Trade, Poverty, Development

Getting Beyond the WTO's Doha Deadlock

Rorden Wilkinson, James Scott, Rorden Wilkinson, James Scott

• Publisher: Routledge
• Year: 2012
• ISBN:
  9781136238734
• Edition: 1
• Topic: Politica e relazioni internazionali
• Subtopic: Politica

This work seeks to look beyond the seemingly endless deadlock in the WTO's Doha round of trade negotiations that began in November 2001 and were first scheduled to conclude by January 1, 2005. As well as offering an incisive analysis of the ills of the round, with particular attention directed at the poorest and least developed countries, the book expands on how the round could be moved forward elaborating on the Statement on the Doha Development Agenda that was negotiated in Johannesburg.

Thework as a whole provides the reader with a critical analysis of the implications of the negotiations for development and poverty reduction as well as proposals for moving beyond the current impasse. The volume brings together contributions from serving and former ambassadors to the WTO, key practitioners, and civil society representatives along with those of leading scholars. Eachchapter explores an area of critical importance to the round; and together they stand as an important contribution to debates not only about the Doha round but also about the role of trade in the amelioration of poverty in the poorest countries.

Part I

The round

1 The poverty of the Doha round and the least developed countries

James Scott and Rorden Wilkinson¹

The November 2010 meeting of the G-20 in Seoul concluded with a familiar “must do harder” refrain on the World Trade Organization’s (WTO) Doha round (DDA—the Doha Development Agenda). Couched in terms of a belief that “trade can be an effective tool for reducing poverty and enhancing growth in developing countries [and] LICs [low income countries] in particular”² and of “recognizing the concerns of the most vulnerable,” the leaders of the G-20 countries publicly stated their commitment to a “successful, ambitious, comprehensive, and balanced conclusion” consistent with the original Doha mandate, encouraging WTO members to build on “the progress already achieved” with a view to reaching a conclusion to the round in 2011.³ Few commentators took this aspect of the Seoul Declaration seriously, taking it to be a well meaning, but nonetheless overly optimistic, attempt to paper over the deep divisions that had emerged among member states over key aspects of the negotiations.

Just 12 months later, at the G-20’s November 2011 meeting in Cannes, attempts to paper over the cracks were conspicuous by their absence. The Final Declaration put it bluntly: “It is clear that we will not complete the DDA if we continue to conduct negotiations as we have in the past … we need to pursue in 2012 fresh, credible approaches to furthering negotiations, including the issues of concern for Least Developed Countries and, where they can bear fruit, the remaining elements of the DDA mandate.”⁴

The G-20’s Cannes Declaration is a sobering acknowledgement of the stasis currently afflicting the round. The negotiations have coughed and spluttered since the round was launched in 2001, passing through the not-as-yet-repeated high point of July 2008 when they came closest to an agreement before abruptly coming to a halt shortly thereafter. A number of the leading players, particularly the United States, have shown little in the way of the kind of engagement required to conclude the negotiations. Moreover, what we previously identified as an “emerging consensus”⁵ on the likely results of the round for the poorest countries has since been consolidated and few now believe the round would offer the least developed countries (LDCs) much, if anything, of value.

Yet, while this consensus has crystallized, just how little the poorest countries would gain from the round is not wholly apparent. The results of most computable general equilibrium (CGE) simulations (more-often-than-not the preferred method of estimating how much is to be gained from concluding a trade round) offer little comfort for LDCs.⁶ Since the round was launched these simulations have steadily predicted that the outcome of the DDA for developing countries will be small, with the poorest and most vulnerable faring worst. Moreover, as the round has progressed the predicted gains for both of these groups have fallen significantly.

That said, the extent of just how poor the gains are for developing and least developed countries as a group is not fully captured by economic models. While they offer useful insights into the aggregate picture of the projected Doha gains (or losses as is increasingly the case), CGE models often lack the kind of specificity that enables a detailed insight into not only the projected gains but also the consequences and congruity of particular kinds of liberalization for particular developing and least developed countries.

