Go to Brad DeLong's Home Page
J. Bradford DeLong
University of California at Berkeley and NBER
As I put this talk together, I discovered that I am more gloomy about prospects for successful management of future international financial crises than I had thought: that the main lessons I at least have drawn from the experience of the Mexican crisis of 1994-1995 is that the political foundations of successful international economic management are much weaker than I had previously believed. The peso crisis of 1994-5 was in the large successfully handled--yes. But it was a near-run thing. And a lot of the political factors that made for success in this case are unlikely to be repeated in the future.
Let me--briefly--run through what happened.
At the beginning of 1994 there was a substantial consensus that the Mexican government was doing most things right. Over the preceding half decade the government budget had shifted from substantial deficit to surplus (thus no longer draining Mexicans' savings pool); businesses were privatized; import restrictions had been lifted; and--with the NAFTA--tariffs against the U.S. and Canada were being eliminated. Since 1989, the government deficit had fallen from five percent of GDP to zero. The inflow of capital had risen from zero to five percent of GDP. Investment in Mexico had boomed, as investors worldwide looked forward to a continuation of market-oriented policies that would boost Mexican demand for manufactured goods, and make Mexico a highly attractive export platform.
At the beginning of 1994 there was some worry about the level of he Mexican exchange rate. Rudi Dornbusch, for example, thought that the peso needed to be devalued against the dollar by perhaps twenty percent to keep Mexico from pricing itself out of world trade. Guillermo Calvo and the Congressional Budget Office agreed. Others pointed to the fact that in 1992, 1993, and again in 1994 the Mexican government had had to intervene in the foreign exchange markets to keep the value of the peso from rising: foreign investor confidence in Mexico was very high. How relevant were purchasing power parity considerations when foreign investors saw such a bright future for Mexico?
By October of1994 things had changed. Mexico's central bank had flooded the country with money during the presidential election year, causing some to lose confidence in Mexico's long-term commitment to moving closer to price stability (even though Mexican government officials hastened to reassure everyone that the policy of inflation had been the result of a "political" business cycle, and would not continue). Political assassinations raised the possibility that Mexico's ruling elites had started settling their differences with guns. The 1994 presidential election had not been very clean--although few had expected it to be. And an armed guerrilla insurrection had been mounted in the south, in Chiapas.
Each of these shocks led a few billion more dollars of foreign investment to leave Mexico, and thus led to the loss--given that Mexico had pegged the peso to the dollar--of a few more of its foreign exchange reserves. But it was no surprise that Mexico's ruling PRI was a corrupt oligarchy, that the peasants in the south were angry at a rural power structure that had stolen their land, that the government manipulated the economy to try to boost political support. All these were true--but if Mexico was a good place to invest in 1992 and 1993, there was no news that should have convinced anyone in 1994 and 1995 that it was a bad place to invest. Mexico's economic fundamentals--a balanced federal budget, a successful privatization campaign, financial liberalization--were strong enough in the spring of 1994 to elicit a strong and unqualified endorsement of Mexico's economic management by the IMF.
The Mexican government's response to each episode of capital flight in 1994 was to run its reserves down a little more, and wait for the resumption of the flow of foreign investment capital into Mexico. But the flow never resumed. Mexico ran out of reserves, and announced the devaluation of the peso in December 1994.
When the peg was abandoned, the peso fell by far more than the twenty percent that Rudi Dornbusch and others had predicted. It fell by more than sixty percent. Each investor in Mexico feared that other investors would pull their money out no matter what the cost, and that the last to withdraw money would lose the greatest amount of invested principal--through near-hyperinflation, as the Mexican government printed pesos to cover its peso-denominated debts; through capital controls, which would trap money in Mexico for an indefinite time (and eat up a substantial fraction of its value); or through formal default: a repeat of 1982'. With $5 billion in reserves, with $23 billion in dollar-indexed liabilities that were being pulled from Mexico as fast as they matured, and with no one willing to lend in hard currency for fear of becoming losers in the financial crisis, Mexico alone had no good options. Either push interest rates sky-high to keep capital inside the country, and cause a Great Depression; or start printing money at a rapid rate to meet its obligations and see a hyperinflation develop.
