be ecologically harmful. As a person of the left, he is also concerned that attempts to incorporate labor standards into the wto treaty have so far been largely unsuccessful. Indeed, these two fears are the main basis of his charge (i) against the wto, that it encourages ‘economic’ values to trump ‘non-economic’ ones, which I have not yet considered.

Singer rightly notes that resistance to incorporation of environmental and labor standards has come mainly from non-oecd governments, on grounds that such incorporation would disproportionately threaten their comparative advantages – thus amplifying the very dimension along which I have agreed that current trade rules are unfair. He acknowledges the point, but severely underplays it – not mentioning, for example, that non-oecd anxiety about environmental and labor-standards measures was a main cause of the breakdown of the 1999 Seattle Ministerial – and applies nothing like his usual argumentative rigor to consideration of it. Admitting that stricter environmental and labor standards in trade rules `could’ impinge on non-oecd comparative advantage he rhetorically asks ‘[B]ut what is the alternative?’ (2002, p. 95). This is surely not good enough. Less expensive labor, and much lower regulatory costs faced by companies, are the overwhelmingly greatest sources of non-oecd comparative advantage, the main reasons why anyone trades with or invests in them at all (except in commodity extraction, where Singer’s environmentalist concerns are also engaged in a way troubling to non-oecd interests).

As I mentioned in the first section of this chapter, I disagree with Singer’s belief that there are intrinsically valuable things besides human welfare (specifically, in his case, animal welfare). I of course agree that biodiversity and a healthy ecology are important to human welfare, but that is why I think they are important. Now, pursuing this disagreement here, were I to try to do so adequately, would take me too far away from topics in development economics and too deep into metaphysics. The ultimate basis of my disagreement with Singer is that I consider it to be a metaphysical fact, forced by considerations of coherence with a broader scientific world view, that there are no values to be ‘found’ in the world independently of what collective bargaining amongst agents reveals. In particular, that inflicting or tolerating suffering by sentient beings is wrong is not such a fact. (For the reader interested in the relevant arguments, I recommend Binmore, 1998.) Animals cannot bargain, nor can unborn or infant people. If current adult people wish to bargain on their behalf, I wholeheartedly encourage them, but then such wishes are to be treated as aspects of the welfare of those people. The fact that I can go no further (here) into this topic of controversy with Singer means that I cannot claim to wholly answer his case for worry about the wto. I do insist, however, that in his failure to seriously consider the conflict between his favored `non-economic’ goods[13] and the economic welfare of people in non-oecd countries, he dodges what ought to be the toughest issue for most people who think that policies giving overwhelming emphasis to economic growth are in some degree wicked.

Let me conclude this section with a forthright value claim. Each person gets only one finite life. If that life is stunted and blasted by poverty, especially poverty against which the person has no effective remedy because of the actions and attitudes of other people, this is a moral atrocity. In the present world, only terroristic behavior, mainly perpetrated by some governments, is as or more morally objectionable than failure to take poverty-alleviation seriously. For reasons I have given in this chapter, I therefore think that an ethical stance that would undermine the wto instead of supporting its own efforts at reform of the global order is morally irresponsible at best and outright immoral at worst. Now, as I said at the outset, Singer joins me in favoring reform of the wto, so I am not accusing him of either moral irresponsibility or immorality. (There are many other commentators, regularly heard from, who I would so accuse.) However, for reasons I also hope to have made clear, I don’t think that focusing on highly abstract charges like Singer’s (i)-(iii) is a potentially productive way of promoting poverty alleviation at the global level.[14] Instead, we should begin by concentrating on specific ways in which current practice is unfair, and try to effect specific institutional reforms. Development economists, including those who work for the wto (and the other global economic institutions) have given careful thought to many such reforms – unlike, I regret to say, philosophers. I will conclude this chapter by listing some of the best suggestions (among those that are relevant the aspects of unfairness I have considered here).

7. Improving the wto

The list of reforms I will now enumerate must, for reasons of space, be a mere list. I will not be able to provide arguments for and critical discussions of each one. In many cases, however, the considerations that would make them improvements are self-evident. In any event, my purpose in listing them is mainly to indicate how much more substantive they are than anything that follows from mere abstract reflection. (A reader seeking a more comprehensive and less rushed survey should consult Hoekman and Martin, 2001.)

Most of the unfairness associated with the wto results simply and directly from power imbalances. Power imbalances, in turn, stem primarily from differential economic resources. Therefore, the most important reforms for which moral critics of current trade rules and practices should press involve commitment by oecd countries to provision of meaningful technical assistance to non-oecd ones. Rhetorical pledges to such commitment are widespread in wto statements and agreements – especially the Doha Declaration and associated documents – but run far ahead of institutionalized actions.

