There is little empirical basis for the comforting consensus among economists that globalisation did not cause the rise in wage inequality and unskilled unemployment in developed countries over the past two decades. That is the contention of Professor Adrian Wood of Sussex University's Institute for Development Studies, writing in the September issue of the Economic Journal. Hence, policy-makers battling against protectionism should not deny that trade has adverse effects on inequality. Instead, they should explain that education, redistribution and job creation provide better and cheaper ways of tackling the problems that trade causes for unskilled workers.
The misleading consensus, which identifies new technology as the main cause of labour market inequalities, allegedly rests on a large body of research. But according to Wood, on balance the evidence gives more support to the globalisation view. What''s more, the consensus has arisen partly from an error of logic. There is strong evidence that technical change has caused most of the fall in demand for unskilled workers over the past two decades. But this downward trend in demand has been happening for many decades, and was formerly more or less matched by the long-term downward trend in the supply of unskilled workers caused by the expansion of education.
The crucial question, which most economists have failed to ask, is why the fall in demand for unskilled workers suddenly accelerated in the last two decades. And despite the diffusion of computers, there is little hard evidence that the speed of technical change, or its bias against unskilled workers, has increased in this period.
A more compelling case can be made that the acceleration of the downward trend in demand for unskilled workers was caused mainly by changes in the pattern of world trade, which involved the relocation of most unskilled manufacturing activities from developed to developing countries. The decline in the share of manufacturing in total employment in developed countries accelerated from about 1970, and most of the lost manufacturing jobs were unskilled. This deindustrialisation is clearly correlated – both over time and across developed countries – with the rise in imports of manufactures from developing countries.
Critics of the globalisation view argue that its prediction of a fall in the relative prices of labourintensive goods is not consistent with the evidence since the 1970s. During this period, however, the process of production of many goods has been split up, with the labour-intensive stages being ''outsourced'' to developing countries and the skill-intensive stages kept in developed countries. More relevant than movements in the relative prices of goods produced in different sectors is thus what has happened within each sector to the price of imports, relative to the price of domestic output. It has clearly and consistently fallen as the globalisation explanation would predict. There is broad agreement on policy responses between the economists who attribute the rise in labour market inequalities to globalisation and those who attribute it to new technology. All favour better education and training, redistribution of income, and using taxes and public spending to raise the demand for unskilled labour. Almost all oppose restrictions on trade, and the risk of encouraging protectionism has led most economists to err on the side of understating the impact of globalisation on inequality.
However, in Wood's view, it is more effective to meet protectionists on their own ground, to concede that globalisation has increased inequality, and to show that other policies provide a better solution than protection. Northern governments have begun to implement these other policies, with the protectionists inadvertently playing a socially useful role by counterbalancing the reluctance of the skilled gainers from globalisation to pay the taxes needed to finance assistance to the unskilled losers.
''Globalisation and the Rise in Labour Market Inequalities'' by Adrian Wood is published in the September 1998 issue of the Economic Journal. Wood is Professor of Economics at the Institute of Development Studies, University of Sussex, Brighton BN1 9RE. The writing of his paper was funded by the Economic and Social Research Council (ESRC).