But as an economy Mexico was not insolvent. It was merely illiquid. Had investors had been willing to roll over Mexico's short-term debts, a moderate devaluation coupled with a small recession would restore external balance. Hence a support package: the first attempt being a set of loan guarantees, and the second attempt being a fund set up by the United States, the International Monetary Fund, and stray other amounts of perhaps $40 billion for Mexico to borrow.
How do we know that Mexico was not insolvent but illiquid, and that a support package was a good deal rather than throwing good money after bad? We know now because it worked. We know because of the speed of Mexico's recovery: Mexico registered a $7.4 billion trade surplus in 1995. We believed then that it was a good deal because Mexico's government appeared serious and its fundamentals appeared strong.
Thus in retrospect it looks as though the Mexican peso crisis was just what central banks were designed for: the private market was deranged not because of bad fundamentals but because of a fear that others in the private capital market had panicked. In such a case there is a role for a lender of last resort to make the bridge loans that the private market would be eager to make if it were in its right mind, and if it had confidence in the successful resolution of the crisis. And the U.S. government and the IMF did indeed act as a lender of last resort.
But the source of my relative gloom is that it was a near-run thing.
As President Clinton began to try to move approval for the loan guarantee version of the support package through congress, Speaker Gingrich's lieutenant, Majority Leader Armey, demanded that the Administration gather over 100 House Democratic votes for the package before he would lift a finger to recruit Republicans. Perennial presidential candidate Patrick Buchanan called the support package a gift to Wall Street: "not free-market economics [but] Goldman-Sachsanomics." Ex-consumer advocate Ralph Nader urged the Congress to vote down the support package and to demand that Mexico raise wages. Columnists in the Wall Street Journal demanded that support be provided only if Mexico first returned the peso to its pre-December parity.
Isolationist Republicans and anti-NAFTA Democrats claimed that NAFTA had caused the crisis, and vowed to fight the package. Newly-elected House Republican Zach Wump emerging from one briefing by Federal Reserve Chairman Greenspan crowing that he was opposed becasue this was "an issue made for talk radio." Newspapers like the New York Times and the Washington Post, usually staunchly internationalist, ran not-very-coherent op-ed pieces denouncing the support package. Senator Dianne Feinstein lectured Treasury Secretary Robert Rubin: "I know no one in the financial community who is against this. I know no one in my constituency who is for it."
On January 30, Speaker Gingrich announced that he could not put together a vote before the middle of February.
On January 31 the White House ditched the first version, and moved ahead in cooperation with the IMF with a different support package that drew on IMF reserves and on the Treasury's Exchange Stabilization Fund.
Now, the authorizing legislation governing the ESF assumes that it is for short-term exchange market interventions to stabilize the dollar's value in terms of other major international reserve currencies. It had never entered anyone's mind that the Executive Branch had the power to use the ESF to stabilize the peso against the dollar. Until driven by necessity--and under cover of a letter from Speaker Gingrich, Majority Leader Dole, and Minority Leaders Gephardt and Daschle stating that they would make no objection to use of the ESF for such purpose--I know of no one in the Treasury Building who had previously imagined that the ESF could be stretched so far.
Thus a policy that promised to (a) stem an international liquidity crisis, (b) avoid a likely Great Depression in our neighbor to the south (and hence avoid a large increase in illegal immigration), (c) avoid regional slowdowns in Texas and California, and (e) probably make money for the U.S. Treasury--turned out to be impossible to push through the Congress in early 1995. It was not that anyone disagreed with the argument that Mexico was suffering from a liquidity crisis in which substantial support from other countries could do a lot of good at minimal risk--the debate was never joined on those terms. What seemed to excite rage was that the U.S. government wanted to do something nice for Mexico. And, even worse, to do something nice for investors in Mexico.
