Globalisation until now has meant the free movement of international private capital but not of the other things like labour and technology. Indeed, if anything, developed countries have sought to slow down the flows of the latter by having stricter immigration and intellectual property rules. The theory of comparative advantage is useful only selectively.

Their economists, however, have tended to focus almost entirely on the benefits that accrue to the developing countries when they allow the free flow of (mostly) western capital. Some of them have famously claimed that capital has no nationality but only owners, which is sad but true.

Developing countries, on the other hand, have been chary of removing controls on the free flow of capital into their economies. After all, they have seen only too well what happens when these nation-less owners suddenly decide to take their money out.

The presence of large amounts of foreign capital also makes economically weak countries vulnerable to political pressure by the countries to which the owners of foreign capital belong. My recent exertions on behalf of the Reserve Bank of India have demonstrated to me how severe these can be.

Economists, bless their na�ve little souls, don't recognise this problem or, even if they do, choose to ignore it. So what we get is papers like the one reviewed here: rich in data, econometrics and conclusions but somewhat lacking in insight.

Sebastian Edwards, who has made a name for himself, in his recent paper* asks whether "restrictions on capital mobility reduce vulnerability to external shocks." That is if a country makes it less easy for capital to flow in and out, does it reduce the risks to itself of sudden contractions?

The answer is no, not really. Edwards finds that "the marginal effect of higher capital mobility on the probability of a capital flow contraction (CFC) is positive and statistically significant, but very small� The point estimate of the marginal effects of capital mobility on the probability of a CFC is approximately 0.1 per cent and very stable. . ."

In plain language this means that there is a risk but it is very small. And the way to neutralise that risk is to have a flexible exchange rate. Flexible exchange rates, says Edwards, "greatly reduce the probability of experiencing a capital flow contraction." Not just that. "The benefits of flexible rates increase as the degree of capital mobility increases."

But what if a benighted country suffers from severe short term domestic supply constraints and, to keep inflation low, runs up a high current account deficit? Ah, says Edwards, "a higher current account deficit increases the probability of a capital flow contraction, while a higher ratio of FDI to GDP reduces that probability."

So the answer to a high current account deficit is the classic textbook one: make it up on the capital account but make sure the money comes in as FDI and not as quick bets on the stock market. Countries which take that route because they don't create conditions conducive to higher FDI flows are playing with fire. I wonder why India keeps coming to mind.

The chief attraction of Edwards's paper lies in the different problem he has tackled. Till now, economists have tended to focus on sudden stops -- Mexico 1994, East Asia 1997, Russia 1998, Brazil 1999, Turkey 2001, Argentina 2002, and so on. India 1989 somehow has not attracted much attention.

Economists have not, however, examined the more pervasive problem of steadily increasing bleeding that happens more widely and frequently. Edwards also takes a broader view of capital mobility. This makes the analysis even more useful.

*Capital Controls, Capital Flow Contractions, and Macroeconomic Vulnerability, NBER Working Paper No. 12852, January 2007.