CHAPTER ONE

The First Crisis of the Twenty-first Century

ON THE EVENING OF January 10, 1995, I stood on the Great Seal woven into the carpet of the Oval Office and swore to uphold the Constitution of the United States as Secretary of the Treasury. Confirmed earlier that day, I had been waiting all afternoon for the official document that would allow me to take the oath of office. Once the papers arrived from Capitol Hill, a small group of family, friends, and colleagues assembled at the White House for a hasty ceremony.

As soon as the formalities were over, I said good-bye to my wife, Judy, and our other guests and remained behind with President Bill Clinton, Treasury’s top international official, Larry Summers, and a few of Clinton’s senior advisers, for an emergency meeting about the financial crisis in Mexico.

I told the President that the Mexican government faced an imminent threat of default and that, in the hope of preventing it, we were recommending that he support a massive, potentially unpopular, and risky intervention: providing billions of dollars to the Mexican government to avoid a collapse in its currency and economy. Then I asked Larry to explain the situation in more detail. It took him ten minutes to spell out our essential analysis and recommendation, which we’d finished formulating in a meeting with Fed chairman Alan Greenspan hours earlier. If our government didn’t step in to help, and help quickly, the immediate and long-term consequences for Mexico could be severe. But the real reason for acting was that critical American interests were at stake.

The alternatives to the massive intervention we were recommending were not promising. If Mexico defaulted on its foreign obligations, Larry and I went on to explain, the flow of capital out of Mexico would probably accelerate and the peso would collapse, likely triggering severe inflation, a deep and prolonged recession, and massive unemployment. And that would surely have a substantial impact on the United States. Mexico was our third-largest trading partner, which meant that many American companies and workers would be hurt. We presented estimates that a Mexican default could increase illegal immigration by 30 percent, a half-million additional refugees a year. The flow of illegal drugs could intensify as well.

A crisis in Mexico might also hurt us indirectly, by affecting other countries. Fears of a Mexican default were already producing wobbles in developing markets throughout the hemisphere, a phenomenon that came to be known as the “Tequila Effect.” Such a chain reaction could lead investors to pull back from emerging markets around the world indiscriminately. That, in turn, could affect economic conditions in the United States—since roughly 40 percent of our exports went to developing countries. According to an estimate made by the Federal Reserve Board, a Mexican default and the consequent “contagion” that was possible could, in a worst-case scenario, reduce growth in the United States by ½ to 1 percent a year. We weren’t proposing intervention for the sake of Mexico, despite our special relationship, but to protect ourselves. That was our case for asking Congress to provide billions of dollars in loan guarantees, as part of a package to be coordinated with the International Monetary Fund (IMF).

As Treasury Secretary, I avoided using words such as “panic” and “meltdown,” preferring less vivid terms such as “contagion” and “loss of confidence.” I’d learned while working in the White House as director of the National Economic Council (NEC) that the words public officials use can make an enormous difference, and that was even truer at the Treasury. I had to describe what was happening in Mexico accurately without being inflammatory. A meltdown, though, is precisely what we were worried about—and not only because of its effect on current economic conditions. With the implementation of the North American Free Trade Agreement (NAFTA), Mexico was hailed as a role model for developing countries pursuing economic reform. The public failure of that model could deal an enormous setback to the spread of market-based economic reforms and globalization.

But there were arguments against intervention as well, which we also laid out for the President. I emphasized that, for a variety of reasons, our rescue package simply might not work. What’s more, intervention would almost surely be criticized as “bailing out” wealthy American and European investors who had speculated on developing markets. Putting public funds on the line was likely to be massively unpopular and politically risky. Leon Panetta, the White House chief of staff, was even more blunt in warning Clinton about the downside risk. Leon favored intervention, but he told Clinton that a failed rescue effort could cost him the election in 1996.

When we finished our presentation, the room was heavy with a collective sense of how big a problem this had rapidly become. Larry had tossed out a figure of $25 billion as the amount of U.S. assistance that might be necessary. George Stephanopoulos, a senior adviser to the President, said that surely we meant $25 million. No, Larry said, we meant “billion with a B.” That was more than the annual budget of the Department of Justice, enough to buy a fleet of B-2 “Stealth” bombers.

Sitting on a sofa in the Oval Office during my first hour on the job, I was answering questions from the President that I had been asking others only a couple of weeks before. Larry had phoned me in December, while I was on vacation in the Virgin Islands, to bring me up to speed on the unfolding Mexican situation. I didn’t know much about Mexico’s economic problems, and I didn’t understand why a peso devaluation was urgent enough to interfere with fishing. I assumed that Mexico was one of a large number of countries with similar problems and that Larry, a consummate professional in the field of international economics, would take care of whatever needed to be done. But I intended to be a hands-on Secretary, and I liked that Larry was including me, even though I was still in the netherworld of being designated but not yet confirmed as Treasury Secretary.

It didn’t occur to me that day that Mexico’s problems would shortly blossom into a full-blown economic crisis that embodied the heightened risks of a more global economy. In retrospect, the Mexican episode also offers much insight into the Clinton presidency. The Bill Clinton I watched come to the aid of Mexico was one the public too seldom saw. His seriousness of purpose, his depth of substantive understanding, and his keen intellectual quest for the right decision on Mexico are a continuing reminder to me of the way in which he remains, in important respects, a misunderstood figure. At a broader level, the dilemma Clinton faced with Mexico suggests to me that our politics may not be well suited to coping with the new risks of the global economy.

   

GETTING MY ARMS AROUND a problem like the Mexican crisis meant thinking about it as systematically and dispassionately as possible. The situation, as I rapidly came to understand following my preconfirmation conversation with Larry, was this: After the outgoing Salinas government had spent over $15 billion in a futile attempt to prop up the peso at the fixed rate of around 3 pesos per U.S. dollar, the newly installed government of Ernesto Zedillo had, in late December 1994, surrendered to overwhelming pressure in foreign exchange markets and allowed the Mexican currency to float freely. With only around $6 billion of its foreign exchange reserves left and far more than that in short-term debts coming due, Mexico had little choice. But with the government no longer providing support, the peso fell rapidly to around 5 pesos to the dollar. As the Mexican currency continued to slide, doubts grew about whether the government would be able to repay its debt, much of it very short term and linked to the dollar. Fearing a possible government default, investors were selling Mexican bonds, as well as the peso. In sum, Mexican authorities had lost control of their country’s finances.

All of us working on the problem agreed that Mexico, now essentially cut off from private lenders, almost surely could not solve the crisis through its own policies alone. The Mexican government’s bond auctions were attracting few bidders, even at dollar interest rates approaching 20 percent. In the short term, the private sector was very unlikely to produce loans on the scale needed to prevent default.

Nor, with requirements this large, could the international financial institutions—the IMF and World Bank—arrange a rescue on their own, as they had in many other cases. Michel Camdessus, the French managing director of the IMF, was unknown to most Americans despite his tremendous influence. Skillful and audacious, Camdessus was prepared to weather the anger of his organization’s European shareholders to make a stabilization loan to Mexico of unprecedented size. But the sums needed exceeded the IMF’s available capability. The only realistic chance of avoiding disaster was help from the United States. The questions for me then became the possible consequences of financial chaos and default in Mexico, the danger of the program failing, and the possible costs of that failure.

What has guided my career in both business and government is my fundamental view that nothing is provably certain. One corollary of this view is probabilistic decision making. Probabilistic thinking isn’t just an intellectual construct for me, but a habit and a discipline deeply rooted in my psyche. I first developed this intellectual construct in the skeptical environment of Harvard College in the late 1950s, in part because of a yearlong course that almost led me to major in philosophy. I started to employ probabilistic decision making in practice at Goldman Sachs, where I spent my career before entering government. As an arbitrage trader, I’d learned that as good as an investment prospect might look, nothing was ever a sure thing. Success came by evaluating all the information available to try to judge the odds of various outcomes and the possible gains or losses associated with each. My life on Wall Street was based on probabilistic decisions I made on a daily basis.