A second body of literature has taken a different approach to exploring the likely outcome of the DDA.⁷ Rather than focusing on the projected gains from liberalization based on scenarios drawn from different points in the negotiations, these studies use detailed examinations of the pattern and progress of the negotiations as a basis for exploring the likely opportunities accruing to different countries from a concluded trade deal. This literature too, however, is not entirely satisfactory; though like the CGE models it nonetheless conveys a sense of the poverty of the Doha round. The strength of these studies is that they have been able to offer more precise and tailored analyses of the impact of particular kinds of liberalization for specific countries. Their weakness, however, is that they have only been able to talk in vague terms about the relative imbalance of opportunities resulting from a potentially concluded DDA without offering a detailed and precise global portrait.

There has, however, been little engagement between these two bodies of literature—their methodological and disciplinary differences more often than not proving too great a divide. One result is that few modeling the likely outcome of the round connect their studies to the political machinations of the negotiations in pursuit of a cogent answer as to why the structure of the round is likely to yield such poor results. Likewise, few political economy approaches engage with the econometric literature beyond a fleeting reference to poor aggregate projections as a means of reinforcing the point that Doha is unlikely to result in significant gains. This lack of engagement presents something of a problem. While both literatures agree that the likely results of the round will be poor, what neither offers alone is a pointed appreciation of just how poor, and indeed problematic, the results are likely to be and how such a situation has come about.

Our aim in this chapter is to get a more complete sense of just what the poorest countries are likely to gain from the DDA by bringing together the insights of the econometric and political economy literatures. We do this by fusing a review of the CGE projections of the likely gains from the DDA with an examination of the passage of the Doha negotiations. We find that, when viewed in combination, the econometric and political economy literatures point to an outcome for developing countries that is not only poor in the aggregate but also locks them into a particular kind of (largely raw material-based) production and crowds out the possibilities for diversification. Thus, we argue that the focus of negotiations in the round should not only be on finding ways of increasing the liberalization “cut” to enhance the projected aggregate gains for the least developed. It should also be on addressing the invisible barriers to industrial diversification that are thrown up by particular patterns of negotiating. In extension, we argue that for the Doha round to be more successful, and to better approximate the sentiments of the G-20 statement quoted earlier, a qualitative shift in the negotiations is required. In pursuit of our aim, the next section reviews the econometric literature on the projected gains from Doha. Thereafter, the paper turns to the passage of the negotiations before offering our concluding comments.

Simulating the DDA

Since their introduction around the time of the Tokyo round (1973–1979)⁸ estimates of the prospective economic gains from trade liberalization based on CGE and partial equilibrium modeling have ballooned. Hess and von Cramon-Taubadel compiled a set of 1200 such studies published between 1994 and 2006.⁹ Most of these studies utilize the freely available model developed by the Global Trade Analysis Project (GTAP), or else they are based on the World Bank’s LINKAGE model (which also uses GTAP datasets).

Inevitably (both because of the proliferation of these models but also their claim to offer more precise assessments of the outcome of any prospective trade deal), CGE modeling has come to play an increasing role in WTO negotiations, informing the negotiating positions of the member states.¹⁰ It is unsurprising, then, that peaks in the production of these studies have arisen at key moments in the negotiations, such as in the immediate run-up to WTO Ministerial Conferences. What is interesting about these studies is that there has been a marked diminution in the predicted gains arising from the DDA across the course of the negotiations¹¹—a factor that has contributed to the relatively agnostic approach that many negotiators have developed to the round’s conclusion. The average predicted global welfare gains made by the studies sampled by Hess and von Cramon-Taubadel preceding the launch of the DDA, for instance, peaked in 1998 at $250 billion, before falling to a trough in 2003 of around $35 billion, and subsequently climbing again to around $90 billion in 2006 (see Figure 1.1). These figures, of course, mask wide variations. For example, for full liberalization of goods and services Brown, Deardorff, and Stern (2003) claim global welfare gains of some $2,080 billion, while Francois, van Meijl, and van Tongeren (2003) predict only $367 billion.¹² Similarly, when examining a likely DDA outcome on trade in goods (that is, through agricultural and Non-Agricultural Market Access [NAMA] liberalization), Harrison et al. (2003) predict global welfare gains of $186 billion, with non-Organization for Economic Cooperation and Development (OECD) countries receiving $97 billion; while Anderson and Martin (2005) predict only $38.4 billion globally with a meager $6.7 billion going to developing countries.¹³

Figure 1.1 Average welfare gains predicted by CGE studies, 1996–2006

Source: Adapted from Sebastian Hess and Stephan von Cramon-Taubadel, “A Meta-Analysis of General and Partial Equilibrium Simulations of Trade Liberalisation under the Doha Development Agenda,” The World Economy 31, no. 6 (2008): 812.

