ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Trade Liberalisation for Development? Who Gains? Who Loses?

A number of models - developed in particular in the World Bank - show large gains for developing countries through trade liberalisation at the World Trade Organisation. These models are, however, not just simplistic, they also suffer from a number of fundamental flaws. The actual gains are far smaller for developing countries and far greater for the rich countries. The models ignore the risks of displacement, economic downturns and rising debt.


Trade Liberalisation for Development? Who Gains? Who Loses?

K S Jomo, Rudiger von Arnim

impacts (e g, Ocampo, Jomo and Khan 2006, especially the chapters by Cornia, Lee and Hoekman & Winters). Not surprisingly, such CGE models rely on poor theory (Ackerman and Nadal 2004), overestimate the benefits and neglect the costs of free trade.

World Bank projections2 of the gains from complete trade liberalisation (Anderson

A number of models – developed in particular in the World Bank – show large gains for developing countries through trade liberalisation at the World Trade Organisation. These models are, however, not just simplistic, they also suffer from a number of fundamental flaws. The actual gains are far smaller for developing countries and far greater for the rich countries. The models ignore the risks of displacement, economic downturns and rising debt.

K S Jomo ( is with the Department of Economic and Social Affairs, United Nations, New York and Rudiger von Arnim ( is at the Department of Economics, University of Denver, Colorado, US.

Economic & Political Weekly

november 29, 2008

lmost a decade has passed since developing countries’ delegates walked from the negotiating tables and protesters took to the streets in Seattle in 1999, as they saw their interests and concerns neglected at the World Trade Organisation (WTO). Since then, rich countries have pledged to put development first, explicitly making it the aim of the ninth round of trade negotiations started in Doha, Qatar, in late 2001, dubbed the “Doha Development Agenda”. The ongoing Doha round of multilateral trade negotiations at the WTO ground to a halt in Cancun in 2003, experienced further hiccups in Hong Kong in 2005, and appears destined to stall for a while longer.

The political imperative of delivering a push for development in the South, especially after the record of almost three decades of Washington Consensus-style reforms, is clear. What, though, are the links between trade liberalisation and develop ment? What do empirical modelling exercises tell us?

Gains from Trade Liberalisation?

The World Bank has provided crucial support to the WTO in pursuing trade liberalisation. The Bank’s projections have been made on the basis of a computable general equilibrium (CGE) model of trade and production,1 the so-called LINKAGE model. Like so much of international trade theory, the models make problematic – or at least moot – assumptions, such as full or fixed employment, while focusing only on the effects of price changes.

Indeed, trade liberalisation is likely to cause unemployment or to lower incomes in previously protected, internationally uncompetitive activities. Further, studies of the effects of trade liberalisation do not point to significant employment generating and Martin 2005) have been significantly revised downwards from earlier estimates just a few years before. More than 70% will accrue to rich countries, including twothirds of the global benefits from agricultural trade liberalisation, and even more for non-textile manufacturers. More than two-thirds of the static gains to developing countries from trade liberalisation accrue to Argentina, Brazil and India in the case of agriculture, and to China and Vietnam in the case of textile and garments.

As full trade liberalisation is not under negotiation in the Doha round, Anderson and Martin (2005) also considered several possible Doha round scenarios of trade liberalisation. Their most realistic scenario projects welfare gains in 2015 of $96 billion, a third of their estimated gains from full trade liberalisation – again with much greater gains for rich countries, which stand to gain $80 billion, or 82% of the potential gains of full liberalisation, compared to $16 billion for developing countries, or only 18% of such gains.

Furthermore, contrary to the claims of advocates of agricultural trade liberalisation, eliminating agricultural and export subsidies in the Organisation for Economic Cooperation and Development would actually cause net welfare losses in developing countries! The supposed gains from agricultural trade liberalisation are likely to bring greater benefits to a few rich agriculture exporting countries – rather than to most of the developing world, let alone the bulk of the poor. Countries already enjoying preferential market access – such as Mexico (through the North American Free Trade Agreement), central American countries (via Central American Free Trade Agreement), most African countries and some of the other least developed countries – are likely to lose most from trade liberalisation due to preference erosion.


Other estimates suggest even more modest gains, with ostensible gains – and their impacts on poverty and inequality – very sensitive to assumptions, definitions and measurement (e g, Ackerman 2005). Another CGE model-based Doha agricultural trade liberalisation scenario found that rich countries would gain $19 billion, China and south Asia $1 billion each, while other developing countries would lose $3 billion (Bouet et al 2004).

The likely contribution of such scenarios to poverty reduction varies greatly, and is further limited by the declining contribution of economic growth to poverty reduction due to rising inequality. In view of the historically critical role of trade policy reforms favouring growth and employment – as opposed to trade liberalisation – for economic development, the consequences of trade liberalisation for sustainable development are dubious (Chang 2007; Reinert 2007).

Using a simplified, but structurally similar model, Taylor and von Arnim (2006) show how heavily trade liberalisation simulation results depend on assumptions. Allowing a bit more realism – unemployment, for example – makes clear that Africa will not gain, on balance, from trade liberalisation. Their exercise suggests that sub-Saharan Africa is likely to experience welfare losses, even assuming the absence of macroeconomic shocks. The region is likely to experience a worsening trade balance, debt problems are likely to increase, and any short-term gains in employment and GDP are could evaporate quickly under the pressure of such strained balances.

Even though his model’s details differ, Kraev’s (2005) “alternative” analysis of the effects of trade liberalisation on GDP has a methodology and aims compatible with those of Taylor and von Arnim. Endogenising output, employment and the current account in a CGE framework allows him to estimate future risks or past losses due to trade liberalisation. As soon as the current account and employment are endogenised, trade liberalisation induces macroeconomic volatility – with mostly negative effects for developing regions.

