Writing in the Mail and Guardian Online, Fredrik Erixon and Razeen Sally argue that “Oxfam’s one-sided trade liberalisation is a policy of self-harm for developing countries in the WTO”. The evidence they put forward is as follows:
According to World Bank and Organisation for Economic Cooperation and Development (OECD) figures, since 1980 developing countries with a total population of about three billion — mostly in Asia — have more than doubled their trade-to-gross domestic product (GDP) ratios, doubled real per capita incomes and have cut average import tariffs by more than one-third. That leaves “less-globalised” developing countries with a combined population of about 1,5-billion, which have stagnant trade-to-GDP ratios and per capita incomes and much lower cuts in average import tariffs. The “new globalisers”, unlike the rest, have also seen dramatic reductions in poverty and improvements in human welfare indicators.
This is rather weak stuff, since it describes a situation - well-off countries with low tariffs, and poorer countries with higher tariffs - which could just as easily have come about through growth causing tariff reduction rather than tariff reduction causing growth. The inability of Erixon and Sally to provide less ambiguous evidence doesn’t strengthen their case.
In fact, there seems to be good evidence that the relationship between trade barriers and growth is far less clear-cut than they claim and may even work in the opposite direction. Halit Yanikkaya reports here that “trade barriers are positively and, in most specifications, significantly associated with growth, especially for developing countries … consistent with the findings of theoretical growth and development literature”, while DeJong and Rippoll say “the relationship between tariffs and growth is negative and significant among the world’s rich countries, while positive (but typically insignificant, depending upon the particular estimator being employed) among the world’s poor countries”. Lastly, Emma Samman sifts through the data here and concludes:
1) countries that reduced their tariffs appear to have fared no better than those that did not;
2) deeper tariff cuts may be associated with lower growth than more moderate cuts; and
3) more generally, the large diversity of experiences precludes drawing any sweeping conclusions.
That last point is important, but taken together these studies suggest that higher trade barriers are not significantly harmful to poor countries, but that cutting them too quickly might be. This is entirely consistent with Oxfam’s position as described by Erixon and Sally, and suggests that the best thing rich countries can do is drop their own tariffs unilaterally rather than trying to extract similar cuts from poorer countries. So Oxfam are right.
The evidence above also suggests a few more important points:
1) This really should not require repeating after how many decades of development theory, but we cannot go on assuming that what applies to rich countries applies equally to poor ones. Clearly, in terms of trade policy, it does not. Based on the evidence, it is perfectly reasonable to oppose both trade barriers in rich countries and pressuring poor countries to liberalise.
2) The time dimension of economic change is important. Countries with under-developed, undiversified economies and lots of people on low incoms seem to adapt pretty poorly to sudden change of the kind that trade liberalisation can cause. Models using comparative statics, like the World Bank estimates of the gains from free trade referenced by Erixon and Sally, don’t cope with this very well.
3) Maybe trade policy isn’t all that important. It doesn’t always show up as significant in growth regressions, and when it does the effect is rarely large. One of the few studies to find really serious effects, Dollar and Kraay’s “Trade, Growth and Poverty”, was also one of the shoddiest, as demonstrated by Samman, Dani Rodrik, and these guys. But at least Dollar and Kraay’s work inspired this excellent paper by Birdsall and Hamoudi, which establishes that the apparent decline of ‘non-globalising’ countries is mostly accounted for by the collapse in export revenues of commodity-dependent countries in the early 1980s. So much current research into international development seems to treat state policy as all-powerful (especially when it’s implicitly criticising the exercise of that power), but it seems to me that governments fundamentally don’t control the shifting patterns of world trade.