This was the background I brought to the question of whether we should intervene in Mexico. With an enormous number of competing considerations, the key to reaching the best possible decision was identifying all of them and deciding what odds and import to attach to each—probabilistic decision making at work. Doing that also meant recognizing that our knowledge would never be as complete or perfect as I—or the rest of the team at Treasury—would like. Moreover, even with the most systematic and thorough work, a decision, though informed by the facts and analysis, would never emerge automatically from the yellow pad on which I scribbled notes. The final component of decision making was the intangible of judgment. The process of decision making that we evolved in the Mexican crisis—and that I would use over and over again in my time at Treasury—was familiar to me from my life in the private sector. But the range of considerations was much broader. For example, we had to think about the damage that a failed intervention could do to America’s credibility. If we attempted to help Mexico and did not succeed, our backing would be a less useful tool in some future crisis.

Success had dangers as well. Even if our efforts helped stabilize Mexico, we might create a problem of what is known as “moral hazard.” Investors, after being insulated from the consequences of risk in Mexico, might pay insufficient attention to similar risks the next time, or operate on the expectation of official intervention. In Mexico, investors had become complacent, following a herd mentality in buying short-term dollar-linked bonds throughout 1994 without paying sufficient attention to the danger that the central bank’s currency reserves might not be sufficient to maintain their promised convertibility into dollars. We worried that our program to prevent Mexico’s failure might encourage investors to make similar mistakes again in the future.

It was my good fortune to be able to think through these issues with Alan Greenspan and Larry Summers. In our backgrounds, our professional training, and our temperaments, the three of us were alike and very different. Alan is a conservative free-marketeer and an economist grounded in both macro policy and an acute empirical understanding of the American economy. Before entering government, he had his own private-sector consulting firm and traded actively for his own account. He is a precise man with an exceedingly good and understated wit. Larry, whose parents are both Ph.D. economists and who has two uncles who won Nobel Prizes in economics, was one of the youngest professors ever to receive tenure at Harvard. He is a forceful, self-assured theoretical economist with a good feel for the practical, both in politics and in markets. I had a pretty good conceptual understanding of economics, had spent a career in trading operations and management on Wall Street, and had been involved in Democratic politics. People who know me are familiar with my distrust of definitive answers and my habit of asking questions. While our personalities differed, they meshed—perhaps because our analytical approaches to a problem like Mexico proved highly compatible. Equally important was the spirit in which we worked. Though none of us is without ego, there was a remarkable lack of it in our meetings. Each of us tried to work with the others to find the best answer, not to show off his intellect or defend preconceived notions. Another crucial component of our relationship was the mutual trust we developed. For four and a half years, Alan, Larry, and I had breakfast or lunch at least once a week, along with many other meetings and discussions. After I resigned in 1999, Larry and Alan continued the tradition. To the best of my knowledge, nothing any of us said in any of those private meetings ever leaked out. (For this book, they gave me permission to refer to these conversations.)

I had seen Greenspan periodically during my time at the White House but hadn’t known him very well before I became Treasury Secretary. When we were both thrown into the peso crisis, we got to know each other rather quickly. I was deeply impressed with the way he thought about the problem. Alan, who believes strongly in the discipline of markets, was very focused on the issue of moral hazard. This was why he had opposed the government rescue of the Chrysler Corporation in 1979. But, despite his opposition to the idea of government intervention in markets, Alan weighed the moral hazard against the risk of having Mexico go into default. He was a pragmatist, trying to find the best way to balance competing considerations.

   

ALAN, LARRY, AND I AGREED about what had caused the crisis. Mexico, despite reforms in many areas, had made a serious policy mistake by borrowing too much in good times, leaving it vulnerable when sentiment shifted. And when markets began to lose confidence, the government put off facing reality for as long as possible. It borrowed still more, at shorter and shorter terms, issuing dollar-linked debt and spending its limited dollar reserves on holding up the peso, which had an exchange rate fixed to the U.S. dollar. At the same time, creditors and investors—both Mexican and foreign—were paying little attention to the buildup of economic imbalances. Their continued financing allowed the problem to become almost unmanageable when the crunch finally came.

The trouble really began in the early 1990s, when Mexico’s current account deficit—basically the trade deficit plus net interest payments and some similar items—began expanding rapidly. To cover this gap, the country needed dollars, which it attracted by issuing government bonds. At first, it sold peso-denominated assets. But later, as investors became less willing to take on the exchange rate risk, the government started issuing large quantities of Tesobonos, short-term obligations whose value was linked to the U.S. dollar. For a while, these bonds proved attractive to Mexicans and foreign investors. But Mexico’s large current account deficit combined with a fixed exchange rate was not sustainable indefinitely. To make matters worse, Mexico’s banking system was weak and under strain.

Underlying imbalances like Mexico’s are the real cause of resulting crises, but often some event that might otherwise not have created trouble serves as a trigger. In this case, a violent insurgency in the Chiapas region at the beginning of 1994 and the assassinations of two leading Mexican politicians created a deep sense of alarm in financial markets. Mexican bonds began to look much riskier and started to trade at steep discounts. Domestic and foreign investors became less willing to keep money in Mexico. The central bank had to sell more and more of its foreign exchange reserves as it struggled to meet the demand for dollars while holding the exchange rate unchanged. At the same time, the Mexican government found it more and more difficult to roll over its debt, despite offering higher and higher interest rates.

As so often happens in financial markets, these negative effects became self-reinforcing. As investors feared that the exchange rate might fall, they moved into dollars and drove the government’s reserves down still further. This in turn made a peso decline more likely and exacerbated fears of a government default. The promise to repay Tesobonos with however many pesos were required to keep investors whole in dollar terms came to look less and less credible. With the foreign exchange reserves running out, the authorities made a last-ditch attempt to save the fixed-exchange-rate system with a partial devaluation, but that didn’t stem the tide. Domestic capital continued to flee, foreign market confidence plunged, and the government was forced to let the exchange rate float freely. Market attention shifted to the huge quantities of Tesobonos coming due in the weeks and months ahead. The demand for new bonds had dried up. So the government would have had to flood the market with pesos to pay off the maturing Tesobonos—which would send the exchange rate down further.

The Mexican crisis is usually viewed as a failure of Mexican policy. But it was, crucially, also a failure of discipline on the part of creditors and investors—a point about crises that would become very important a few years later, when we faced the return of the same kinds of problems elsewhere and on an even larger scale. Lured by the prospect of high returns, investors and creditors hadn’t given sufficient consideration to the risks involved in lending to Mexico. Once investors became nervous, however, their reaction was swift and unforgiving. Mexico promptly lost access to international capital markets and couldn’t refinance the short-term Tesobonos. Most observers believed that in the long run Mexico would be able to repay its debts. But in the short run, with less than $6 billion left in foreign currency reserves and almost $30 billion in dollar-indexed bonds coming due in 1995—$10 billion in the first three months—Mexican and foreign investors wanted out. For better or for worse, there’s no international law enabling countries to reorganize their debts in bankruptcy court. Thus, our declining to intervene would likely have led to the default of a country that mattered to us in many ways.

Mexico is a good example of a situation—often encountered by policy makers as well as by those in the private sector—in which all decisions had the potential for serious adverse consequences and the key was to find the least bad option. In this case, the dangers of not acting were severe economic duress in Mexico, a contagious decline in emerging markets, and a setback to American growth and prosperity. The risk of acting was failure—potentially endangering repayment of billions of dollars of taxpayer money—or, if we succeeded, moral hazard. Alan, Larry, and I all opposed making the holders of Tesobonos whole. But we concluded—I think rightly—that Mexico couldn’t be rescued without the side effect of helping some investors.

