In the seventies, things were clear. Aid and trade represented two different worlds, aimed at different constituencies, governed by different bodies, and of the same order of magnitude. In the new century, the picture is blurred. Trade now has the lion's share of exchange flows between the North and the South. In 2004, low and middle-income countries exported $1,985 billion to high-income countries; while they received only $84 billion in the form of aid (net official assistance and official aid). For these countries, the amount of aid was even smaller than foreign direct investment inflows or workers' remittances.
Trade is now increasingly seen by many as a necessary, if not sufficient factor of growth. On the other hand, aid has been criticised, as too many loans and grants seemed to have been given endlessly and repeatedly to the same countries without noticeable achievements. Donors suffered from an “aid fatigue” and reconsidered their modes of intervention in developing countries. They asked for more efficiency in aid and for “policy coherence”. They wanted to design trade policies that would make sense from a development perspective. The Doha Round that started in 2001 was based on that premise. For developing countries that could not benefit from a multilateral liberalisation based on reciprocal market access, various preferential schemes were put in place, which gave them full market access (at zero tariff rates) to the North. The European Union launched the Everything But Arms initiative; the United-States implemented the African Growth Opportunity Act. Of course, there were some limitations in product coverage, or through the application of rules of origin. But, all in all, these preferential market accesses were a blessing for some of the least-developed countries. Moreover, trade-related adjustment costs and their social implications were running high on the political agenda, not only in the South but also in the North. Domestic issues, such as competition policy, public subsidies or labour laws were increasingly mixing with international issues.
As a consequence, the Yalta agreement between trade and development communities broke up. The convergence between them resulted in the rising concept of “Aid for trade”. What developing countries needed was not only technical assistance to trade, but more generally, aid designed at alleviating the social cost of trade liberalisation and reducing transactions costs of various kinds, many of them “behind the border”. With such an extensive definition, aid for trade encompasses not only trade policy and regulations and trade promotion schemes, but also assistance to (domestic) economic infrastructure. In 2005, aid for trade, under this definition, represented $26 billion and 36.5% of sector-allocable official development assistance (that is, excluding humanitarian aid and food aid). Also, aid for trade is fragmented between a myriad of bilateral and multilateral donors despite efforts at coordinating the whole process through initiatives such as the Integrated Framework. The latter is targeted at least-developing countries and coordinates six multilateral institutions with (in principle) strong involvement of local constituencies.
What does recent research say about the intricate relationship between aid and trade and the prospects of aid for trade? Is it better to give more aid or more market access to developing countries? The traditional argument would go this way: direct instruments are always better. In that respect, aid should be preferred to market access as a development tool.
However, this line of argument does not take into account side effects of aid on relative prices. For instance, aid that would finance non-tradable goods such as health or education expenditures could undermine the developing country competitiveness, by a change in the relative prices of tradables versus non tradables, that is, the internal terms of trade of the developing country. Aid could also affect the external terms of trade, between the developing country and the donor country. For instance, the fact that aid could be “immiserising” for the developing country, by reducing its terms of trade relative to the rich country, is well-known and described as the “transfer paradox” in the development literature. Interestingly however, the transfer paradox is demonstrated to arise only in specific trade policy settings and in particular, when price-related policies (like tariffs) are present before the transfer. If this is not the case, as for instance when the developing country is imposing quantitative restrictions on imports, the transfer paradox does not occur. Indeed, in such a case, the change in the terms of trade is shut down by the quota. On the contrary, if the developing country is taxing imports through tariffs, the probability that aid could be immiserising is greater, because the transfer can exacerbate the already distorted terms of trade. In such a case, imports would suffer twice, from the tariff and from the deterioration of the external terms of trade. This kind of reasoning shows immediately how the trade-policy space and the aid-policy space (ie. transfers) can interact with respect to growth and development matters. The link between trade and aid policies is empirically illustrated by a recent study by Rajan and Subramanian that shows that in countries that receive more aid, labour-intensive and export sectors grow more slowly than capital-intensive and non-exportable sectors. This evidence is suggestive of the aforementioned relative price effects and that foreign aid may indeed reduce the competitiveness of developing countries.
What should be done in front of such potential interactions? A way out from the risk of the Transfer Paradox can be to make aid conditional to tariff liberalisation in the developing country. The developing country’s trade policy and aid are then regarded as strategic complements. Another solution can also be to avoid the terms of trade deterioration by ensuring that the content of foreign aid includes enough tradable goods. In that respect, Aid for trade seems to be a good candidate, because technical assistance is often import-intensive.
However, a direct evaluation of Aid for trade is not yet available and many pitfalls must be overcome. First, as in the general literature on openness and growth, it is difficult to empirically distinguish between policies and outcomes (that is trade flows). Most empirical studies are looking at flows, while the theoretical discussion concerns both flows and policies. Second, it is difficult to assess the “pure” effect of aid for trade, because of possible endogeneity effects between the two terms. The causality can go from trade to aid: the donor government might listen to private interests and prefer to allocate aid to commercial partners. Alternatively, a common geo-strategic factor can induce the tightening of both aid and trade linkages between two countries. The political economy linkage is obviously significant, but not in a uniform way.
Empirically, the direction of causality between aid and trade flows depends as well on the (donor, recipient) pair of countries. When the causality goes from aid to trade, a study by Lloyd, Morrissey and Osei (2001) finds that aid first reduces trade flows and increases them only after two years. When the causality goes the other way round, from trade to aid, donors are shown to give more aid to recipients that buy proportionately more from them, reflecting a strategic motive in aid allocation.
Despite these pitfalls, it seems that aid for trade has a positive impact on trade flows. However, the magnitude of the impact depends on the degree of market access in the north. Simulations show that a reduction in trade-transactions costs would benefit South Asia more than Latin America or the Middle East North African region, and far more than Sub-Saharan Africa.
Aid for trade is not the miracle and one-size-fits-all solution that would reconcile development and globalisation. Our discussion suggests that on the contrary, economists and policy-makers should scrutinise the specific design of each measure, assess its potential impact on the terms-of-trade and consider the specific (donor, recipient) country pair in the case of bilateral relations.
This article draws on A.Suwa-Eisenmann and T.Verdier, Aid and Trade, forthcoming in the Oxford Review of Economic Policy, published in a previous version as CEPR discussion paper 6465.
Bhagwati J., Brecher R. and Hatta (1985), The Generalized Theory of Transfers and Welfare: Exogenous (Policy-imposed) and Endogenous (Transfer-induced) Distortions. Quarterly Journal of Economics 3, 697-714.
Lahiri S. and Raimondos-Moller P.(1995), The Welfare Effect ;of Aid under Quantitative Trade Restrictions, Journal of International Economics 39, 297-315.
Hoekman B. (2007), Aid for Trade : helping developing countries benefit from trade opportunities, in D. Njinkeu and H.Cameron (eds), Aid for Trade and Development, Cambridge, Cambridge University Press.
J.S. Wilson, C.Mann and T.Otsuki, (2004), 'Assessing the Potential Benefit of Trade Facilitation : a Global Perspective', World Bank Policy Research Working Paper 3224, World Bank : Washington, D.C.