Undervaluation: a second-best solution for growth? – part II

If an undervaluation strategy can indeed help support growth, how can it be implemented? Indeed, how does such a strategy relate to inflation targeting and to capital account liberalization?

Mr. Sta Ana is proposing a real exchange rate target, i.e. real undervaluation. As we know from the “impossible trinity”, a central bank cannot pursue a (nominal) fixed exchange rate, open capital accounts and an independent monetary policy simultaneously. One solution could be to (partially) diverge from one or both of the latter.

In the discussion on inflation targeting, I agree that a less doctrinaire approach could look at longer-term inflation prospects, which are also dependent on exchange rate stability. A mature monetary policy could also seek to prevent overvaluation, which could lead to sudden depreciations and hence risks to inflation over the longer term. Yet a change of approach should be well thought out, also for the impact on financial stability. In Europe and the US, there is a discussion on taking financial bubbles into account in monetary policy, which may require raising rates or “leaning against the wind” to prevent growing imbalances. But note that this latter strategy would go in the opposite, more “hawkish” direction. If monetary policy seeks to keep down the exchange rate – i.e. lowering interest rates more than justified by inflation expectation, a “dovish” policy – this can lead to negative real interest rates and the emergence of financial bubbles. This lends credence to the so-called “Tinbergen Rule”, that for each policy target there must be one tool, and that conversely, one tool cannot achieve two goals consistently. Look at what is happening in Turkey right now, where the central bank (CBRT) is clinging to a low policy rate to discourage capital inflows, even as inflation rises. Reserve requirements are being used in parallel, but are so far only partially effective in slowing credit growth. This is also clearly political; Turkish Prime Minister Erdoğan has attacked what he calls the [foreign] “interest rate lobby” and CBRT Governor Başçi has said that Turkey has “the most creative monetary policy in the world”. It is yet to be seen how effective this policy will be. In any case, this underlines that the central bank cannot achieve undervaluation alone, and again that consistency with government (especially fiscal) policy is needed. Run-away inflation would undermine real undervaluation.

One could also consider restrictions on the capital account. The case for a market-based control on inflows (i.e. a Brazilian tax on portfolio flows, or Chilean Unremunerated Reserve Requirement) is well-presented, but not uncontroversial. The IMF has also recently worked on a framework for capital flows, including capital flow management policies, and there has been significant debate on the issue. In the 1990’s, the IMF argued that, like trade restrictions, capital controls are an unnecessary distortion. This point of view has changed. There is some acceptance that – while the first-best strategy is more financial deepening and the development of a strong macroprudential framework – there could be a case for developing countries to liberalize capital accounts slowly, and even reinstate capital controls when other policy options have been exhausted. The point is that, while open capital accounts may be optimal in an idealized world of efficient financial markets, bringing risk diversification and savings to the most productive places, these conditions are not given in practice due to both national and global factors. In the volatile post-crisis environment, with financial de-leveraging and spillovers from unconventional monetary policy, we are even further from the idealized neo-classical world (if we were ever there in the first place). Kose, Prasad and Taylor (2009) argue that there are thresholds in financial development, below which it is not optimal to open the capital account. I am partial to this view, which seems to have informed the IMF’s recent work. Meanwhile, Rodrik (2008) ever the pragmatist, thinks that capital controls should simply be a permanent part of the international monetary system. This loses sight of the fact that capital controls tend to lose their effectiveness and be circumvented over time. Moreover, there are macro-prudential policies, such as limits on foreign exchange-denominated borrowing or open foreign exchange positions, which work similarly without explicitly targeting foreign investors. As Romeo L. Bernardo has written in this paper, such tools are also used in the Philippines. In the longer run, a sound macro-prudential framework, of the kind being developed in many developed and developing countries, could support financial stability, de-burden other areas of macro-economic policy and address the specific risks of cross-border flows. Yet for an intermediate phase toward development, it may be wise to be cautious with international investment flows, and to work on building the pre-conditions for open capital accounts (sound supervision, deepening, open equity markets) before the first-best solution can be pursued.

Jon Frost

The author is an economist at De Nederlandsche Bank (DNB), the central bank of the Netherlands. The views expressed here are his own and do not necessarily reflect those of DNB.