Crisis Prevention and Resolution: Lessons from Argentina, Address by Anne Krueger, First Deputy Managing Director, IMF
July 17, 2002
First Deputy Managing Director
International Monetary Fund
National Bureau Of Economic Research (NBER)
Conference on "The Argentina Crisis"
Cambridge, July 17, 2002
Mr Chairman. Ladies and Gentlemen.
I am delighted to join you here today for this important discussion of Argentina's economic difficulties, and the challenges they pose for policymakers both in the country and at the international financial institutions. Ordinary people in Argentina are paying an enormous price for the current crisis and it is incumbent upon us in the international community to do all we can to help the country recover as quickly as possible—and to help prevent other countries from suffering a similar fate. I hope that the discussions here today can play some small part in that process.
I would like to focus on some of the lessons that Argentina's recent experience holds for the Fund's efforts at crisis prevention and resolution. Observers of financial crises are fond of recalling Tolstoy's famous remark that "all happy families resemble each other; each unhappy family is unhappy in its own way". Argentina is a case in point. It faithfully applied many of the lessons that we thought we had learned from the previous crises of the mid and late 1990s—but a combination of new mistakes and some old ones brought it to grief all the same. Thus, inevitably, we now have a new set of lessons to learn—and some old ones to remember a little better.
Let me begin by sketching out what seems to have gone wrong in Argentina, a subject that other speakers are covering in greater length and detail today. I will then draw out some of the key long-term lessons for the Fund and its members, before concluding with a brief discussion of the immediate challenges Argentina faces and the Fund's role in helping to address them.
2. What went wrong?
Argentina's fall from grace has certainly been a spectacular one.
Between 1990 and 1997 its economy outperformed that of most other countries in Latin America, growing by more than 6 percent a year. Contagion from the tequila crisis in 1995 was severe, but short-lived with growth soon resuming. Argentina's performance was recognized internationally with President Menem's appearance alongside President Clinton at the 1999 annual meetings of the Fund and Bank.
Now, of course, things look very different. Most Latin American countries regained momentum after the slowdown in 1999. But Argentina descended into a protracted recession that is now in its fourth year. Real GDP is expected to decline by between 15 and 20 percent this year alone—the worst performance in the region by far. Meanwhile, the foundations of Argentina's economic strategy have collapsed beneath it, with the abandonment of the currency peg, the freezing of bank deposits, and the suspension of payments on external debt.
To trace the roots of this sad decline, we need to go back to the early 1990s. With the economy beset by recession and hyperinflation, the Convertibility Plan was introduced in 1991 into an attempt to break Argentina's inflationary psychology once and for all. Impressively (but not without precedent for an exchange-rate based stabilization) it reduced inflation from four figures to single digits within three years, laying the foundation for the resumption of growth. Not surprisingly, given the depths to which the economy had fallen in the 1980s, the rebound was initially a spectacular one. But when growth slowed to a more sustainable 4 to 5 percent a year between 1993 and 1998, this was still Argentina's best performance for many decades.
The Mexican and Asian financial crises posed a serious challenge to Argentina and its fellow emerging markets. Argentina came out relatively well, helped in part by the fact that the falling dollar boosted its competitiveness in European markets.
The IMF and the international community more broadly drew a number of policy lessons from what had happened in Mexico and the Asian crisis countries. In several respects, Argentina's policies were in line with those lessons. For example, both before and after the tequila crisis, it took steps to strengthen its banking system, welcoming foreign ownership and improving regulation and supervision. Argentina was also an early volunteer for assessment against international standards and codes for policy transparency, with initial reports published in the spring of 1999.
So what went wrong? With hindsight, two factors came together in a destructive cocktail: weak fiscal policy and mounting overvaluation, the latter reflecting relatively high inflation, a stronger dollar, and insufficient domestic flexibility (for example, in the labor market). The last point is especially important—under a firmly fixed exchange, you need other sources of adjustment to maintain competitiveness.
Looking first at fiscal policy, it is important to examine the finances of the public sector as a whole—including both federal and provincial governments. The federal government ran modets deficits throughout the mid-1990s. Adjusting for the cycle, the federal government's fiscal position deteriorated steadily during the boom years, although this partly reflected a recognition of unfunded pension liabilities.