We also worried that the Mexican crisis could affect the global movement toward trade and capital market liberalization and market-based economic reforms. NAFTA had just gone into effect on January 1, 1994. If Mexico went into default a year later, in part for failure to properly manage the influx of foreign capital, the case for further reform might be set back in the United States and abroad. Larry, who had served as chief economist at the World Bank before joining the Clinton administration, was especially concerned with this problem. “Letting Mexico go,” he argued, would send a discouraging signal to other developing nations—such as Russia, China, Poland, Brazil, and South Africa—that had been moving forward with market-oriented reforms. Though we took turns playing devil’s advocate, Larry, Alan, and I all came to a rough consensus in the days before my swearing in. All of us came to think that the risks of not acting were far worse than the risks of acting. Alan captured all of our views when he called a support program the “least worst” option.

On the afternoon of January 10, the three of us, joined by a number of others, including my successor at the NEC, Laura D’Andrea Tyson, had our last meeting to confirm our recommendation to the President while waiting for my confirmation papers to arrive. Larry and I shared Alan’s view that we should put up a substantial amount of money, significantly more than we thought would be needed. In this, we were employing a corollary to Colin Powell’s doctrine of military intervention. The Powell Doctrine, which became well known during the Persian Gulf War, says that the United States should intervene only when American interests are at stake and that intervention must be with an overwhelming level of force.

Of course, no one could say with certainty how much force would be needed to overwhelm the problem in Mexico. One benchmark was the total value of the outstanding Tesobonos, which at that point was about $30 billion. Even that might not be enough, taking into account other government debt, the external debt of Mexican banks, and the potential for “capital flight” as domestic holders of pesos converted them into dollars. Knowing the IMF would also put up a significant amount of money, we proposed $25 billion in U.S. loan guarantees—which had the same financial risk to our government as loans but with some technical advantages.

No “right answer” or formula can exist for how much money is enough in such circumstances, because restoring confidence is a psychological matter that varies from case to case. In this instance, Tesobonos were on the minds of market participants and we decided to make available more than we thought Mexico would actually need. Like a big military arsenal, a large financial one can make a considerable psychological difference to the markets. If investors believe that a government has sufficient resources to right itself and that reforms are in place to deal with the underlying problems, the outflow should stop.

   

WHEN WE GOT THE MESSAGE that my confirmation papers had arrived, Larry and I bundled up our notes and hurried over to the White House. We could not have been bringing the President a more difficult decision at a worse time. Only nine weeks before, he had been dealt a severe political blow—the Democrats had lost both houses of Congress for the first time in forty years. Newt Gingrich and his Contract with America were gracing every magazine cover, and President Clinton was fighting to reestablish himself politically. And here we were, coming into his office on January 10 asking him to make what was likely to be an unpopular and politically risky decision—which also had a real risk of not succeeding—based only on the policy merits.

As usual, it didn’t take Bill Clinton long to grasp the situation. He had a few questions for Larry and me. Is there a real risk of cataclysmic consequences if we don’t do this? Clinton asked. We said yes. Second, the President wanted to know whether there was a good chance our program could prevent those consequences. While there was no guarantee of success, I repeated, the chances were good. Finally, the President asked how much money we could lose if the rescue didn’t work. Larry explained that the loan guarantees would be offered in increments of about $3 billion at a time. If the medicine didn’t seem to be helping, we should be able to stop our losses short of the full $25 billion.

Once he heard our analysis and the seriousness of the situation, Clinton responded without hesitation that he would have to live with the political hazards. “This is what the American people sent us here to do,” he said. I also remember the President saying that he wouldn’t be able to sleep at night if he didn’t come to Mexico’s aid. Often, when I’ve heard criticism of Bill Clinton as indecisive or driven by politics rather than policy, I’ve remembered and cited that night as a response. He gained nothing politically by helping Mexico and risked much at a time when his political capital had already been greatly diminished.

When our discussion was done, Clinton walked over to his desk, picked up the phone, and asked to be connected to the congressional leaders of both parties. Within a couple of hours, Senators Bob Dole (R-KS) and Tom Daschle (D-SD) and Representatives Newt Gingrich (R-GA) and Richard Gephardt (D-MO) all promised to back his emergency request for the loan guarantees. Larry and I went to see them all on Capitol Hill the next day, to solidify their support. At the outset, even Alfonse D’Amato (R-NY), the new chairman of the Senate Banking Committee, who had been investigating Clinton relentlessly over Whitewater, was supportive. D’Amato said we ought to put up more than $25 billion, so that the financial markets wouldn’t think they could “overpower” us. With that encouragement, we increased our proposal to $40 billion. A critical component of the proposal also required Mexico to commit to various economic reforms and to pledge its oil export earnings to assure repayment.

Despite this support from the congressional leadership, the reaction when Alan and I went to Capitol Hill to explain the plan was overwhelmingly negative. Meeting with more than a hundred legislators from both parties on January 13, we got our first taste of just how difficult getting Congress to act was going to be. Several members asked for a promise not to put U.S. tax dollars at risk. Some questions were very sensible but hard to answer. Senator Joseph Lieberman (D-CT) pressed us on why the Japanese and Europeans weren’t sharing the risk with us. I responded that our allies were making their contribution through the IMF. A less diplomatic response would have been that I believed our allies should have also contributed bilaterally, because a crisis in Mexico and possible contagion would have affected them as well. But they weren’t going to, perhaps in part because they considered Mexico our problem but also because they didn’t share our judgment about the global danger a Mexican collapse would create. In any event, none of this changed the fundamental point: acting was in our interest. Afterward, Larry and I went on a full-scale media and political blitz to press our case. Among the calls I made was to the governor-elect of Texas, George W. Bush, who offered his support for our effort. Like many border-state politicians, Bush instinctively grasped what was at stake and became a strong public supporter of our aims and efforts.

I’m still not sure I fully understand the depth of the negative reaction to our package in Congress. At one level, congressional opinion simply mirrored public opinion. Xenophobia may have explained some of this opposition, but many people just didn’t see any need to risk our tax dollars on this effort. Perhaps we could have done a better job of making the case. But the situation was probably too novel and too complicated to be assimilated quickly. In 1995, the notion that a poor country’s macroeconomic miscalculations could affect the largest economy in the world simply didn’t register with a lot of people. A few years later, when the Asia crisis took hold, it still didn’t. For most Americans, the global economy remains an abstraction, with little meaning in their daily lives.

The opposition we encountered in Congress also seemed to reflect entanglement with other issues. Many Democratic legislators had bucked their supporters in organized labor to vote for NAFTA. Now the opponents of NAFTA were taunting them—look, we told you so. On the Republican side, some of the new group of highly energized freshmen were eager to fight the President and skeptical of international engagement. Why help a country that sends us narcotics and illegal immigrants, especially when that help would benefit Wall Street at the same time? I tried—with little success—to explain that our purpose was not Mexico’s or Wall Street’s well-being but America’s. A Mexican default would exacerbate the very problems they were concerned about. But my arguments made little headway.

Some members of both the House and Senate—such as Senators Chris Dodd (D-CT), Paul Sarbanes (D-MD), and Robert Bennett (R-UT)—understood the issues and worked to help us at many critical junctures. But most members willing to support the package wanted conditions that were either politically impractical or not germane to reestablishing stability, or that simply couldn’t be worked out with Mexico as part of this program. For example, some Democrats insisted on new labor standards to protect Mexican workers. Jim Leach (R-IA), the conciliatory and internationalist-minded Republican chairman of the House Banking Committee who supported our proposal, was willing to accept some Democratic demands as the price of passage. But that incensed Leach’s colleagues, who didn’t see why Democrats should be calling the shots now that Republicans ran Congress. Some of them said they wouldn’t support any rescue package with labor standards. This messy conflict provided a foretaste of future battles over globalization, including trade liberalization. The constituency for free trade wasn’t large to begin with—but if we were going to have it, everyone wanted his particular interests protected.

As the negative reaction mounted, congressional leaders who had agreed to support us at the outset seemed to grow more wary. They weren’t persuading skeptical colleagues and appeared to be reducing their efforts to do so. Even some of our committed backers seemed worried about looking too enthusiastic. One prominent supporter kept sending us letters raising “concerns” about our proposal. I later realized that this was a paper trail qualifying his endorsement that he could point to if we failed. But for Greenspan’s credibility, the reaction from the GOP would have been even more negative.

