BANNER-WAVING anti-capitalists and advocates of free trade (such as The Economist) seldom agree on much. But one thing readily accepted by both sides in the globalisation debate is that the World Trade Organisation (WTO), the Geneva-based outfit that oversees multilateral trade arrangements, makes a difference. Almost nobody doubts that the WTO and its predecessor, the General Agreement on Tariffs and Trade (GATT), have helped to boost global trade, by encouraging countries to lower their tariff barriers in successive rounds of trade negotiations.
Until now. In a brutally contrarian paper*, Andrew Rose, an economist at the University of California, Berkeley, claims that the WTO has been ineffective. “There is no compelling empirical evidence showing that the GATT/WTO has actually encouraged trade,” he writes. How can this be? Hasn't global merchandise trade risen, on average, by 6% a year since the GATT was created in 1948—far faster than global output?
Yes, says Mr Rose, who is as keen on trade as most economists—but the WTO apparently had nothing to do with it. Being a member of the WTO makes no statistical difference to how much countries trade with each other, after taking into account other things that are widely assumed to influence trade, such as income and distance from foreign markets and suppliers. In a companion paper, he suggests why it might be that the WTO has so little effect: membership seems to have little or no influence on trade policy. Although the WTO is ineffectual, Mr Rose argues, one agreement that does boost trade is the Generalised System of Preferences (GSP): a labyrinth of trade preferences that rich countries unilaterally give to poor ones. The GSP, and preferences in general, are viewed with suspicion by free traders, who see multilateral trade agreements—in which everybody receives the same treatment—as superior. Mr Rose finds that the GSP has doubled trade, whereas the WTO has made no difference.
This is startling stuff. Are Mr Rose's statistical methods wrong? Apparently not. He uses a standard approach to explain trade patterns and he tests his results carefully. Whatever you make of the findings, he has uncovered quite a mystery.
The simplest explanation for Mr Rose's puzzling numbers is that the WTO is indeed useless. But this is a conclusion that Mr Rose himself regards as contrary to common sense. There are other possibilities. The timing of membership, the pacing of trade reform and differing attitudes to membership among particular groups of countries may all be clouding the truth.
As Mr Rose notes, de facto and de jure membership of the system are not the same thing. Countries may sometimes behave as though they are members even when they are not, in anticipation of joining. Conversely, many developing countries first joined the GATT on the understanding that they would be spared its obligations (and benefits). Until recently, the GATT system made few demands of poor countries, which are now the majority of the WTO's 144 members, and granted them second-class status in return. Many poor countries were allowed to keep their trade barriers high. India, a founding member of the GATT, still has tariffs that are among the highest in the world.
Over the past two decades, many poor countries have opened their economies unilaterally (or been pushed into it in exchange for loans from the IMF and World Bank). In contrast, the industrial countries joined the GATT early, but then reduced their merchandise tariffs rather slowly over subsequent decades. Seen in the aggregate, all this may look as though the GATT had little or no effect. Yet it could still be true that the rich countries' long and gradual liberalisation would not have happened without it. And if the rich countries had not liberalised, neither would most of the developing countries.
Even if the quantifiable role that the GATT has played in boosting trade among members is modest, a rules-based multilateral trading system has broader benefits. As Mr Rose points out, the system may be a kind of public good, encouraging trade among all countries, members and non-members alike. Also, the WTO oversees a system of rules and dispute-settlement procedures that goes well beyond tariffs. Besides benefiting members directly, these processes may indirectly benefit non-members, by establishing pro-trade norms and standards of behaviour. Also, the system gives members benefits that go much wider than expanded trade—advantages which Mr Rose is failing to capture. For instance, foreigners are likely to be more confident about investing in poor countries if they are WTO members.
Mr Rose's finding that the GSP is strongly trade-promoting is incidental to his main results—he cites it mainly for methodological purposes, to establish that his approach can detect some kinds of trade expansion. No less than his puzzling findings on the impotence of the GATT, however, his results on the potency of the GSP are both peculiar and troubling. Most previous research on the GSP has underlined the harm it has done to poor countries. One recent paper** by Caglar Ozden, an economist at the World Bank, and Eric Reinhardt of Emory University, argues that the GSP is anti-trade: on balance, the system actually delays a poor country's efforts to liberalise. Even if that were not so, the GSP seems certain to distort trade in a way that worsens its quality, regardless of any influence, good or bad, on quantity.
The Doha round of multilateral trade negotiations is struggling; the pendulum is swinging towards preferential bilateral and regional agreements. Most economists believe this is bad. Mr Rose's arresting results challenge that view. They demand, and will no doubt receive, careful scrutiny.
* “Do we really know that the WTO increases trade?” CEPR Discussion Paper 3538.
** “The Perversity of Preferences” World Bank working paper, forthcoming.
|Copyright © 2002 The Economist Newspaper and The
Economist Group. All rights reserved.|