But this was not the whole story. Much of the fiscal problem reflected a lack of discipline at the provincial level, exacerbated by a transfer of responsibilities from the federal government. Once off-balance sheet borrowing is taken into account as well, the consolidated budget deficit was around 2½ percent of GDP even as the boom peaked in 1997 and 1998. The figures would have been larger still had privatization receipts not been used to finance current expenditure.
One facet of the problems in the public finances was the mounting differential between the wages and salaries paid by the federal government and those in the private sector. In 1994 the average federal employee was paid 25 percent more than the average private sector employee; by 1998 the differential had risen to 45 percent.
Judging from international comparisons, the public sector had become bloated. By the mid-to-late 1990s, public sector employment accounted for 12.5 percent of the labor force in Argentina, compared to barely 7 percent in Brazil and Chile, 6 percent or less in Indonesia, the Philippines and Thailand, and 8-10 percent in Russia, Turkey and South Africa. Indeed, the figure in Argentina was more in line with industrial countries like Germany, Italy and Spain.
The protracted weakness of the public finances soon made its impact felt on Argentina's consolidated public debt burden, which rose from less than 33 percent of GDP in early 1990s to more than 41 percent in 1998. Compared to the target of 60 percent of GDP laid down by the European Union in the Maastricht Treaty, the 41 percent reached in 1998 does not sound very threatening. But emerging market countries in general—and Argentina in particular—have greater constraints on their ability to shoulder debt than industrial countries. For example, Argentina had less capacity to raise tax revenue than industrial countries. It was more vulnerable to external shocks and to shifts in market sentiment. And much of the debt had to be serviced in foreign currency, which was made more difficult by Argentina's low export-to-GDP ratio. Argentina's total foreign currency debt was around five times the size of its foreign currency receipts from exports of goods and services. In addition, the fixed exchange rate regime under the convertibility plan further reduced the degrees of freedom for fiscal deficits and debt. External debt had already climbed to 50 percent of GDP in 2000, but of course markets feared that in the absence of fiscal and other policies strong enough to support the convertibility plan, the currency board would collapse, the exchange rate would plummet, and the debt would ballon. These fears were duly realised and by the end of 2001 the debt-to-GDP ratio stood at 130 percent.
The Fund is often accused of being too tough on fiscal policy, a line of criticism that was much in evidence in the early days of the Asian crisis. But in Argentina's case the opposite was true. Fiscal problems in Latin America have often arisen when complacency has set in during booms. This was the same old story—and the Fund should have been more forceful in urging action. Having said this, it is never easy to persuade authorities of the need to tighten their belts when foreign investors and lenders are providing an apparent vote of confidence by pouring capital in.
Now what about the convertibility plan?
With hindsight, an earlier exit to a more flexible exchange rate regime would clearly have been preferable. The real exchange rate appreciated significantly during the 1990s, although most of the adjustment took place in the early years of the decade, as inflation was brought down following the establishment of the currency board. Wages and prices were insufficiently flexible to maintain competitiveness, resulting in weak export growth relative to other Latin American countries.
Whether and when exactly the currency board could or should have been abandoned remains a matter for debate. One option would have been to do it as soon as the credibility of the monetary strategy had been established, say after three years. Another possibility would have been in 1996–97, when the economy was rebounding after the tequila crisis. In hindsight, this may have been the last chance for an orderly exit.
Of course, this is much easier to say now than it would have been to achieve then. The convertibility plan enjoyed wide popular support as a bulwark against a return to Argentina's inflationary past. Reformers also saw it is a useful source of leverage to bring about the open, market—oriented structural reforms that would in time have served to boost competitiveness. An unexpected exit could also have been highly disruptive economically given the depth and breadth of foreign borrowing.
How strongly should the Fund have argued for a change? At the end of the day, the formal position is clear: it is for a member—not for us—to choose its own exchange rate regime and to put in place the supporting policies necessary to maintain it. Our Articles of Agreement require us to support the basic policy strategy adopted by our members, as long as it has a reasonable chance of success. Of course, defining "reasonable" is like asking the length of a piece of string, but it is certainly true that while currency boards impose difficult constraints on policymakers, they can be sustained if fiscal, financial sector, and structural policies are sound enough.