One legislator who did grasp the full dimension of the problem was the new Speaker of the House, Newt Gingrich. Gingrich was concerned enough about populist opposition to the rescue that he asked Alan to phone Rush Limbaugh on his behalf, which Alan did. When Larry and I went up to Capitol Hill to meet with him the first time, Gingrich really seemed to get it. Toward the end of our meeting, he described Mexico as “the first crisis of the twenty-first century.”

GINGRICH MAY NOT HAVE BEEN the first to use such a phrase—the IMF’s Michel Camdessus described the crisis using the same words—but he effectively captured the reality we faced. Many elements of the Mexican crisis had been present in previous events, such as the Latin American debt crisis of 1982. Then, as in 1995, the Mexican government essentially ran out of foreign reserves. In 1982, Mexico’s default triggered an economic decline that spread throughout much of Latin America and beyond. Banks that had lent heavily to developing countries in the preceding years pulled back dramatically, pushing one country after another into default. We did not want Mexico’s difficulties this time around to precipitate another global debt crisis. But the world had changed over the preceding dozen years in ways that made this crisis very different and, in some respects, even more dangerous and difficult to contain. The international financial system had grown in scale, complexity, and velocity, so that the developed and developing worlds were now tied together as never before. Simply put, the potential for financial contagion across developing countries seemed considerably greater, and their economic health affected ours in more complex ways.

The most obvious change was the growth of international trade with developing countries. Many people don’t realize that these countries purchase 40 percent of our exports. As a result, millions of American jobs now depend on the ability of consumers in the developing world to buy what we produce. Capital flows have increased even more dramatically. It’s no longer just banks but investment banks, endowments, pension funds, mutual funds, and, through them, retail investors who have assets in the developing world. In the twelve years from 1982 to 1994, private capital flows to emerging markets had increased more than six times, from $24 billion to $148 billion. I had been at Goldman Sachs during the earlier crisis in 1982 and it had barely registered with me. The 1995 Mexican crisis was a high-profile event throughout the financial system.

By 1995, global finance had become immensely more complex than in 1982, as emerging-market debt shifted from banks to widely held securities. The 1980s debt crisis took considerable arm-twisting by the United States and other governments to make commercial banks renegotiate their bad loans to Mexico and other debtors. But that was in many ways a large-scale version of a “workout” session that banks hold all the time with troubled borrowers. With Mexico in 1995, some people proposed that we again “coordinate the banks,” but the banks were now far from the only creditors. In place of bank loans, a vast variety of debt instruments and derivatives had been devised in the intervening years. Mexican debt was diffused, with bearer bonds—which are not registered in the name of the owner, thus making the owners difficult to identify—held privately by various institutional and individual investors all over the world. In addition, portfolio investors held stock in Mexican companies, which few had in the early 1980s.

With many participants in the financial system, including the big investment banks, holding an array of emerging-market securities, a financial crisis in Mexico could spread much more widely and less predictably, creating a potentially powerful ripple effect. People facing large trading losses in one emerging market might suddenly decide that other emerging markets seemed more risky and liquidate positions in all their securities, even if the countries were apparently unrelated, such as, for example, Mexico and Poland. Firms might also have to raise capital to cover the initial losses, forcing the sale of other positions. Massive downward pressure could develop in other developing markets and even create pressure in industrial-country markets.

A final change was the extraordinary acceleration of market reactions. Throughout most of the 1980s, emerging-market sovereign debt had been illiquid, changing hands only in privately negotiated transactions with large point spreads. In 1995, highly liquid capital moved at the speed of light through fiber-optic cables. Traders had an array of terminals on their desks, with complete information about all prices at all times. Orders could be executed at any hour. The result was that developments in markets in one place could have instantaneous effects in any other place, and crises could spread much more rapidly.

The combination of these factors made the Mexican crisis different in kind from anything anyone had experienced before and made Gingrich’s phrase memorable. Almost as soon as the crisis broke, I began picking up reports that the loss of confidence was affecting markets as far from Mexico City as Warsaw and Bangkok. There was no rational economic tie between Mexico’s liquidity crisis and the Eastern European financial markets. But the psychology of markets is that investors who are far too complacent one day may quickly change and become a stampeding herd the next. In a world of instantaneous reactions, the tendency to react rather than to think is not necessarily irrational. In the race to an exit that not all will fit through, speed can be lifesaving.

In describing what was happening, I found myself trapped in a kind of Catch-22. On the one hand, I needed to underscore the dangers in order to motivate reluctant legislators—and the public—to support our rescue package. On the other hand, frank talk about what might happen could provoke the very reaction we most wanted to avoid. Explicitly raising our worst fears about global contagion could create a self-fulfilling prophecy. Some people at the time pointed out that I lacked the experience of my predecessor, Lloyd Bentsen, in dealing with Congress. That was true but not my real problem. My real problem was relinquishing our strongest tool: fear. The only way to navigate the twin hazards of complacency and panic was by choosing my words very, very carefully, softening concerns and using calculated ambiguity.

When Alan Greenspan, Secretary of State Warren Christopher, and I testified before the House Banking Committee on January 25, the hostility was typical of the whole process. I had to answer charges that our proposal was a bailout for Wall Street and big investment banks disguised as help for a neighbor. Bernie Sanders (I-VT) said I should “go back to your Wall Street friends, tell them to take the risk and not ask the American taxpayers.” I tried to explain that I wouldn’t spend a nickel of taxpayer money for the sake of rescuing investors. Again and again, I returned to my arguments that our proposal to help Mexico was driven by our national interest. These numbers are always hard to calculate, but we made a rough judgment about the potential costs of a prolonged Mexican crisis to the United States—700,000 jobs affected, a 30 percent increase in illegal immigration, and so on.

Ross Perot, who testified to the Senate the following week and danced on what he took to be NAFTA’s tomb, received a much warmer reception. And Senate Banking Committee chair D’Amato had gone from supportive to antagonistic. It was around this time that Pat Griffin, the White House’s highly capable liaison with Congress, expressed annoyance with me at a meeting in the chief of staff’s office for putting the President in a box. He felt that the decision to help Mexico had been made without adequate focus on the political risks and had left Clinton in an untenable position. I answered that the President had understood the political risk and decided to take it. As markets began to recognize the extent of congressional opposition to our proposal, they weakened further, not only in Mexico but also in Argentina, Brazil, and other emerging-market countries that tended to move in sympathy. And again, emerging-market countries as far away as Asia and Eastern Europe were affected.

In my office at Treasury, we embarked on a constant process of analysis and discussion. Included in our regular meetings was a group of officials who would become the core Mexican team: Jeff Shafer, David Lipton, and Tim Geithner, as well as my chief of staff, Sylvia Mathews, and Dan Zelikow, who became head of our Mexico task force. I got in the habit of referring to this group as “we” and “us” because in most cases our decisions were reached together after long days and nights of vigorous exchange of views. The Fed’s top international official, Ted Truman, often joined us, as did Greenspan at important moments. Treasury and Fed officials were in turn consulting closely with top IMF officials, especially Michel Camdessus and his highly respected deputy, Stanley Fischer, a former chairman of the Economics Department at MIT.

Our Treasury meetings were characterized by searching questioning and debate, all for the sake of the fullest possible exploration of alternatives. This was a discussion, rather unusual for Washington, in which rank hardly mattered. A thirty-four-year-old deputy assistant secretary and the Treasury Secretary both felt fully entitled to express their views. That informality reflected my experience both on Wall Street and inside the White House about what kind of discussions tended to be the most illuminating and productive. So if someone, particularly someone junior, who was often closest to an issue, seemed to be holding back, I tried to draw out his or her view. What mattered to me was the merit of the argument, not the title of the person who made it.

