J. Bradford DeLong
The five major financial crises of the 1990s have five important lessons to teach: (1) a more globalized world is likely to be a more unstable world; (2) comparatively small weaknesses in macroeconomic policy draw swift and massive punishment from international financial markets; (3) severe crises can happen where there are no visible weaknesses in economic policy; (4) the current IMF judgment-based system for supporting countries in crisis works (and would work better if the IMF were better funded); yet (5) the political support for our current system for dealing with financial crises is extremely shaky.
Let's look at these in more depth.
First, the 1990s saw five major financial crises--the western European exchange rate mechanism crisis of 1992, the Mexican crisis of 1994-1995, the East Asian crisis of 1997-1998, the Russian crisis of 1998, and the Brazilian crisis of 1998-1999. The magnitude and the frequency of these crises came as a surprise. As Paul Krugman has pointed out, a decade ago the near-consensus of economists was that recent financial crises had been caused by too little globalization, not too much. This near-consensus expected that continued international economic integration would be accompanied by decreasing virulence and frequency of financial crises, not increasing virulence and frequency.
Why? Because the financial crises of the 1980s had largely arisen as a result of bad news or bad policies that required large sudden swings in countries' current-account balances. With the limited degree of globalization then found in the world economy, exports in many countries could not be expanded swiftly enough. Thus a shortage of foreign exchange caused a crisis. And with limited globalization, there was no pool of mobile capital to delay (at a healthy interest rate, of course) the necessary current-account adjustment for those countries that found their underlying fundamental balances swinging widely.
Economists by and large expected the decade of the 1990s to see increased globalization. Thus economists by and large expected the 1990s to be a relatively placid decade, at least as far as financial crises were concerned. They were wrong: globalization has turned out not to be a cure but a cause of financial crises. The benefits of free trade and free investment may well outweigh the costs (I believe that they do by a substantial margin). But no one should advocate increased international economic integration as a path toward macroeconomic stability. The fact that a globalized world is likely to be a more unstable world is the first of the lessons of the 1990s.
Second, both Britain in 1992 and Mexico in 1994 followed macroeconomic policies that were very close to the model of financial orthodoxy as prescribed by the IMF staff in their Section IV consultations. The British government was willing to spend a fortune--both in Treasury cash and in political capital--to maintain its fixed exchange rate vis-a-vis the DM. The only weakness was that markets judged that had the link to the DM not existed, no one in the British government would support establishing it--and that no one in the British government would advocate a restoration of the link if market speculation were to break the European Exchange Rate Mechanism. The Mexican government was fiscally very strong and the Bank of Mexico was committed to reducing inflation--except that in an election year the Bank of Mexico would temporarily shift priorities.
In both cases the policy weaknesses were relatively minor. Yet these relatively minor policy weaknesses generated major losses of market confidence in the governments' exchange rate policies. The lesson is that in today's world there is very, very little room for error on the part of a country that seeks to peg its exchange rate--hence today's consensus belief that a country seeking to peg its exchange rate needs to do so in a decisive and irrevocable manner (like dollarization), or it would be better off not to try.
Third, in retrospect it is very hard to recall why there was an East Asian financial crisis at all. Yes, Thai banks were underwater. Yes, many of Korea's chaebol had extremely shaky finances. Yes, Suharto's government was corrupt (and, in East Timor, murderous). But the overhang of bad debt in East Asia was nothing that several more years of rapid economic growth like that seen in the first half of the 1990s would not rapidly shrink down to size. It was only the financial crisis itself that made the overhang of debt large and dire enough to cause a large-scale financial crisis.
The lesson is that it is not enough for macroeconomic and financial fundamentals to be sound. Macroeconomic and financial fundamentals must be strong enough to remain sound even in a large-scale financial crisis. The possibility that a crisis may in the end be caused by nothing more than fear that there will be a crisis--and that a crisis may then generate the weaknesses that sustain and validate the possibility of a crisis--is very, very real.
This means especially that it is extremely dangerous for a developing country to borrow in hard currency. The exchange rate depreciation that accompanies a financial crisis greatly increases the home-currency value of foreign debt. Thus a country whose government, banks, or businesses have borrowed on a large scale in foreign currency will find itself vulnerable to economic chaos: the flap of a butterfly's wing that diminishes confidence in economic growth can set in motion a self-sustaining downward spiral. Debt-heavy and especially foreign currency-denominated debt-heavy financial structures are too risky for emerging economies in today's world.
Fourth, however, the support provided for East Asia (and Mexico) coupled with the natural bounce-back of market economies from depression has worked. Growth and prosperity have quickly been restored. In the past year real GDP in South Korea has grown by 13 percent; in Malaysia it has grown by 12 percent; in Thailand 7 percent; and in Mexico 8 percent. Of the emerging market economies most badly hit by the financial crises of the 1990s only in Brazil and Indonesia is real economic growth--3 percent over the past year--still disappointing. Real wages in Mexico, Thailand, and Indonesia are still well below pre-crisis levels: recovery is far from complete. But there is good reason to think that recovery will soon be complete in a much, much shorter period of time than after previous financial crises like 1982.
A couple of years ago everyone seemed to agree that the international financial institutions needed major reform. Half the critics (the Jeffrey Sachs-Joseph Stiglitz wing, with which I tend to agree) believed the institutions were too scrooge-like: he IMF was forcing countries into deflationary policies that caused severe depressions. Half the critics (the Ralph Nader-Wall Street Journal wing) believed the institutions were too generous and liberal: the IMF's generosity had encouraged overlending and overproduction that had caused widespread crisis. But even though the directions of the proposed reforms were directly opposed, everyone seemed to agree that such major reforms were absolutely necessary.
Yet here we are. So the natural question is: what went right?
What went right was that our global institutions did a much better job of handling the crises of the 1990s than the folk wisdom holds. For more than a hundred years we have known what to do in a financial crisis: show up with a lot of money to restore confidence, lend freely to fundamentally sound organizations that need cash, and rapidly close down and liquidate businesses that aren't going to make it even if the crisis is successfully resolved. This keeps the financial system working. This was what was done, with approximately $20 billion of public money committed to Mexico and approximately $60 billion committed to East Asia. Mexico had one and East Asia had two bad recessionary years, followed by a resumption of rapid economic growth. In my view at least, we got more than our money's worth out of the IMF and its partners in the 1990s.
But, fifth, there is no doubt that the political underpinnings of our current system are weak. As Representative Barney Frank told economist Barry Eichengreen at one congressional hearing, the end of the Cold War has robbed the internationalists of 50 votes in the U.S. House of Representatives. The peso support package of 1995 that has worked out so well for Mexico almost failed to happen, and in the end happened only because the U.S. Federal Reserve and Treasury were willing to skate the borders of their authority, with Congress willing to stand by and watch them do so with the understanding that heads would roll if things did not turn out OK. The IMF received a blizzard of criticism for both the macro and micro aspects of its role in East Asia in 1997 and 1998. The fact that the critics had opposed and inconsistent complaints does not change the fact that supporters seemed few and far between.
Thus the fifth and last lesson is that work needs to be done to rebuild support for a crisis-management system that seems--ironically--to have worked relatively well in the past decade. And if support cannot be rebuilt, than a new system of crisis management that can command broad political support needs to be built in its place.
Sign up for Brad Delong's (general) mailing list