Oliver Blanchard went to a IMF conference on capital flows, and wrote up some thoughts. It's well worth reading the whole (short) post, if only to see where the IMF's thinking on this is right now, but I want to highlight a few bits.
First, while the issue of capital controls is fraught with ideological overtones, it is fundamentally a technical one, indeed a highly technical one. Put simply, governments have five tools to adjust to capital flows: monetary policy, fiscal policy, foreign exchange intervention, prudential tools, and capital controls. The challenge is to find, for each case, the right combination.
Regular readers will guess that I disagree with this completely. The issue of capital controls is not just "fraught with ideological overtones", it is also fraught with distributionary consequences. If governments choose to restrict capital flows using one of the five tools Blanchard lists, then they will be benefitting some groups over others. Which of the five they employ will also involve winners and losers. From this perspective there is no "right" combination, only choices that advantage some members of (domestic and global) society and disadvantage others. The challenge for political leaders is to find the combination that will allow them to remain in office. The challenge for interest groups is to push for the combinations that will benefit them. But this is not a technocratic problem, or at least not just one.
Blanchard gets close to understanding this a bit later, when he writes:
The nature of specific investors must inform the policy choices. We often think of inflows and outflows as coming from primarily from decisions by foreign investors. The reality is that many of these inflows and outflows often come from decisions by domestic investors. When this is the case, targeting nonresidents is largely misguided.
Layna Mosley, one of my professors, has done a lot of very good work examining how, when, and to what extent international investors place pressures on domestic governments*. There is less work (that I know of) that seeks to explain how, when, and to what extent domestic political actors (including investors) pressure their governments for certain types of policies related to capital flows. But surely this is a political question that requires a political answer. To the extent that the IMF isn't thinking about those issues they are probably missing the boat.
It is not clear that the diversity of approaches we observe in practice comes from different circumstances, or from suboptimal responses. It was interesting to observe for example that Chile relies on foreign exchange intervention, not on capital controls, but India, instead, relies on capital controls, not on foreign exchange intervention. Are these corner solutions really optimal?
Again... what is meant by optimal? Different policies will benefit different actors. In many of the cases under discussion there is no reason to think that we're on the Pareto frontier, but even if we are the actual policy choices reflect distributional concerns. Instead of trying to figure out whether these policy choices are optimal, we should be thinking of them in terms of bargaining theory. And because these policies involve international as well as domestic actors, we need to complicate the model to include multiple levels of analysis. Blanchard seems to realize this towards his conclusion.
There were some issues that I would like to have seen explored more fully.
One was the multilateral angle. As my IMF colleague Min Zhu said in his opening remarks, “ensuring that countries reap the full benefits of capital flows is a shared responsibility between advanced and emerging market economies, between surplus and deficit countries, between capital-exporters and capital-importers.” The challenge is to translate this into practice. What is the actual responsibility of source countries? Should they take it into account in conducting monetary policy, and if so, how? Should we worry about the “beggar thy neighbor” effect of controls? Some of the evidence presented at the conference suggested that these spillovers across recipient countries were not very large. Theoretical and further empirical work is badly needed here.
These are good questions, and they do need more work. But they are political questions, so economists are not very well suited to answer them. I understand that Blanchard's position within the IMF means that he has to focus on the technocratic rather than the political, but if he wishes the IMF to be an effective institutions in the future he should at least be thinking of the political implications of capital flows. This has been the IMF's weakness for decades; it's high time for them to get better on this score.
*I discussed this work briefly here, but interested folks are encouraged to read the actual research (linked in that post).