Meetings produced the best results if those who disagreed with the accepted view were encouraged to speak out. So if a meeting seemed to be moving toward a consensus, I would make a point of soliciting dissenting views. Disagreeing with me was socially approved rather than discouraged. If no one disagreed, I would encourage someone to play the role of devil’s advocate. I might say, “This is where we’re heading, but we need to know the contrary view so we can consider it.” And I, or someone else, would take up the other side. Just as important as the freedom to disagree, I think, was that this group of high-powered intellects in large measure avoided investing their egos in their arguments. It was a common search for the best answer in the midst of a worsening crisis.

As congressional opposition solidified, our group naturally began to consider alternatives. One possibility was acting unilaterally, without a vote by Congress, by drawing from the Exchange Stabilization Fund (ESF), the pot of money that the Treasury uses for currency interventions. Congress created the ESF at the time of America’s departure from the gold standard in 1934 to allow the Treasury to stabilize exchange rates. At that time, no one envisioned a crisis like Mexico’s, but in our view responding to the Mexican crisis fit within the purpose of the ESF. The fund had about $35 billion and, as Treasury Secretary, I had considerable discretion over when and whether to use it, subject to the President’s approval.

Senator Bob Bennett—who supported us throughout this crisis—had suggested the ESF early on, but we had initially decided not to use it in part because we thought Congress should be involved in a decision of this magnitude for the country. However, as Congress clearly showed no disposition to endorse our decision, we belatedly focused on the ESF as a potential alternative. Some in Treasury captured one problem in this approach with the phrase that the ESF was “a weapon you could use only once.” Our concern was that members of Congress might be so outraged by unilateral action that they might legally disarm us from using the ESF again. But with the package we called “Mexico I” struggling in Congress, the option of tapping the ESF gained force.

As this discussion continued at Treasury, I was in frequent contact with my Mexican counterpart, Guillermo Ortiz, who had been brought in after the managed devaluation had failed in December. I had gotten to know Ortiz a bit in the late 1980s, when, as an official at the Hacienda, as the Mexican Finance Ministry is known, he had handled bank privatizations. Like many of Mexico’s senior economic officials, he was a highly capable economist, with a Ph.D. from Stanford University. A thin, serious fellow with a taut demeanor, he looked even thinner and more serious than I remembered when he visited the Treasury as the newly installed Mexican finance minister. Ortiz was not given to overstatement or self-dramatization, so when he told me on January 28 that despite the formal announcement of a $7.8 billion IMF program two days earlier, the situation was worsening, I took him very seriously. More than a billion dollars’ worth of Tesobonos were coming due the following week, and the Bank of Mexico’s currency reserves were running out. That meant default was getting very close.

Tony Lake, the President’s national security advisor, had deputized Sandy Berger, his number two, to take the lead for the National Security Council (NSC) on dealing with Mexico. That evening, Sandy, Leon Panetta, and I met in Leon’s office in the West Wing. After some discussion, we decided to press ahead with our effort to rally congressional support, setting a deadline of Monday, the thirtieth. On Sunday, when he got back from church services, Clinton again made calls to leaders in both parties. We still thought we might be able to convince Congress to act.

I woke up on Monday with a sense of deep concern. As I had feared, the Mexican markets began to sell off sharply, with the peso dropping almost 10 percent to more than 6 pesos per dollar, its lowest level yet. We had originally assumed that Mexico could remain solvent at least through February. But despite the assumption many people make that government has better information than the private sector, the opposite is often true. Mexico informed us that its reserves had fallen to around $2 billion on the same day the International Herald Tribune reported it. That could have meant a generalized financial collapse within days in Mexico.

That evening, Sandy, Leon, Larry, and I again gathered in Leon’s office. As we considered our options, Gingrich phoned from Capitol Hill with bad news. The best-case scenario, in his opinion, was that congressional passage would take another two weeks. A few minutes later, Guillermo Ortiz called from the Hacienda and spoke to Leon. He delivered a message we already understood: The Mexicans were out of rope; we were the only hope. At about 11:00 X, the President, just back from a fund-raising dinner, joined us in Leon’s office. As the meeting had stretched into the evening, someone had sent out for Domino’s Pizza. The President, still in his tuxedo, gazed longingly at the grease-spotted pizza boxes. The Secret Service didn’t like him to eat food brought in from the outside. The rest of us took our chances.

Once again, Larry and I presented the possible consequences of a Mexican default to the President. We proposed abandoning the effort to get loan guarantees through Congress and instead tapping the Exchange Stabilization Fund for loans. Partly because the ESF had only about $35 billion and needed a cushion for other possible needs, and partly because the IMF was willing to increase its contribution, we lowered our U.S. proposal to $20 billion. Michel Camdessus, in a moment of daring unusual even for him, had promised to find or provide another $10 billion on top of the $7.8 billion the IMF had previously committed. That brought the total amount available, including a bit more from other sources such as Canada, to just under $40 billion. Sometimes the press referred to our proposal as a $50 billion package, but we avoided using that number because it included other short-term contributions from the Bank for International Settlements that Mexico couldn’t practically use to pay off Tesobono holders or finance imports.

The essential goal remained the same: to allow Mexico to restructure its debt from short term to long term and to implement reforms in order to reestablish financial stability and regain access to private capital. By acting on the basis of executive authority, we would avoid the problem of satisfying the immense range and number of conditions proposed in Congress. However, as under the previous proposal, Mexico would have to agree to significant policy reforms negotiated with us and with the IMF, including stronger fiscal and monetary policy, important structural measures, and fuller and timelier reporting about its financial condition. The Bank of Mexico would also be required to pay significant interest on the loans and to make revenues from oil sales available to the United States in the event of nonpayment.

This time we were even more specific with the President about the political risk. A poll published in the Los Angeles Times a few days before had showed that the American public overwhelmingly opposed our efforts to help Mexico, by a margin of 79 to 18 percent. I cited these numbers. And I stressed, once again, that the plan might not work. No precedent existed for action of this kind on this scale, and none of us could predict with substantial confidence what would happen. All the choices were bad. But the alternative to intervention remained much worse.

Despite the way opposition had solidified, Clinton’s hesitation was no greater than the first time we had gone to him, three weeks earlier. “Look, this is something we have to do,” he said. Once again I was deeply impressed not only by his willingness to take on a big political risk but by how relaxed he seemed about doing so. Leon Panetta also made the point later that Clinton seemed to welcome being able to do something difficult and important for the country on his own.

After the meeting, I went back to my Treasury office and called Alan to relate the President’s decision, which Clinton discussed with the four congressional leaders at the White House early the next morning. Because the decision was made when it was the middle of the night in Europe, we didn’t have time to consult most of our Group of Seven (G-7) allies in advance of announcing our new proposal. The next day, they were furious at Camdessus for offering another $10 billion without consulting them, and very upset with us too. Six countries were angry enough to abstain from the official IMF vote of approval. When German Chancellor Helmut Kohl met with President Clinton a few days later, he said the G-7 finance ministers were all irritated with me and that I should send them each a bottle of whiskey as a peace offering. I didn’t send any whiskey, but Larry and I did try to make amends for the lack of consultation, both over the phone and when we met our G-7 colleagues at a meeting in Toronto a week later. Relations were repaired, and all eventually endorsed the IMF program.

For me, it was the first of many experiences of dealing with the ambivalence of our allies about U.S. leadership and the difficulties of exerting that leadership effectively with sovereign states that have agendas and political needs of their own. The lesson I took from that episode was the great importance of working with other countries to build support for what we thought was the right path on international policy. Thereafter, we spent considerable time and effort consulting with our counterparts around the world, especially through a process that brought Larry together with the deputies of the other key finance ministries.

   

THE PESO and the Mexican stock market, the Bolsa, climbed 10 percent following our announcement of “Mexico II” on January 31. Markets in Brazil and Argentina moved in sympathy. But the respite didn’t last long. As we’d feared, some in Congress were furious that we’d made what amounted to the largest nonmilitary international commitment by the U.S. government since the Marshall Plan without their consent. And Mexican markets, realizing that Congress might not allow us to proceed, soon resumed their decline.

