Introductory statement Speech at the IIF International Capital Markets and Emerging Markets Roundtable

Check against delivery.

Ladies and gentlemen

Ever since the ground-breaking work of Solow in the fifties most policy makers and members of academia – especially in the advanced economies – have inclined to favour free capital flows. Many crises later, the view has somewhat shifted, in particular, regarding emerging markets and developing countries, which is the group of countries that I will focus my remarks on.

Not only can capital flow reversals and in the extreme form sudden stops create significant costs in terms of output losses, financial instability and even political instability. But from all we know today it is surprisingly difficult to find empirical evidence of capital account liberalization being beneficial in terms of output gains, investment, productivity gains, or risk sharing for countries that have not progressed sufficiently far in their institutional development.

Much of the work on this has been carried out by the IMF and indeed by Maurice Obstfeld himself, so Maurice should correct me if I am wrong, but essentially liberalized capital flows should go hand in hand with functioning judicial systems including well enforced property rights, developed and well regulated financial systems as well as stability-oriented macroeconomic policies. There are even longer lists, but these seem to be the minimum requirements.

So, given that these conditions are not always met fully, should countries resort to capital flow measures if faced with volatile flows?

By the way, I am using the IMF terminology of capital flow measures. These are basically capital controls plus those macroprudential measures that aim to limit financial stability risks arising from cross-border capital flows.

I think we all agree that sound domestic policies should always be the first line of defence and that capital flow measures should not substitute for warranted macroeconomic or exchange rate adjustments.

But it is also useful to state an obvious but often overlooked fact: Unlike restrictions on the current account, the imposition of capital flow measures is not under the jurisdiction of the IMF, thus, any country can basically adopt such controls as it deems necessary to regulate capital movements. However, capital flow policies are part of the Funds surveillance and as such open to an assessment of their impact on the International Monetary System.

Objectives of capital flow measures usually are to maintain monetary independence, to mitigate exchange rate pressures and to safeguard financial stability. However, capital controls also come with a cost, like impairments of financial markets and possible reductions of economic growth due to their distortions.

Evidence on the effectiveness of capital flow measures is mixed, or in the words of Olivier Blanchard: "still surprisingly inconclusive". Work on macroprudential measures has only begun in earnest after 2009, thus, the knowledge about these policies is still limited.

Significant results have been found on the composition of capital flows. There is evidence that they can lower banking inflows and portfolio inflows. But there is no significant evidence for a clear impact on net or gross flows. While there is solid evidence that capital flow measures help countries to conduct an independent monetary policy, there seems to be little or no effect of capital flow measures on the exchange rate.

Although the literature on the impact on financial stability is relatively young, capital flow measures have been found to lower credit growth and to curb bank leverage.

Where does that leave policy makers? There is still a lot we do not know. It seems capital flow measures can help a country to deal with volatile capital flows and reduce the likelihood of a financial crisis. But of course one has to consider the costs and distortions.

I think the Institutional View of the IMF has got it mostly right. To safeguard financial stability countries should be able to use all tools at their disposal, depending on which are best suited to the task, but as non-distortionary as possible and not as an excuse to shirk difficult decisions on macroeconomic policy.

Given that capital movements are not comparable to trade in real goods and given the diversity of its membership, I see no value in extending the IMF’s mandate beyond the existing arrangements regarding capital flows. This would necessitate also a change of the IMF’s Articles of Agreement, which could be opening a can of worms.