There are several reasons for the wide spread in predicted gains, though this is not the place for a detailed analysis.¹⁴ The principal differences lie with the assumptions made, particularly the choice of Armington elasticity (which sets the elasticity of substitution between products of different countries), and whether or not tariff cuts are assumed to be made on bound or applied rates. As with all CGE modeling, the outcome is highly dependent on the assumptions made. The gains for developing countries in particular are highly dependent on the Armington elasticity chosen.¹⁵ Doubling the Armington elasticity, for instance, increases predicted global welfare gains by 96 percent, gains of developing countries as a group by 119 percent, and gains for sub-Saharan Africa by 423 percent. The Armington values in the standard GTAP model are taken from estimates made for seven countries (Argentina, Brazil, Chile, New Zealand, Paraguay, United States, and Uruguay).¹⁶ It is an open question how valid they are for the rest of the world and across sectors,¹⁷ and the Armington elasticities used in key World Bank studies have been argued to be greater than justified by econometric evidence.¹⁸ As noted earlier, their choice has profound implications for the expected distribution and size of benefits from the DDA. Notably for Africa, the choice of Armington values can have dramatic consequences for the continent’s predicted gains, or indeed convert predicted gains into losses.¹⁹

Compounding the issue of falling returns, some of the assumptions in the standard GTAP model are inappropriate for developing countries and correcting these anomalies can further reduce the expected gains. First, in many developing countries tariff revenues constitute a significant portion of total tax revenue. Tariff liberalization will lead to falling tariff revenue, which must be made up elsewhere if government spending levels are to be maintained.²⁰ Valenzuela, Anderson, and Hertel have examined the effect of modifying the GTAP standard model such that tariff revenue losses are balanced by increased indirect consumption taxes (since most developing countries lack the capacity to raise direct taxes). They find that the expected welfare gains in developing countries as a group fall by 17 percent, and for sub-Saharan Africa by 25 percent.²¹

Second, CGE models usually assume either full or less than full but fixed employment. That is, the model assumes that any person who loses his or her job will find new employment in a different sector. This removes by assumptive fiat the central political issue with regard to trade liberalization—how trade liberalization affects the level of employment. Particularly in developing countries, which tend to have high levels of un- or under-employed labor, a more realistic situation might be for displaced workers ending up joining the ranks of the unemployed.²² Some attempts have been made to integrate employment into CGE models,²³ but the caveat applies that the effects of trade reforms on employment are too complex and context specific to be readily modeled. A broad outline can be drawn that in countries in which production is falling, employment will likewise fall.²⁴ For some countries, notably China and Vietnam, a reduction in agricultural employment as a result of the DDA is likely to be counterbalanced by an increase in employment in manufacturing, because these two countries are frequently found to be the biggest potential beneficiaries of liberalization in non-agricultural goods (so-called NAMA). Other countries that are not competitive in manufactures, particularly in Africa, are likely to see unemployment rise. So countries that are already expected to be net losers from the DDA, particularly if they are seeing both shrinking agricultural and manufacturing sectors, may be expected to suffer greater losses if employment effects were included. Even for those countries that are expected to benefit from liberalization, there are impacts on the quality and distribution of jobs that are of political significance but are lost in the CGE framework.

Third, there is a mismatch between the model of development (and particularly within that, of industrialization) that developing countries generally subscribe to and which underlies CGE models. A key element of development, or at the very least a key aim of developing countries, is to change the range of products that they produce, either to industrialize or to move into higher value added sectors. Though it finds little support within neo-classical economics, one of the most important policies used to this end among all industrialized (both old and new) countries has been the use of trade restrictions to support the growth of new and infant industries.²⁵ CGE simulations calculate the welfare gains from moving factors of production into those in which the country has, at present, a comparative advantage, and from allowing consumers access to goods at the cheapest price. This sits uneasily, however, with the aims developing countries have of creating new areas of comparative advantage, often through the use of targeted trade restrictions, and of sacrificing present consumption for higher future returns through repositioning their economy within the global trade system. Of course, this tension merely reflects the on...