When I went up to Capitol Hill to testify before the House Banking Committee a week later, members of both parties blew off steam. Representative Maxine Waters (D-CA) asked whether the bondholders who would be made whole were my “Wall Street buddies.” Beyond the hearing room, the reaction was even harsher. A group of Republican freshmen in the House tried to find a way to forbid us from extending loan guarantees without congressional approval. Although Gingrich was still personally supportive, he clearly understood the political realities well enough to conclude that he couldn’t turn his caucus around.

In retrospect, it seems to me that many members of Congress probably meant to oppose us without actually stopping us. They didn’t want to be blamed for failure. Gingrich was quoted in one newspaper article telling Panetta that if the President took responsibility for the rescue plan, he would hear a “huge sigh of relief” from Congress. The legislators understood what needed to be done but didn’t want to have to vote for it. But even such halfhearted opposition was not without cost. Attempts to criticize the program without actually stopping it created market concern that the program was at risk, thus working against the objective of reestablishing confidence and perhaps putting more taxpayer funds at risk than would otherwise have been necessary.

Other attacks were truly meant to stop us and were getting quite ugly, especially a concerted effort at personally vilifying me. One rumor was that I had a secret account somewhere holding Mexican securities. The nastiest official statements came from a Republican freshman congressman from Texas named Steve Stockman, who accused me of various “suspicious” conflicts of interest and called for an investigation into whether I had arranged the rescue package for my own benefit. He said Goldman Sachs, which had underwritten the privatization of some of Mexico’s nationalized industries, now might face liability from investors who lost money there. When I had joined the administration in 1993, my equity in Goldman Sachs had been converted into debt, and I had gone far beyond the requirements of the Office of Government Ethics—and paid a significant amount of money—to neutralize my position so that I had neither benefit nor risk tied to the success or failure of the firm.

To me, the attacks were an illustration of how harsh and ugly the political process had become. Critics weren’t content to disagree with our policies; they impugned my motives and asserted hidden conflicts. At that point, I was still somewhat surprised that opponents would make ad hominem attacks as a way of dealing with policy disagreements. As time went on, I came to recognize that, to some extent at least, Washington unfortunately functions this way.

In February, the daily reports Dan Zelikow prepared on Mexico made for gloomier and gloomier reading. Alfonse D’Amato, by now a relentless critic of ours, said our approach had “all the potential of being a very real debacle.” Bob Dole, who had signed a statement favoring our use of the ESF, was also in the process of revising his position, albeit more quietly. In a word, the political situation looked grave.

Guillermo Ortiz spent much of early February in Washington. He was negotiating the details of the new IMF program with Stan Fischer and others at the Fund, and also negotiating with Larry and his team on our bilateral program, which required Mexico to follow the IMF program and had some additional elements. Meanwhile, Mexican markets were uneasy and deteriorating. Ortiz was spending days and nights talking to the IMF and to us about conditions and difficult policy measures, all the while wondering whether he could sell the agreements to a suffering public back home. He looked ashen and exhausted. As great as the stakes were for us, we could only imagine what the crisis felt like for him.

A few of our own officials with primary responsibility for the problem also showed the pressure in their faces. I remembered similar stressful reactions from traders at Goldman Sachs when losses mounted. At one point Jeff Shafer, who was conducting our negotiations along with Larry, looked so distressed that I told him what I had told traders many times in the past—that a thousand years from now none of this was going to matter much. I told Jeff that he was extraordinarily capable and doing his best in difficult circumstances—and that was all anyone could ever do. One way or another, we’d make our way through this.

One source of remaining uncertainty was the new president of Mexico, Ernesto Zedillo. Zedillo was an economist, educated in both Mexico and the United States, with a doctorate from Yale. But I had not met this man to whom we were about to lend $20 billion, and no one in the administration knew him well. We didn’t have a sufficient sense of how committed Zedillo would be to following through with the difficult reforms that were going to be required for the program to work. And we needed to be sure that Ortiz was speaking for him in all cases in our negotiations. So in a phone conversation with Zedillo, I proposed sending Larry down to meet with him. Zedillo thought that was a good idea.

This trip involved dilemmas of substance and perception in both countries. At a substantive level, our economists had a series of proposals to reform aspects of Mexico’s economic policy and reestablish confidence. But the program would never work if we imposed these measures. We had to reach a meeting of the minds with Mexican officials, and they had to take ownership of the program. The problem of public perception was related but distinct. On the one hand, we didn’t want the Mexican public to feel we were infringing on their sovereignty. On the other hand, we wanted the American public to feel precisely that we were imposing strict economic conditions on Mexico to protect taxpayer money. The tension between these contradictory demands pervaded our discussions. We spent countless hours fine-tuning the wording of our public statements to avoid erring in one direction or the other.

We solved the perception problem around the trip with a cloak of secrecy. We put in a presidential order for an Air Force plane, and before dawn Larry and David Lipton were off on a mission. We made every effort to keep their trip quiet, and luckily no one saw them slipping in and out of Los Pinos, the President’s dwelling in Mexico City. More important, our substantive concerns were eased. Larry came back deeply impressed with Zedillo. The new president of Mexico understood exactly what he was doing. Moreover, Zedillo clearly had full confidence in Ortiz. We didn’t need to worry about his negotiator getting out ahead of him.

Furthermore, President Zedillo was firmly committed to economic reform despite the difficulties that lay ahead for his country. The single most important aspect of this reform was interest rates. The IMF had negotiated that the interest rate would be about 2 percent per month. However, inflation was expected to be 4 or 5 percent or higher. We wanted to restore confidence in the peso and knew that no one would hold pesos if their position lost value because the rate of return was negative in real terms—that is, if interest rates were lower than the rate of inflation. But the Mexican team negotiating in Washington had rejected higher interest rates. In the meeting with Zedillo, Larry raised this problem after forty-five minutes of polite conversation on the full range of issues regarding the rescue. Zedillo thought for only a moment, then said, “I spent my whole career at the Bank of Mexico writing articles saying that Mexico should have positive real interest rates. Now is not the time to abandon that idea.” Although some critics take issue with the need for high interest rates in a financial crisis, this approach was absolutely essential in Mexico for two related reasons. It created confidence that credible policies were now in place to restore stability and, in the context of that confidence, it offered investors an attractive rate of return to induce them to hold pesos.

On February 16, I hosted a dinner at the Jefferson, the pleasant old hotel on Sixteenth Street that served as my home in Washington for six and a half years. In a private dining room at the back of the restaurant, Panetta, Berger, and the rest of our group at Treasury convened for one last examination of the program about to begin. In a few days, I was expecting to sign an agreement that would commit us to lend $20 billion to Mexico. Though we retained the power to withdraw unilaterally at any point, this was our last real chance to change our minds.

In previous discussions, Larry had laid out the analysis—and the risks—very clearly. In theory, if Mexico offered a high enough interest rate, then people—whether ordinary Mexicans or foreign investors—would choose to hold pesos rather than buy dollars at a particular exchange rate. The greater the confidence that the government’s policies and IMF-led financial support would succeed in restoring financial stability, the lower the interest rate that would be needed to persuade investors to hold pesos. But if people feared that the program would not work—that their pesos might quickly lose their value as the exchange rate plunged further and inflation accelerated, putting more pressure on the peso and creating a vicious cycle—they would demand a terribly high interest rate to compensate for that risk. And as interest rates climbed, we might reach a point at which higher rates could actually become counterproductive in attracting capital. Rather than attracting capital and increasing the demand for pesos, higher rates could reduce the demand for pesos by threatening to push the government’s debt burden to a level where default seemed inevitable, or trigger a collapse in the already weak banking system. In that case, the plan would fail and our billions of dollars in loans from the ESF would merely have helped finance some of the capital flight as money poured out of Mexico. Larry, Alan, the rest of the team, and I had spent endless hours trying to gauge how high interest rates could go without being too high. But what if no interest rate existed that was high enough to attract capital before rising above the level that would scare capital away?

