Economics: Not everyone stands to get fat on Gatt

THERE is a blithe assumption among policy makers that everybody benefits from free trade. After all, the elimination of trade barriers merely removes an obstacle in the way of business deals, and most economists assume that people will not buy or sell unless they perceive it to be in their own interest. Since more trade means more deals, it also means more benefits.

However, this is not necessarily so. An interesting study by the Organisation for Economic Co-operation and Development and the World Bank finds that the partial liberalisation proposed in the General Agreement on Tariffs and Trade (Gatt) talks would add some dollars 213bn ( pounds 142bn) to world income in 2002 and every year thereafter, measured in 1992 dollars. The members of the OECD - the developed countries - gain nearly two thirds of this total.

Of course, individuals and businesses can lose from new competition. But it is less widely recognised that whole countries and regions can also be losers. Several regions of the world, notably sub-Saharan Africa, north Africa and the Mediterranean countries, lose a total of about dollars 7bn a year, according to the OECD and the World Bank. This may not seem a lot, but incomes in black Africa average just dollars 280 a head per year. As the first column of the table shows, a successful Gatt round would cut real income in Indonesia, sub-Saharan Africa; the Maghreb and the Mediterranean countries.

By contrast, the world's big gainer from a successful Gatt round is the European Community, whose income rises more than most because it has more protection to dismantle. France does well, despite the protestations of its government. The United States, the aggrieved party in Europe's attempts to unpick the Blair House farm agreement, would gain a mere 0.2 per cent. The biggest gainer of all, proportionate to its income, is China.

What is going on? The first conclusion of the study is that the real gains overwhelmingly come from reducing the protection and subsidisation of agriculture. Out of the overall dollars 213bn gains, about dollars 190bn arises from liberalising farm trade alone.

This reflects both the highly distorted nature of food markets, and some lacunae in the study. It did not assess the gains from freeing trade in services, because of the difficulty in measuring barriers. And it did not take account of non-tariff barriers to manufacturing trade (such as voluntary export restrictions of Japanese cars, or consumer standards that only home producers meet).

These omissions are for the sound reasons of uncertainty. But they mean that the study is only partial. The gains from trade liberalisation may be greater than the authors estimate. The study takes account of tariff cuts, but average industrial tariffs are now a mere 6.4 per cent, down by a half from their levels at the end of the Tokyo round of trade talks in 1979. But the Gatt round will also chip away at the non-tariff barriers, which cover 18 per cent of manufactured imports and 38 per cent of farm imports.

The losses for some of the world's poorest countries are explained by the dynamics of farm trade. As the EC and others reduce their farm subsidies, they will also have to reduce the dumping of surpluses on world markets. That will in turn firm up the world market price for many commodities: the study estimates that wheat prices will rise by 3.5 per cent, sugar by 8 per cent and so forth. This will boost production in low-cost areas in the developing world but also lead to losses of income for areas that import more food than they export.

This pattern of big overall gains combined with some losses for selected groups of countries also applies if there is a full liberalisation of trade ('full free trade' in the second column of the table), going well beyond what is proposed in the Uruguay Gatt round. The price changes in agriculture are more dramatic: for example, as subsidies in Japan and the EC come down, world wheat prices rise by 17 per cent.

The authors estimate that total gains would then be dollars 450bn, of which the OECD countries gain dollars 290bn. The aggregate losses are also larger, at dollars 40bn. Nigeria loses 1.8 per cent of its real income, and the rest of sub- Saharan Africa loses 0.9 per cent. The Maghreb and the Mediterranean lose more than 2 per cent, while the European Community gains 3 per cent.

Let me cite one more interesting result of the study, before discussing some of the implications. These computer results, like most work on trade, assume flexible labour markets. In other words, if someone becomes unemployed because of new competition from overseas producers as barriers come down, the pay rates in that area fall, and employers soon raise their demand for labour again. The unemployed are quickly absorbed back into some gainful occupation.

This may hold true in the long run, but it is clearly over-optimistic in the short run. So the authors also experiment with a computer model that assumes that labour markets are inflexible: real wages are relatively rigid, so that the impact of liberalisation is to create some transitional unemployment as employers cut labour costs by means of redundancies rather than a pay squeeze.

Ironically, the overall output gains from liberalisation are even greater on this basis than they are with flexible labour markets. The reluctance of employees to accept pay cuts means that real incomes are higher, and so is their spending. Those in work benefit more from the fall in farm prices.

The gains for the developed world from the Gatt round become even greater, with EC income up 3.6 per cent. Black Africa no longer loses, but Brazil, Mexico and Canada join the Maghreb and the Mediterranean countries on the list of losers. For example, Brazil loses because reducing its high protection for manufacturers leads to high unemployment.

These contradictory results, depending on whether labour markets are assumed to be flexible or not, do not make policy analysis any easier. The authors prefer the conventional presentation in which pay is flexible. This is reasonable, since trade liberalisation takes a long time to have its full effects, during which time the labour market is likely to adjust.

On this basis, policy makers should compensate sub-Saharan Africa's losers. After all, there are potentially large gains to the developed world from liberalisation: a small increase in aid flows would offset any putative Third World losses and would represent a fraction of the total gains from the Gatt round. Sadly, Africa cannot expect to benefit much from the opening of Western markets. The study finds that Africa would benefit more from bigger transfers than higher OECD growth.

Given the sharp cut in European farm incomes implied by reductions in support prices towards market levels, it is not surprising that French farmers are up in arms. But the study also shows that there are large potential income gains within the developed world that could and should be used to compensate farmers who lose.

The Common Agricultural Policy is bound eventually to crack, simply because the trend of farm productivity is ever upwards, pushing the EC into larger and larger surpluses unless there are sharp price cuts or ever more stringent production quotas. The OECD-World Bank study projects that the Gatt cut in EC farm subsidies will merely hold the crop surplus to 13 per cent of consumption, compared with 15 per cent anyway. And the livestock surplus will be 6 per cent instead of 12 per cent of consumption. The CAP's mountains need more radical cures.

* Trade Liberalisation: global economic implications by Ian Goldin, Odin Knudsen and Dominique van der Mensbrugghe, OECD and World Bank (from HMSO).