Polaski (2006) introduces unemployment and separates agricultural labour markets from urban unskilled labour markets in an otherwise “standard” CGE model to conclude that: (1) global gains from further trade liberalisation will be very modest; (2) in sharp contrast to the World Bank’s full employment models, developing countries’ gains come overwhelmingly from market access for manufactured exports; and (3) the largest gains will accrue to countries such as China, while the poorest (in sub-Saharan Africa) will be net losers.

‘Aid for Trade’ Rationale

Very little of the trade liberalisation literature discusses its dynamic and longer term consequences. However, recent efforts to persuade developing countries to accept further trade liberalisation have led to new inducements in the form of “aid for trade” (A4T). Trade liberalisation advocate Jagdish Bhagwati (2005) has made three revealing arguments in support of A4T:

  • to compensate governments for their loss of tariff revenue, which can account for up to half of total tax revenue in some of the poorest countries.
  • to compensate producers, workers and others for the loss of uncompetitive production capacities in agriculture, industry and even services.
  • to develop new internationally competitive productive and export capacities and capabilities.
  • Such recognition of the need to compensate for the loss of revenue and economic capacities as well as the high and uncertain costs of developing new competitive economic capacities and capabilities underscores the limitations of the simplistic comparative static arguments associated with traditional trade theory.

    Standard models used to estimate the distribution of benefits from further trade liberalisation are flawed in important ways, and ignore the risks of labour displacement, economic downturn and increasing debt in the developing world. Unfortunately, the reality is worse. Widely recognised power imbalances in the WTO also undermine developing countries’ potential “gains from trade”. Many of the poorest countries are already expected to be net losers following further trade liberalisation, and clearly, much more than “aid for trade” will be needed to ensure that the Doha round is truly develop mental.


    1 Any CGE model is essentially a system of equations, each describing the “behaviour” of firms, households, governments and so on, and LINK-AGE happens to be a particularly large one with more than 40,000 equations. The effects of trade liberalisation are estimated by removing tariffs and subsidies, which enter the price equations effecting demand decisions. The behavioural functions are based on theory, and all too often – most certainly in the case of LINK-AGE CGE models – rely on dubious simplifications, false assumptions, incomplete information and misleading comparative statics. Stiglitz and Charlton (2005) have emphasised that transition, implementation and adjustment costs are often ignored.

    2 Ackerman (2005) as well as Taylor and von Arnim (2006) have criticised the theory and methodology underlying these estimates. For example, the Armington trade specification overemphasises the potential benefits of liberalisation, and introduces unknown, but possibly huge biases. The Bank model’s detailed disaggregation and very complex architecture obscure what the model is actually doing, distracting policymakers as well as the informed public from its fundamentally flawed assumptions – among them full employment, flexible exchange rates and balanced trade. The Bank model assumes trade elasticities greater than those supported by econometric evidence, thus exaggerating welfare gains and downplaying adverse terms of trade trends. Furthermore, the Bank model assumes uniformity in tastes and equal access to resources, postulating that all governments will be able to replace lost tariff revenue by taxing households, and thus, keep its budget in balance.


    Ackerman, Frank and Alejandro Nadal (2004): The Flawed Foundations of General Equilibrium: Critical Essays on Economic Theory (London: Routledge).

    Ackerman, Frank (2005): “The Shrinking Gains from Trade: A Critical Assessment of Doha Round Projection”, GDAE Working Paper No 05-01, Global Development and Environment Institute (Medford, MA: Tufts University).

    Anderson, Kym and W Martin, eds. (2005): Agricultural Trade Reform and the Doha Development Agenda (Washington: World Bank).

    Bhagwati, Jagdish (2005): “World Bank Funds Should Be Used to Lubricate Trade Liberalisation by Developing Countries”, Financial Times, 26 September.

    Bouet, A, J C Bureau, Y Decreux and S Jean (2004):

    Multilateral Agricultural Trade Liberalisation: The Contrasting Fortunes of Developing Countries in the Doha Round (Paris: CEPII).

    Chang, Ha-Joon (2007): Bad Samaritans: The Myth of Free Trade and the Secret History of Capitalism

    (New York: Random House). Kraev, Egor (2005): Estimating GDP Effects of Trade Liberalisation on Developing Countries (London: Christian Aid).

    Ocampo, J A, K S Jomo and S Khan, eds. (2006): Policy Matters: Economic and Social Policies to Sustain Equitable Development (London: Zed Books).

    Polaski, Sandra (2006): “Winners and Losers: Impact of the Doha Round on Developing Countries” (Washington: Carnegie Endowment for International Peace).

    Reinert, Erik (2007): How Rich Countries Got Rich and Why Poor Countries Stay Poor (London: Constable & Robinson).

    Stiglitz, Joseph E and Andrew Charlton (2005): Fair Trade for All (New York: Oxford University Press).

    Taylor, Lance and Rudiger von Arnim (2006): Modelling the Impact of Trade Liberalisation: A Critique of Com- putable General Equilibrium Models

    (Oxford: Oxfam).

    – (2007): “World Bank Trade Models and the Doha Debate” in Philippe Arestis and G Zezza eds. Advances in Monetary Policy and Macroeconomics

    (Hounmills: Palgrave Macmillan), pp 257-79.

    november 29, 2008 Economic & Political Weekly


    Dear reader,

    To continue reading, become a subscriber.

    Explore our attractive subscription offers.

    Click here


    (-) Hide

    EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

    Back to Top