Interest rates were the most critical issue, but other policies were also crucial to reestablishing confidence and therefore growth. These included a commitment to a floating exchange rate to avoid a rerun of the previous crisis; a budget plan that showed that the government could tackle its debt burden; reform of the banking system, revealed by the crisis to be close to insolvent; and much greater transparency so that investors felt adequately informed. The more credible the government’s commitment to a strong reform program, the less pressure there would be on the exchange rate and the more leeway Mexico would have on interest rates.

As we sipped our coffee at the Jefferson, I went around the dinner table and asked all present what they thought the odds were that the plan would work. Dan Zelikow, who had seen real economic dysfunction as an adviser to the first democratically elected government in Albania, thought the odds of success were only one in three. Larry thought our chances were substantially better but didn’t offer precise numbers. David Lipton gave the most optimistic specific prediction: better than a 50 percent chance. What was striking was that everyone agreed we were taking a significant risk.

In a sense, the plan had two distinct, but intertwined, risks. The first was that the Mexican government would simply be unable to follow through on the tough steps needed to rebuild confidence and attract private capital again. The second was that official money—from the United States and the IMF—would not be sufficient to provide the breathing space needed while policy reforms took hold. Ironically, the bigger and more certain the promise of official money, the less was likely to be needed.

Making matters more complicated, our G-7 allies were still protesting Camdessus’s decision to add another $10 billion from the IMF. In response, Camdessus suggested restructuring our deal. On the morning of February 21, the day for signing our agreement with Mexico, Camdessus told me that the IMF could provide only $7.8 billion, plus contributions from elsewhere. That was inconsistent with his original commitment; now he wanted to go ahead with the extra $10 billion only if it came as bilateral loans from other countries, similar to our ESF commitment. That was a problem for us, since Mexico needed to have the entire $17.8 billion available and I had always told Congress that the total IMF contribution would be $17.8 billion. I sympathized with Camdessus’s difficult position. But for him to do this now would harm the program and seriously undermine our credibility in Congress.

Camdessus had provided strong leadership in difficult circumstances and I had great respect for him, but this wouldn’t work. With Leon sitting in my office, I called back and said, “Michel, this is not what we agreed to. And if you insist, I am going to go out and make a public statement. We are going to hold a press conference and announce that you have changed the deal. And I’m not going to go ahead with the Mexican program.”

Michel said, “You can’t do that.”

I replied that, in fact, we could. The moment was dramatic, but in the end, Camdessus came around, and our strong relationship with him—so important in the years ahead—was not harmed. My approach in general is to try to see both sides and work to find common ground. But sometimes there is no good alternative to an adamant stand calmly taken.

We signed the Mexican rescue agreement as planned that day in the Cash Room at the Treasury Building. The Cash Room was where citizens once came to trade paper dollars for gold. The location seemed appropriate, since the closing of the gold window and the creation of the ESF in 1934 had made the action we were about to take possible. But we were all concerned. After the signing, I walked back to my office in worried silence with Sylvia Mathews and David Dreyer, another senior adviser. David tried to cut the tension with humor. “I guess we’ll never see that money again,” he joked. Sylvia and I didn’t smile. It’s funny to me now, but it wasn’t then.

A night or two after that, when the positive market reaction to the agreement had already dissipated and markets were once again dropping, Larry came into my office and offered to resign. It was about eleven in the evening, and we were both still at work. Larry felt personally responsible for an effort that might well fail. I told him that his talk about resigning was ridiculous. While I understood how Larry could feel his responsibility so keenly, I told him he wasn’t any more responsible than the rest of us and he was taking the matter much too personally. What we were doing was right, and we were all in it together. We’d just have to hang on and get through it, one way or the other.

In the next few weeks, we all felt the pressure. Jeff Shafer told me a story somewhat later about having a drink with friends before a baseball game at Camden Yards in Baltimore on a rare evening off. When a friend asked him something about the Mexican “bailout,” the term that most irritated us, Jeff’s response—“It was not a bailout!”—was loud enough to stop conversation in the crowded bar.

I didn’t discuss my own feelings with anyone at work, but I too had focused on what the possibility of failure could mean for me. Losing $20 billion in public funds, especially on such a controversial and high-profile matter, could substantially taint how I would be seen as the Secretary of the Treasury. But even if I had to step down, I could deal with that. I felt better thinking that I’d helped set up the National Economic Council at the White House, which was working well. No one could take that away from me, no matter what happened afterward.

As markets continued to fall, Larry and I had a difficult phone call with Guillermo Ortiz. This was after we had signed the agreement but just before the first disbursement of funds. As we explained how bleak the situation looked, Guillermo, though sounding overwrought, tried to paint a rosier scenario for us. We weren’t persuaded, but I understood he could do little else. After the call, we went right over to the Roosevelt Room in the White House for a meeting with Panetta and Berger. I felt, in light of the circumstances, that we had an obligation to raise the question of whether to exercise our right to withdraw from the arrangement unilaterally.

“Letting Mexico go” at this stage would turn the possibility of default into a virtual sure thing—but I thought I should raise the issue even though I personally believed we should still proceed. My question was greeted with surprise. Only Erskine Bowles, the deputy chief of staff, who, like me, had worked as an investment banker, related to why I was even posing the possibility of not following through on this program we had already agreed to. Leon said that he didn’t think that option was viable. The administration was committed to a plan of action and had to stick with it even if the chance of failure had increased. The cost to the administration of reversing course—in terms of lost credibility—would be enormous. I, on the other hand, imagined the congressional hearing where I’d be called to account, with one of our very vocal critics leading the inquisition. So, Mr. Secretary, you thought that there was only a small chance that sending billions of dollars of American taxpayers’ money would help? And you sent the money anyway?

That discussion illustrates a difference between making decisions on Wall Street and in government. There is a strong impetus to stick with presidential decisions, even when circumstances change, because the world is watching to see if you keep your commitments. Credibility and reliability are powerful values. Thus, changing direction may sometimes be worse than proceeding with something that could be wrong. In the private sector, reliability and credibility are also very important, but you can change course much more easily. When a Wall Street trader decides to cut his losses or a corporate head cuts back in a troubled business, no one complains about inconsistency. Nonetheless, there are times when high-profile government decisions should be reversed despite the damage to credibility. I didn’t think that was the case here, but I did think the issue should be raised.

On March 9, the day we were to release the first $3 billion loan, the peso fell dramatically, closing for the first time at more than 7 pesos per dollar. Rumors—in this case accurate—circulated on Wall Street that we were contemplating not releasing the money on schedule. Despite the commitment of additional funds from the World Bank and the policy reforms that Ortiz was about to announce in Mexico City that evening, we were all deeply concerned that market confidence simply wasn’t going to rebound. But when the time came to decide, we approved the release of the money.

The roller-coaster quality of that period was caught for me by the visit Larry and I paid the next day to the hearing room of the Senate Banking Committee. As I was answering questions, including hostile ones from Senators D’Amato and Lauch Faircloth (R-NC), my staff kept slipping me notes about the peso, which was rising even more dramatically than the prior day’s fall. Larry, who was testifying alongside me, passed me a note saying, “I think this thing might actually work.” While one of the senators was talking, I scribbled back a response: “I think it might.” Once again, though, our optimism was short-lived. The March 10 rally was followed by a steady decline. A month later, we went through the same agonizing decision again about whether to disburse the second $3 billion loan.

By mid-May, the Mexican central bank data we saw showed the first, very tentative signs that the program was beginning to work, although markets didn’t seem to reflect much progress and still looked fragile. We sent a memo to the President that pointed to some encouraging indicators. The Mexican economy was in a severe recession, but the country’s trade deficit had turned into a surplus, the stock of outstanding Tesobonos had been reduced substantially, and the peso had recovered somewhat. Anticipating the success of our rescue package, Thomas Friedman, the Pulitzer Prize–winning New York Times columnist, described it in his May 24 column as “the least popular, least understood, but most important foreign policy decision of the Clinton presidency.”

