A quarterly magazine of the IMF

December 2002, Volume 39, Number 4

As a relatively new arrival to the IMF, I cannot bear firsthand witness to who said what, when, during the buildup to the 1990s Asian debt crisis. Was the IMF as guilty of ramming accelerated capital account liberalization down countries' throats as is sometimes alleged? A wide polling of "old-timers" yields mixed results. Overall, my sense is that whereas certain antiglobalization polemics grossly overstate the charge, it still has some currency. At the very least, in some of its routine surveillance missions during the early to mid-1990s and in some of its technical assistance advice, the IMF was not forceful enough when it warned countries with weak financial systems and inadequate macroeconomic frameworks that they were opening up to short-term borrowing from abroad too quickly.

These days, everyone agrees that a more eclectic approach to capital account liberalization is required. But what should it be? How can developing countries drink from the waters of international capital markets without being drowned by them?

To begin with, we should acknowledge that this is a tough question, and we should be humble about what we know and don't know. This doesn't mean that we have to toss out the first welfare theorem of economics—which espouses the virtues of competitive markets—and agree with those antiglobalists who have never met a capital control they didn't like. After all, capital controls often breed corruption and always engender distortions. Human ingenuity usually ensures that their effectiveness is eroded over time, through avoidance and evasion. And to sustain effective capital controls indefinitely, a government has to be prepared not only to intervene heavily throughout the trade and financial system—a policy that has highly undesirable side effects—but it must also be disciplined enough (which few are) to limit excessive capital inflows during boom times.

Which types of capital flows are we talking about controlling? Certainly not foreign direct investment (FDI), which tends to be by far the most stable form of external finance and is often accompanied by transfers of foreign managerial skills and technology. And, hopefully, not foreign holdings of equities, which provide great benefits in terms of risk sharing. True, sudden reversals by foreign investors can be a problem for countries with rigidly fixed exchange rates, but that says more about fixed rates than it does about cross-border holding of equities.

The real debate is about debt instruments—both short-term flows and certain types of long-term flows (for example, debt payable in foreign currency or indexed to short-term domestic interest rates). Debt does not have the desirable risk-sharing properties of FDI and equity, and it makes countries susceptible to reversals of sentiment and insolvency problems. That said, let's be clear that it is nonsense to think that all short-term flows are bad: this misperception, in fact, is one of the many weaknesses of the proposed Tobin tax, which would discriminate indiscriminately against all short-term flows. Trade credits, the lifeblood of international trade, are a type of short-term capital flow. Speculative flows normally play an essential role in stabilizing foreign exchange and financial markets. And short-term instruments enable investors to hedge, creating, in turn, a welcoming environment for FDI.

All of which supports the view that the right starting point for thinking about capital controls must be on very focused, temporary measures aimed at stemming massive temporary inflows or outflows of debt.

To come up with appropriately nuanced measures, however, we need to better understand whether and when the benefits of capital flows exceed the costs. For trade in goods and services, most economists agree that the benefits of unhindered flows far exceed the costs. Why isn't the same obviously true for other capital flows? Here is where the results of the past 8 or 10 years of academic research might surprise you. Researchers pretty much agree that from a qualitative perspective, international capital market integration is a good thing. The rub is that it is awfully hard to show that the benefits are important quantitatively.

In a recent IMF paper, Pierre-Olivier Gourinchas and Olivier Jeanne note that modern growth economics attributes most of the difference in output per person across countries not to differences in capital-labor ratios, but to differences in total factor productivity. That is, richer countries are richer mainly because they get more output from the same amount of capital and labor inputs. This suggests that a significant part of the difference in economic well-being across countries has to do with "soft" factors like institutions, corruption, and governance. Opening up the valves on international capital flows might help equalize capital-labor ratios across countries without closing the big gaps in productivity. Therefore, the traditional gains from capital market integration might be less than previously thought.

How about indirect benefits? Certainly capital flows (especially FDI) facilitate the transfer of technology, promote competition and financial development, and create pressures for improved governance. But these benefits are difficult to measure, and it is tough to prove that capital flows provide the only means of securing them.

In principle, international capital market integration also helps countries diversify risk. However, a series of papers by Obstfeld and various coauthors (including myself) show that the welfare benefits of risk sharing are also not as great as one might imagine. One reason is that as long as countries can trade in goods, relative price effects bring about some of the risk sharing that would otherwise take place through equity markets. Another is that consumption just isn't that volatile in most industrial countries, so the benefits to reducing volatility must similarly be limited. Granted, estimates of the potential risk-sharing benefits of financial integration tend to be decidedly larger for developing countries, because they tend to be much less diversified internally and far more dependent on commodities with volatile prices. However, this assessment doesn't take into account periodic debt crises.

But should we go so far as to dismiss the benefits of capital flows as not worth the risk? There was a time when economists found similar negative results concerning the benefits of free trade—and those early results have now largely been reversed. Also, historically, trade and financial integration have tended to move in tandem, probably largely because trade integration requires capital market integration.

Where does this leave us? I don't want the reader to go away thinking that the IMF's new Economic Counsellor thinks capital controls are the greatest thing since sliced bread. In my brief tenure, I have seen a lot more destructive capital and exchange controls than I have seen healthy ones. Even where some limited form of capital control is warranted on economic grounds, actual implementation is all too often dominated by political considerations, and the results are not pretty. A few powerful political stakeholders benefit but only at a high cost to other citizens. And no country that has made substantial progress in capital account liberalization has been inclined to reverse that progress in any long-term sense—not even countries that have suffered crises. Countries that have opened up their capital accounts seem to believe that whatever problems they faced in liberalizing, the benefits will exceed the costs going forward.

Hard work remains to be done on capital account liberalization and its sequencing with other policies to find the point at which the benefits to further capital market integration stop exceeding the costs. An analytical study in a recent issue of the IMF's World Economic Outlook (WEO) stressed the importance of making sure that financial market liberalization does not get too far ahead of trade liberalization. Another WEO study suggested that, to some extent, exact sequencing may be less important than making sure that macroeconomic and regulatory policy are strong prior to liberalization.

In the meantime, there seems to be a good case for keeping an open mind on the issue of capital controls and debt—especially when debating ways to better immunize the global financial system against crises in the twenty-first century.

Whatever historians ultimately conclude about the evolution of IMF doctrine, the view that excessive debt flows can be problematic for developing countries is certainly not new to me. My 1990 Journal of Economic Perspectives paper with Jeremy Bulow offered a blueprint for restructuring the international financial system. Our blueprint focused on what industrial countries could do to level the playing field for FDI, equity flows, and debt. This still seems to be the right place to start.

References:

Jeremy Bulow and Kenneth Rogoff, 1990, "Cleaning Up Third-World Debt Without Getting Taken to the Cleaners," Journal of Economic Perspectives, Vol. 4 (Winter), pp. 31-42.

Pierre-Olivier Gourinchas and Olivier Jeanne, 2002, "On the Benefits of Capital Market Integration for Emerging Market Economies" (unpublished; Washington: International Monetary Fund).

Robert E. Hall and Charles I. Jones, 1999, "Why Do Some Countries Produce So Much More Output Per Worker Than Others?" Quarterly Journal of Economics, Vol. 114 (February), pp. 83-116.

Kenneth Rogoff, 1999, "International Institutions for Reducing Global Financial Instability," Journal of Economic Perspectives, Vol. 13 (Fall), pp. 21-42.