The applause line of the critics was that the U.S. government had undertaken a $50 billion bailout for Treasury Secretary Robert Rubin's Wall Street friends and corrupt Mexican druglords. But the support package did not make investors richer by making anyone else poorer. Stemming financial crises is a positive-sum game: everyone wins. Workers keep their jobs and small businesses avoid bankruptcy. These benefits probably add more to the sum of human welfare than do the higher asset values created. That financiers and investors gain along with everyone else is no argument against acting as a lender of last resort.
Perhaps worse, echoes of the sentiments of the U.S. Congress were audible in the comments of some industrial-country governments with substantial voting shares in the IMF: that the Mexican crisis was not a "systemic problem," and that the rescue program bailed out those who had made imprudent short-term investments in Mexico. Germany and Britain abstained from the IMF executive board meeting that authorized the IMF's contribution to the support package. Ultimately the IMFdid step up to the plate. The IMF executive did win support for its decisions from its directors. The IMF proclaims its willingness to do so again. But would it have risked doing so if the U.S. executive had not been extraordinarily interested in making sure that this rescue package happened?
Any assessment of what the political firestorm over the peso support package means for future management of the world economy must be a depressing one. Faced with what in prospect seemed highly likely to be, and in retrospect seems a classic example of, a "liquidity" crisis in which international support produces huge economic benefits at very little risk, the U.S. Congress could not step up to the plate. The only positive note struck is that the Congress was equally unwilling to put itself on the line by blocking those whowere doing the right thing.
There will, someday, be another international financial crisis. The IMF and industrial country governments will be faced with a run on some currency or class of assets. They will have to make a snap judgment as to whether the run is a self-fulfilling prophecy--a liquidity crisis--that they should neutralize; the result of fundamentals going bad--a solvency crisis--in which orderly liquidation is called for; or a situation in which public-sector involvement is out of place. The U.S. executive branch went ahead with the peso support package because:
- it had, as a result of foreign exchange interventions during the 1980s, $20 billion extra of foreign exchange reserves under executive branch control that it could at least argue it could constitutionally use to fund the support program.
- it had, during the NAFTA debate, tied its political future to the success of economic reform in Mexico.
- a Great Depression or a hyperinflation in Mexico would cause severe economic dislocation in--and greatly enhanced illegal immigration to--Texas and California (although it would have little impact on the rest of the U.S. economy).
- Alan Greenspan and Ted Truman at the Federal Reserve, Robert Rubin and Lawrence Summers at the Treasury, and Michel Camdessus and Stanley Fischer at the IMF were confident that the positive reading of Mexico's fundamentals was the correct one: that Mexico's economic future was bright, and that it was the stampede of capital out of Mexico in 1994-95 that was an irrational panic (rather than the stampede of capital into Mexico in 1992-93 being an episode of irrational exuberance).
- the congressional leaders--new on the job, in their first month, finding their feet--were singularly willing to cede power over the use of federal money to the executive banch.
We are never again going to have a situation in which all five of these elements come together. We may have two or three of them--but not all five. A repeat financial crisis somewhere else will either fail to attract strong enough attention in the executive to override congressional foot-dragging; or executive and Federal Reserve policy makers may not be as confident of their assessment; or the congress may be more interested in establishing that it spends the money and the president merely executes the laws than--as the 1995 congress was--interested in ducking the issue.
European policy makers are not becoming more enthusiastic about rescue packages.
The only source of potential leadership will be the IMF. Would it have been willing to act on its own, over the objections of large shareholders? Probably not. Will it be willing to act on its own in the future? Perhaps. People are trying to strengthen its ability to conduct such lender of last resort activities. I hope so. But I am not very optimistic.
Go to Brad DeLong's Home Page
Professor of Economics Brad DeLong,
University of California at Berkeley; Berkeley, CA 94720-3880
(510) 643-4027 phone (510) 642-6615 fax