Alas, Friedman was getting ahead of himself just a bit. The roller coaster continued for the next few months. With the policy measures imposed by Zedillo, the financial and economic situation looked more stable by the end of the summer. But unemployment was growing, real wages had fallen significantly, and bank balance sheets were severely impaired. Chafing under the duress—and encouraged by signs that the program was taking hold—the Mexican government moved prematurely to lower interest rates. Markets resumed their slide, but the Mexicans quickly reacted and tightened policy to halt the slide.

Despite another rocky period in November, by the end of 1995 the program was taking hold. Investors started to put some money in; foreign exchange reserves started to build up; exchange rates stabilized; interest rates came down a little bit. Everything just started to work. The private sector had begun lending Mexico money again. By the beginning of 1996, the Mexican economy was growing again. The Zedillo government began to repay the U.S. and IMF loans, rolling them over into less conditional private-sector debt.

The speed of the response was remarkable. The 1982 crisis led to what has been called a “lost decade” of negative growth, financial instability, and political and social unrest throughout Latin America. The 1995 crisis caused real suffering on the part of the Mexican poor and middle class—and real wages were very slow to recover—but only one year of economic growth was lost. After the 1982 crisis, Mexico took seven years to regain access to capital markets. In 1995, it took seven months.

In August 1996, Mexico prepaid $7 billion of the $10.5 billion still outstanding from the United States and IMF. When the Zedillo government completed the repayment in January 1997, more than three years ahead of schedule, an anonymous aide of mine was quoted in The New York Times as saying, “This was Bob Rubin’s Bosnia. And today he got the troops out.” Mexico paid us $1.4 billion dollars in interest and left the ESF with a profit of $580 million—the excess over what our money would have earned in U.S. Treasury notes. Senator D’Amato, who had already called the program a “failure,” put out a one-line press release saying he was “pleased” our program had been successful—thanks to “vigilant congressional oversight.”

   

IT SEEMS TO ME that the Mexican crisis has much to teach us about the global economy, new and heightened risks that our country is likely to confront in the future, and the challenges we face in trying to deal with these hazards. These challenges are complicated by volatile financial markets and by our own political processes. I’ve drawn out many of those reflections in the context of my narrative, but a few final observations depend on the whole story.

The first lesson is that our ability to address economic crises beyond our borders is limited. The money we lent to Mexico could not have had the desired effect without the policy choices the Mexican government made. This was the crucial element, both because of the effects of individual policies—especially on interest rates—and because of the confidence engendered by the more amorphous cumulative sense that the Mexicans were serious about getting their act together.

As an episode in public policy making, our decision making in the face of a highly uncertain situation and considerable political pressure showed that the probabilistic thinking that I internalized so deeply in the financial world had real applicability in Washington. And that process was ongoing, as we reevaluated options and policies when the facts changed on the ground in Mexico and in the financial markets. I think, too, that our work demonstrated the value of robust and open intellectual interchange in making government decisions.

Yet in other ways the episode showed me just how challenging decision making is in the context of government. Good decisions are much more difficult to make when disagreement is not just about means but about objectives. The private sector often focuses intensely on customers and employees, but in the final analysis everything comes back to serving the overriding objective of profitability—except perhaps for the relatively small portion of corporate activity devoted to philanthropy and other public purposes. The public sector, by contrast, operates with many equally legitimate objectives. For many in Congress, narcotics and illegal immigration mattered far more than economic issues in dealing with Mexico, and these legislators were not persuaded by our argument that the former problems would get far worse if Mexico defaulted and suffered from severe and prolonged economic duress.

Mexico also demonstrated the difficulties our political processes have in dealing effectively with issues that involve technical complexities, shorter-term cost to achieve longer-term gain, incomplete information and uncertain outcomes, opportunities for political advantage, and inadequate public understanding. Unfortunately, many of the most important economic, geopolitical, and environmental challenges of today’s complicated world fit this profile, raising the question of how effectively our political system will be able to deal with them.

Having said that, the Mexican crisis also showed the strength of our system. Congress, while not able to act itself and often complicating our efforts, also induced greater focus on some important issues, such as moral hazard, and helped assure that all points of view were considered, a value often lost in a more monolithic system. In addition, some individual legislators were tremendously helpful. As I discovered, finding effective legislative allies is key to navigating our system successfully. At one point, Senator D’Amato had proposed measures that would rule out future Treasury use of the ESF in this type of situation—which would have severely hampered us in dealing with the Asian crisis two years later. But Senators Dodd and Sarbanes, who had a deep understanding of the benefits and risks of the global financial system, filibustered D’Amato’s language, which led to a more limited constraint. I also remember an act of graciousness of the kind that occurs too seldom in any walk of life. Frank Murkowski, a Republican senator from Alaska and a former banker, who had opposed our rescue package as unlikely to work, went out of his way when I was testifying later at a hearing on another matter to say that he had been wrong—a gesture unusual in Washington and most other places.

However, Murkowski’s prediction could have turned out to be right. Our program could have been undertaken only by a President—and an administration—willing to take a major calculated risk, substantive and political. We could have failed because of a mistake in our analysis, but also because of unforeseeable circumstances, or simply the foreseeable risk actually occurring. If the odds are calculated accurately at three to one, you’ll lose one time in four. Unfortunately, Washington—the political process and the media—judges decisions based solely on outcomes, not on the quality of the decision making, and makes little allowance for the inevitability of some level of human error. This can easily lead to undue risk aversion on the part of public officials. The same issue exists in the private sector—in my own experience, most seriously in judging trading and investing results. But the private sector somewhat more frequently recognizes the need to look beyond the outcome to reach the most sensible and constructive evaluation.

Some years later, Paul O’Neill, the Bush administration’s newly appointed Treasury Secretary, said he liked the Mexican program because it worked. “We gave them money, it stabilized their situation, and they paid back the money ahead of schedule,” he said. “I like success. I’m not a real fan of even well-meaning failure.” Where O’Neill said he liked what worked, my view was that decisions shouldn’t be evaluated only on the basis of results. Even the best decisions about intervention are probabilistic and run a real risk of failure, but the failure wouldn’t necessarily make the decision wrong.

Finally, what concerns me most is how little the public understands the impact that all of the issues around globalization and economic conditions elsewhere have on jobs, living standards, and growth in this country and how critical U.S. leadership is on these international economic matters. The result, as I realized over and over again during my six and a half years in Washington, is that public support—and thus political support—for trade liberalization, international financial-crisis response, foreign aid, funding for the World Bank and the IMF, and the like—is at best very difficult to obtain.

At one point during the second term, Secretary of State Madeleine Albright and I discussed holding joint public meetings around the country to try to improve public understanding of how global issues, both economic and geopolitical, affect people’s lives. Regrettably, we never did this, but some kind of ongoing public education campaign is badly needed to change the politics around all these concerns, which are so critical to our future. On trade, for example, dislocations are very specific and keenly felt—and lead to strong political action—but the benefits of both exports and imports are widely dispersed and not recognized as trade-related, and thus haven’t developed the level of political support they require.

We also face significant challenges when it comes to the international politics of economic leadership. In Mexico, and later in the Asian financial crisis, U.S. leadership, exercised in correlation with the G-7, the IMF, the World Bank, and others, was necessary for effective response. But even our closest allies are ambivalent about the role of the United States. We are criticized if we don’t lead and resented if we do. At Treasury, the lesson we took was to work all the more energetically with other countries to reach consensus whenever practical, which often meant making accommodations on our part. But we also recognized that at times we would feel a need to push beyond where others wanted to go.

In 1995, I referred to the Mexican crisis as a “very low-probability event.” But my view later changed. The likelihood of a contagious crisis emanating from problems in any one developing country may ordinarily be small. But modern capital markets—with their many interrelationships, size, and speed—combined with the inherent human tendency to go to excess, create a seemingly inevitable tendency toward periodic destabilization that is difficult to anticipate and prevent. Indeed, only a couple of years later, I found myself immersed in another global financial crisis—one far more threatening in its scale, complexity, and potential consequences than what had happened in Mexico.