The pros and cons of prudential capital controls to curb the boom-bust cycle in capital flows have been discussed before (Williamson 2005), but economists now understand better the theoretical case for such policies with a new literature on the welfare economics of prudential capital controls (Korinek 2011). This literature essentially transposes to international capital flows the closed-economy analysis of the macroprudential policies that aim to curb the boom-bust cycle in credit and asset prices. It finds that it is optimal to impose a countercyclical Pigouvian tax on debt inflows in a boom to reduce the risk and severity of a bust. Interestingly, the optimal tax would fall primarily on the flows (short-term or foreign currency debt) that are the least likely to be conducive to economic growth.
The optimal tax has also been quantified in calibrated dynamic welfare optimising models. Models with endogenous and occasionally binding constraints are not tractable and must be simplified in some respects to be solvable, even numerically, but the results may be informative. A nice example of this approach is Bianchi (2011), who finds, in a model calibrated to Argentina, that the optimal tax rate on one-year foreign currency debt increases with the country's indebtedness and fluctuates between 0% and a maximum of 22%.
Obviously, capital controls are not a silver bullet. Like any attempt to regulate the financial sector, they elicit attempts to circumvent or evade and can be used effectively only if they are used with moderation. But this observation applies to all taxes and regulations. Capital controls are not a panacea, but this does not prevent them from being a legitimate instrument in the macroprudential toolbox.