SANTIAGO – Few policy debates are stranger than the one concerning capital controls. Mention the issue to a banker or a mainstream economist and you are likely to get a vehement reply: capital controls do not work, because speculators can evade them at little or no cost, but countries should never adopt such controls, because doing so is very costly. Am I the only one who finds this logic a bit crooked?
The next stage of the conversation is usually just as strange. When dealing with surges of potentially destabilizing capital inflows, capital controls are a no-no, but something called prudential regulation is quite okay. Capital controls, you are likely to be told, discriminate between transactions depending on the country of residence of the parties involved, and that is bad. Prudential regulation discriminates on the basis of the transaction’s currency of denomination or maturity, and that is good.
If this conversation is taking place at a cocktail party, at this point you would be well advised to ask for another drink.
That has been the state of the debate – until recently. The big intellectual change has come from the International Monetary Fund, once an outright enemy of capital controls. In a series of studies, IMF economists have stopped treating the issue as one of good versus evil, and have started treating it as a run-of-the-mill economic choice: capital controls have costs and benefits that can be measured and compared. The Fund now argues that they should be one more tool for prudent policymakers to use as circumstances require.
Another big change has come from fast-growing emerging-market countries, many of which, facing a wall of money coming their way from slow-growth advanced countries (and mindful of the costs that large and volatile capital flows can bring), have chosen to apply capital controls. Brazil, Indonesia, South Korea, Thailand, and Taiwan have all recently introduced disincentives to capital inflows. Other emerging countries, like Peru, have tightened domestic prudential regulations with the same aim in mind.
Those who claim that capital controls are ineffective seldom specifyat what they are presumed to be ineffective. A recent paper by Marcos Chamon, Atish Ghosh, Jonathan Ostry, and Mahvash Qureshi, economists at the IMF´s Research Department, suggests that capital controls can be effective indeed. The authors conclude that, by reducing the share of short-term foreign-currency debt in a country’s total liabilities, capital controls can reduce vulnerability to financial crises.
The IMF economists ask how 50 countries fared during the recent world financial meltdown, with performance measured by economic growth in 2008-09 relative to 2003–2007. Their main finding is that countries with capital controls in place beforehand fared better – their growth rates fell less during the crisis. This was because they had less leverage to be unwound and less footloose capital that could leave in the event of a global financial squeeze.
This does not mean that capital controls can solve all problems: the same paper shows that they are relatively ineffective at preventing lending booms – another key cause of economic vulnerability. This should not be surprising, since many factors besides international debt flows can cause domestic banks to throw caution to the wind and start lending recklessly.
But the paper also shows that domestic prudential policies – such as capping home-mortgage loans at a certain percentage of the property’s value, or increasing banks’ capital requirements during economic upswings – can be effective at restraining lending booms. This suggests that capital controls and prudential policies can complement each other, contrary to what conventional wisdom often assumes.
Nor do these findings suggest that capital controls have no costs. For example, as the MIT economist Kristin Forbes has documented, they increase the cost of financing for small and medium-size companies.
All aspiring economists are taught the discipline’s cardinal rule: governments (and people, for that matter) should undertake an action to the point where, at the margin, its benefits equal its costs. Recent research, by carefully quantifying the potential costs and benefits, allows this standard logic to be applied to the question of whether and when to apply capital controls. For an economist, this is cause for celebration – no cocktail required.