In the Wake of the Crisis: Leading Economists Reassess Economic Policy

14. Remarks on Capital-Account Management and Other Macropolicy Topics

International Monetary Fund
Published Date:
March 2012
  • ShareShare
Show Summary Details
Arminio Fraga

The efficacy of capital controls is usually overestimated by governments and underestimated by market participants. Market players ignore the fact that governments may go way beyond optimal levels of taxes and controls, often because governments tend to ignore long-term costs and side effects. But if participants underestimate what governments may do, that is their problem. It does not matter.

More serious is what happens on the policy side. Policymakers see too little room for arbitrage, and therefore tend to ignore the loss of efficacy over time of most such policies. They also ignore or underestimate the side effects of controls.

Brazil’s experience is a good example. For decades capital outflows from Brazil were banned. As a result, a black market for hard currency developed outside the boundaries of the law. Once money was taken out of the country, it could not be taxed. What is worse, this policy was part of the development of a culture of transgression, an integral part of the lost decade of 1981 to 1993. More recently, taxes on inflows have been introduced, but it is still too early to tell how relevant the side effects will be. The effects may include moving some portion of financial-market activity offshore, with loss of transparency and liquidity for locals, as well as a negative impact on more desirable long-term flows such as direct and equity investments.

This brings to mind a crucial point: the regulation of capital flows can be a useful tool, perhaps a second-best response. One very obvious example currently in evidence is the flood of short-term capital inflows to developing countries. However, while this is true, one must not forget that not all capital flows are driven by supply or push factors such as zero interest rates at the core of the system. Some may be driven by policies that themselves deliver undervalued exchange rates or ultra-high-interest rates in the recipient country, both key drivers of inflows.

So great care must be taken not to postpone or avoid facing more urgent problems. Again using Brazil as an example, nothing would do more for Brazil than policies that allowed interest rates to go down toward where they are in similar countries (say, 2 to 3 percent real rather than 6 percent real). This would greatly reduce the incentive for carry trades.

The same argument applies to China and Korea right now. In the case of China, one could argue that the pegging of the renminbi at an undervalued level is as much a bubble machine as the zero or close to zero interest rates found in the G3. All the money in the world wants to go to China.

More generally, we run the risk right now of seeing the widespread use of controls turn into a protectionist game, leading to large losses for the global economy in aggregate. Here there is a clear need for global coordination, as in the case of trade.

So what should countries do? Is there room for sensible capital controls? One conclusion that jumps to mind is that, in general, capital controls should not be seen as a permanent solution to problems.

Another key lesson, taught to us by Carlos Massad, a former governor of the Bank of Chile, is that policies that reduce volatility are good while policies that suppress volatility are bad.

The principles that apply to the ongoing debate on monetary and prudential policies apply to the case of capital account regulations and controls as well: risk is endogenous and systemic weaknesses must be addressed by macroprudential policies.

The most important case by far is the clear tendency for contemporary financial systems to generate way too much short-term borrowing in the aggregate. I do not believe this is a natural state of affairs but rather a man-made phenomenon, a creature of poorly designed incentives and asymmetric monetary policies, as mentioned by Otmar Issing. That must be fought by authorities at the root-cause level directly through the elimination of bad policies or through instruments such as the taxes suggested by Hyun Song Shin and others.

In a balance of payments context, the risk of allowing too much short-term borrowing tends to be amplified by the added currency mismatch, so there is an additional reason to watch out!

My conclusion is that capital account management tools should be applied in a context that is clearly prudential. They should not, however, be used to support undervalued currencies, in mercantilist fashion. Therefore, in practice this means capital account policies must be judged according to their objectives and not by the tools used in their implementation.

This brings me to a point that has been neglected in these times of zero rates, these times of push rather than pull: governments have a terrible track record of defending overvalued exchange rates. In my view they are likely to be just as bad in defending undervalued exchange rates. This poses two issues:

  • At the country level, care must be taken to avoid creating credit and asset bubbles and losing control of inflation.

  • At the International Monetary System level, the abuse of capital account tools can be just as damaging as protectionist policies, and can lead to the imbalances that we all know so well.

Lastly, I am of the view that central banks are perhaps being asked to do too much: to keep inflation low and stable, and to the extent possible to smooth the business cycle (not to try to abolish it); to be the financial regulator, with special emphasis on system risk; and, lastly, to manage the exchange rate.

The tools at the central bank’s disposal are: monetary policy, prudential tools (reserve and margin requirements, taxes, etc.), and intervention in the foreign exchange market. The limitations of these tools and the short- and long-term constraints that these objectives must meet are reasonably well known, though not always by short-term-focused governments.

The central bank may also be called to the table to optimize the package with fiscal policy as an extra tool, adding an extra degree of complexity to the exercise. In theory, governments may deal with this difficult optimization problem in competent fashion. In practice, the challenges are tremendous, governments are rarely wise long-term optimizers, and problems and errors may easily ensue.

Take the case of intermediate exchange-rate regimes. I believe such regimes are possible, despite the time decay in the efficacy of capital account policies. I just don’t think managed exchange rates are a good idea. They lead to confusion and temptation. Confusion because economic actors are not given clear signals as to how they ought to structure their economic and financial lives (e.g., their balance sheets) with respect to interest-rate and exchange-rate risk. Temptation because governments are often prone to succumb to short-term political pressures and end up defending unsustainable exchange rates. My preference therefore is for being closer to the extremes of the regime spectrum, allowing of course room for prudential policies and, in the case of a flexible regime, for intervention in exceptional circumstances.

Another important case has to do with fiscal policy. Many G3 countries are now running unsustainable budget deficits that in most cases will tend to worsen over time. This approach is highly likely to end badly, but no serious action is being taken yet.

Both of these examples belong in one form or another to the time-consistency family. Great economic tragedies such as hyperinflations, banking crises, and sovereign debt defaults, with their accompanying social pain, are almost always the result of errors of this kind.

I therefore see great merit in the old-fashioned view of building sound long-term institutional constraints, and in some degree of separation of responsibilities. There is great frustration in the air concerning the conduct of economic policies before and during the great financial crisis of 2007 to 2009. This frustration is legitimate and provides great energy to the ongoing effort to rethink policy frameworks and practices. But in a world of fallible and short-term-focused policymaking, care must be taken not to forget some important good old lessons and thus not to let the pendulum swing too far toward excessive complexity and discretion. In particular, I am of the view that a framework that delivers low and stable inflation and sustainable fiscal policy is essential and must be constructed where absent. Inflation targeting and fiscal responsibility laws are a good way to achieve these goals. What is missing is a proper macroprudential framework to supplement the always important (but frequently flawed) microsupervision effort.

    Other Resources Citing This Publication