"Inequality is soaring through the globalisation period - within countries and across countries. That's expected to continue." Noam Chomsky, 2001

May Day protesters against "neo-liberal" or "corporate globalisation" know that Noam Chomsky is right: globalisation has not only exacerbated inequality but worsened world poverty. But this is one of those things "people know that just ain't so". They are wrong in both their conclusions and their analysis.

Before looking at the evidence, we need first to ask what matters more, inequality or poverty? The answer for the great majority of people must be the latter. Many people would prefer more equality to less, but few would prefer a world of uniform poverty to one of unequal wealth, provided the latter brought a reduction in destitution in its train. Fortunately, the evidence that the world has been achieving a big fall in the proportion of destitute people and a significant fall in the absolute numbers of the destitute is strong. Indeed, with the recent acceleration in the rates of growth of large poor countries - above all, China, and, to a lesser extent, India - both global inequality and poverty have been falling.

Suggestive - but heroic - work by François Bourguignon, the World Bank's current chief economist, and his collaborator, Christian Morrison, traces the history of inequality and extreme poverty - defined as consumption of $1 a day at purchasing power parity, in 1985 prices - since 1820*. They show that all the increase in global inequality, prior to 1980, was the result of the growing differences in average incomes across country groups. That is why growth matters. They also suggest that, as development has spread, there has been a steady fall in the proportion of the world's population in extreme poverty, although also a rise in the absolute number up to 1980 (see chart).

Now turn to the period since 1980 - the age vilified for its rush into globalisation. Both global inequality and the proportion of the world's population and number of the world's people in extreme poverty have fallen.

Start with poverty. The latest World Development Indicators from the World Bank, out last month, suggest that the numbers in extreme poverty, defined by the Bank's poverty line of $1 a day, at 1993 purchasing power parity, fell from 1.451bn in 1981 to 1.219bn in 1990 and 1.101bn in 2001. As a proportion of the population of developing countries, the decline was from 39.5 per cent in 1981 to 27.9 per cent in 1990 and 21.3 per cent in 2001.

Surjit Bhalla, an Indian economist formerly at the World Bank, has estimated the number on a roughly comparable basis at 1.581bn in 1980, 1.208bn in 1990 and 899m in 2000**. Mr Bhalla argues that, by 2000, the proportion of the developing country population in absolute poverty was just 18.2 per cent, down from 46.5 per cent in 1980 and 29 per cent in 1990. This is true even when he uses the lower estimates for consumption levels given by household expenditure surveys, rather than the national accounts. On the latter basis, he finds that the incidence of extreme poverty was as low as 13.1 per cent of the developing world by 2000 (see chart).

There are sizeable methodological differences between these researchers, the most important being whether consumption and income averages are taken from the national accounts or sample surveys of household expenditure. But, as Martin Ravallion of the World Bank remarked in The Economist on April 7, "a similar trend of long-term reduction in poverty does emerge" from the different estimates.

Mr Bhalla and Xavier Sala-i-Martin of Columbia University, whose conclusions on poverty are similar to Mr Bhalla's, also argue that there has been a sizeable reduction in global inequality among households. Indeed, the former argues that, by 2000, global household inequality of consumption was back to levels last reached in the 1950s and far below levels at its peak, which occurred in the mid-1970s (see chart). Prof Sala-i-Martin's analysis goes back only to 1970. But he also shows sizeable reductions in inequality since then.

The cause of these reductions in poverty and global inequality are one and the same: the fast growth of Asian countries, above all China. The World Bank's poverty figures bring this point out clearly. Between 1981 and 1999, the proportion of people in the east Asia and Pacific region living on less than $1 a day fell from 56 per cent to 16 per cent. In China, it fell from 61 to 17 per cent. In South Asia, too, the fall was rapid, from 52 to 31 per cent. But, in sub-Saharan Africa, where growth failed, the share rose from 42 to 47 per cent.

If you care about global poverty and, for that matter, about equality, your aim should be to raise the growth rates of poor countries. Eliminating poverty is now a challenge, rather than the unchangeable reality of human existence it used to be - because we know, in principle, how to solve it. "It's the growth", stupid.

Where then does that pantomime villain, globalisation, fit in? The answer is that successful countries have all exploited global market opportunities, predominantly international trade and, to a more variable extent, foreign direct investment, to accelerate their growth. China is now the world's premier example of this strategy. But it is following the example of other countries in its dynamic region.

Successful globalisation has, in short, reduced both poverty and inequality. Opening up to trade is far from a sufficient condition for rapid growth. But, in the right conditions, it has proved an enormous help. In the post-second-world-war era, it is impossible to think of a consistently successful economy that has not put international economic integration at the core of its development strategy. The elimination of destitution, the millennial curse of humanity, is in sight, provided we find a way to spread the benefits of market-led development elsewhere - to sub-Saharan Africa, above all. Now that really is a cause worth demonstrating for.

* Inequality among World Citizens, American Economic Review, September 2002;

** Imagine There's No Country (Washington DC: Institute for International Economics, 2002); The World Distribution of Income (Estimated from Individual Country Distributions), May 2002, http://www.columbia.edu/

martin.wolf@ft.com