Those who are morally concerned about trade rules and practices can begin by lobbying their governments to supply much higher funding levels for the United Nations Council on Trade and Development (unctad) and the International Trade Centre (itc), since these agencies have demonstrated strong competence, within their existing budgets, in enhancing the bargaining capacities and access to information of non-oecd negotiators. Two multilateral initiatives, the Joint Integrated Technical Assistance Program for Selected Least-Developed and Other African Countries (JITAP, 1996) and the Integrated Framework for Technical-Related Assistance, Including Human and Institutional Capacity Building to Support Least-Developed Countries in their Trade and Trade-Related Activities (IF, 1997), have recently been strengthened and expanded following acknowledgment by oecd governments that, as a result of their own insipid and miserly support to unctad and the itc in implementing them, their initial results have been inadequate (Luke, 2002). Since those aspects of the plan that unctad and the itc were able to roll out have been effective – contributing to the success of developing countries in making their strategic weight felt at the Seattle, Doha and Cancun Ministerial sessions – disappointments with these initiatives do not imply that they should be given up on; quite the opposite (Ibid). These programs merit still further broadening and deepening, to extend them to non-oecd countries throughout the world, and to promote leadership roles in their implementation for middle-income developing countries who can profit from them by being funded to partner with the trade agencies in steering them.

oecd countries can and should provide other forms of technical assistance directly through the wto, and through government-to-government commitments. Especially important are efforts to enable non-oecd countries to harmonize and modernize their Customs facilities and procedures (Staples 2002) – since problems in this area are a major source of dead-weight trade barriers – and to be able to identify, verifiably demonstrate and seek redress for unintended or underestimated implementation costs imposed on them by the trims and trips agreements. Direct financial assistance can and should be provided to ensure that all wto members can maintain adequately sized and budgeted delegations in Geneva on a full-time basis. In concert with this, wto Headquarters could be moved to a less expensive site in a non-oecd country, with appropriate infrastructural support provided. Measures such as these all amount simply to shifting more of the costs of enhanced fairness to oecd taxpayers. Persuading them of their obligations in this regard obviously requires moral argument, of the sort that Singer has been in the forefront of articulating. However, the relevant moral arguments are much more likely to be effective if they are accompanied by demonstrations that such specific measures as have been properly implemented to date have paid off efficiently – as they have. Equally importantly, the arguments should be linked to emphasis on Ricardian logic – reminding oecd citizens that building richer markets abroad enhances their own opportunities for exporting their goods and services – rather than in tones suggesting that existing trade institutions are morally corrupt.

‘Throwing money at’ problems is not always an ineffective way to solve them, where we have both empirical experience of, and decent economic models to explain the dynamics of, specific types of applications in which extra resource infusions multiply good effects without drowning them in new rent-seeking opportunities. I have been listing some such types of application. Another is provision of expensive legal resources, through global trust funds, for non-oecd countries bringing cases to the dsb. Friends of fairness should relish the idea that when teams of lawyers from quad countries examine the legal briefs and arguments accompanying cases brought by Ghana or Surinam or Papua New Guinea, they might find them prepared by legal firms with Manhattan or Chicago addresses. The value of the resulting fair fights would surely exceed the admitted costs of rents to the lawyers.

Numerous other measures for improving dsb procedures have been suggested. (see Jackson, 2000 and Delich, 2002 for surveys.) Countries that have won cases should be able to appeal to compliance review panels within short time periods after remedies have been identified; ideally, such panels should be constituted by the original panels that considered the cases in question, and invested with binding writ. Time frames for all dsb procedures can and should be shortened. The dsb could and should encourage collective retaliation against countries found to be in breach of their treaty commitments, as this is one of the few ways in which the imbalance in sanctioning capacities can be addressed. Another is to allow countries that win dsb cases over breaches in one trade mode to respond by withdrawing concessions in another. If Pakistan is harmed by a Japanese tariff, it might not be able to worry many Japanese by use of a countervailing tariff of its own; but withdrawal of a concession to Japanese firms under trims, or abrogation of a Japanese patent right under trips, might keep some extra lights on after hours in Tokyo. (In 1996-97, Ecuador attempted exactly this in a dispute with the eu over banana exports. The resulting dsb opinion did not close the door on such measures, but spun a substantial cloud of confusion through hesitant and opaque reasoning (see Delich, 2002). Finally, Hoekman and Mavroidis (2000) have suggested that a kind of `special prosecutor’s’ office be established by the dsb, which could expedite case assessments and encourage ‘out of court’ settlements, especially in cases where alleged economic damages are small.

It is not widely appreciated that by comparison with its sister organizations, the imf and the World Bank, the wto Directorate has relatively little ability to exercise policy influence, because its staff and discretionary budget are tiny. This makes the animus directed at the organization by radical critics ironic; if it tried to be a site of special economic manipulation independent of its member governments, it would find its capacity for fomenting conspiracies greatly limited. In fact, the wto Directorate is the one player on the scene that is straightforwardly incentivized to try to broaden both the reach and the fairness of wto agreements. Developing countries and their supporters should therefore press for its strengthening.

A general aspect of current trade practice that magnifies power imbalances is structural failure to tote up all injuries to parties other than those represented by particular non-oecd governments when the latter come into conflict with oecd producer interests. For example, when regional trade agreements (rtas) cause trade diversion that negatively impacts third parties, they have no recourse through the wto. A new institutional procedure that addressed this could substantially blunt the potentially disastrous dynamic, discussed earlier in this essay, by which rtas create new rent-seeking coalitions opposed to multilateral progress, and which become wedges by which non-oecd interests are set against each other. Similar logic underpins what would likely be the single most effective (politically feasible) measure for combating the plague of anti-dumping actions. If wto members agreed to harmonize their procedures in domestic anti-dumping inquiries so that domestic users (especially industrial users) of the imported products under scrutiny were automatically represented in official considerations of dumping allegations, then at a single stroke the systematic interference with market access caused by this ‘safeguard’ would be addressed, and non-oecd governments could expect to find procedurally implicated allies within the political bargaining dynamics of oecd countries (Finger, 2002).[15]

This last suggestion leads directly to a more general point. Moral critics of the wto usually direct their arguments, as Singer does, to an abstract and generalized ‘moral conscience’. That this is the typical mode of address adopted by philosophers is unsurprising; they are supposed to be professionally concerned, first and foremost, with the logical power of arguments, independently of considerations of audience. However, this can and does encourage failure to recognize that objectively sound arguments can be effective economic weapons if they help to isolate special burdens of moral responsibility in interest sectors much more specific than ‘oecd citizens’ or ‘wealthy people’ in general. For example, I think it unlikely that meaningful reform in agricultural trade rules – the single most urgent area of manifest current unfairness – will happen until and unless the non-oecd countries and their friends launch a carefully planned, concerted and well-financed public-relations campaign in oecd countries that clearly paints subsidized agricultural interests as the outrageously successful rent-seekers they are. It is a clear and demonstrable fact that their rent-seeking not only does grievous harm to poor countries, but also damages oecd consumers by raising food prices and by burdening tax bills with (very large) subsidy costs. However, it is difficult to make these dry efficiency considerations into vivid and culturally contagious images of the sort that the modern advertising industry is skilled at producing. But philosophical argument can soundly press the conclusion that wealthy, subsidized and inefficient farming corporations in oecd countries are morally vicious. A philosopher who fastidiously objected to seeing the conclusions of his or her good arguments turned into politically effective slogans, sound-bites, ditties and images would thereby, I suggest, show questionable commitment to the seriousness of the moral case he or she makes. Let us advertise against rich-world rent-seekers – starting with farmers – and do what we can to make them targets of cultural derision. We can begin by not mincing words: they are stealing (since that is what rent-seeking is) staggering quantities of money from the world’s poorest people.

There is a tradition in moral philosophy, especially in radical moral philosophy, of regarding advocacy of incremental structural reforms to institutions as missing what is truly morally serious by compromising with a bankrupt status quo. Of course, if someone is convinced that they have a good case for regarding trade for profit amongst private business interests, in general, as a wicked thing, then they will not be persuaded by anything I have said. But it is equally unlikely that such people will persuade oecd electorates to adjust their moral priorities, so however much television excitement they create, their political importance is minor. Fortunately, many moral critics of the wto base their cases on the fact that too little is being done, too slowly, about poverty. If such critics are sincere, they should contrast in their imaginations two possible mass-media advertising campaigns. One would hector whole oecd societies for being selfish and hypocritical about the value of genuine, global, democracy. Another would emphasize that various narrow interests – and not, mainly, the multinational corporations in which millions of people make their livings – are being allowed to prevent the wto from fulfilling its mandate to improve global welfare. Which of these campaigns would be more likely to change the minds of citizens in wealthy countries? Until these minds are persuaded, the interests of poor countries will continue to be damaged by the stalemate over agricultural trade, by preposterous anti-dumping findings, and by demands that poor countries implement complex reforms for which they lack infrastructural capacity. It is not even necessary to go outside the current institutional framework in pursuit of significant changes to benefit the environment. For example, as Esty (2000) points out, it would be fully consistent with the mandate of the wto to pursue zero worldwide tariffs on environmental goods and elimination of (presently extensive) non-tariff barriers on environmental services. Such a measure would almost certainly do more long-run environmental benefit than complaining that national governments, when they meet with one another to discuss world welfare, focus too narrowly on ‘economic’ issues.

It is good that a leading moral philosopher like Singer has put the case for a cosmopolitan and consequentialist ethic of globalization. What follows from such a case, however, when couched only in terms of very abstract values, is frequently non-obvious. Singer’s discussion is in one sense at the level of institutions: he morally evaluates the wto. But in holding it up only to the light of highly abstract principles, with little attention to the strategic wisdom its design embodies, he gives the impression that direct conformity to the principles is the road to goodness for the institution. Here, I have been disagreeing with this. The wto can contribute to an ethically better world mainly be being made more effective. Let us all – economists, philosophers and citizens alike – avoid using an ethical perspective to make the best the enemy of the good. With millions of lives waiting for opportunities to be good lives, this would itself be unethical.

Notes

1. Philosophical preliminaries

3. Unfairness in world trade rules

4. Singer’s charges

References