Things began to go seriously awry for Argentina in late 1998 and into 1999. This reflected a number of factors: contagion from the Russian crisis of the previous summer; the collapse of the real plan in Brazil; the mounting debt burden; and the further loosening of fiscal policy as President Menem sought to stay in office, to name but three. By the end of the following year, things had deteriorated significantly. The De La Rua government was in turmoil. And there was open talk of default as Argentina faced a financing requirement in excess of $20bn a year at a time when the market's appetite for Argentine paper had been exhausted by substantial debt issues in 1999 and 2000.
When Argentina came to the Fund for help in late 2000, what were the options?
The Fund could have refused to lend, on the not unreasonable grounds that the fiscal measures necessary to avoid default had become politically and economically infeasible. This would likely have triggered default and the end of the convertibility plan. The government would have blamed the Fund, rather than "owning" the change of strategy. The authorities clearly remained wedded to their existing strategy and the international community agreed to support a program based on serious fiscal reform.
The support was much larger than in a normal Fund program—$14bn from us and a total package of $40bn. But this was in line with the other big packages agreed for victims of capital account crises in recent years. In that sense, it satisfied the requirement for the Fund to treat its members equally. There were also reasons for genuine optimism: reserves were still quite healthy; there was hope of lower US interest rates; and spreads (while high at 500-600 basis points) were still manageable.
The package provided only a brief respite. US interest rates did fall, but fiscal targets were missed, undermining confidence and widening spreads. When Domingo Cavallo was appointed Finance Minister in March 2001, he well understood the need for dramatic action on the fiscal front. He recognised that Argentina could no longer borrow or print money, prompting passage of the zero-deficit law. Revenue-raising measures were also put in place, including the financial transactions tax. At the same time, however, market confidence was undermined by the move to the dollar/euro peg, by the removal of Central Bank Governor Pou, and by a costly debt swap. Investors were also unimpressed by the various measures aimed at boosting competitiveness, some of which also undermined the effectiveness of tax administration. The authorities had in effect run out of policy tools, yet their commitment to the convertibility plan and continued debt servicing did not waver. With support from various G7 leaders, Argentina received a fresh loan disbursement. Alas, all too soon it was to prove in vain.
So, to summarize, what went wrong? With hindsight:
In other words, Argentina became caught in a vicious cycle of weak activity, overvaluation, and mounting debt.
The Lessons for Crisis Prevention and Resolution
Bearing in mind that Argentina did apply many of the lessons that the international community thought it learned from the Mexican and Asian crises, what more have we at the Fund learned—or re-learned—from Argentina's own experience?
Overarching these lessons is a bigger and even more difficult one. In a way, it is the Fund's eternal dilemma: how far should we go in supporting a policy strategy to which the authorities are deeply committed, but which we and investors doubt may succeed? What probability of success should we look for before we lend: 40, 60 or 80 percent? And how should one trade off the strength of a potential program against the chances of successful implementation? These judgments all have to be made in conditions of great volatility and uncertainty, both economic and political.
Ownership by the authorities—and ideally by society more broadly—is very important if a program is to be implemented effectively. If we had forced the authorities in Argentina to abandon the convertibility plan and/or default on their debt, then we would have got the blame for the economic consequences. More importantly, the resulting lack of ownership and commitment by the authorities would likely have made those consequences more severe in their impact on the Argentine people. (National policymakers are just as aware of this dilemma as we are, of course, which can encourage a measure of gamesmanship in the negotiation process.)
We must remember that while ownership is necessary for the success of a program, it is not sufficient. It has to make sense economically. At the end of the day, a country is responsible for the policies it chooses—and we are responsible for deciding whether to lend in support of them.
The whole point of Fund lending in financial crises is to take risks and provide finance where the private sector will not, but at the same time we are responsible to our members for ensuring that their money restores growth and viability, thereby allowing repayment. Some people will always—and do always—complain that we err too much on one side or the other. That is inevitable, but we need to be as clear and open-minded as we can be in explaining and justifying the decisions we take, bearing in mind that in conditions of uncertainty the judgment is ultimately probabilistic.
4. The Immediate Challenges
Let me conclude by talking a little about the immediate challenges Argentina faces now. You will appreciate that I am somewhat constrained in what I can say here.
The economic situation remains very difficult:
We have seen some welcome progress in key policy areas, including satisfactory amendments to the insolvency law and the repeal of the economic subversion law. The authorities are also making progress in putting in place bilateral agreements with the provinces to ensure an orderly adjustment of their finances. This is an essential component of a coherent fiscal framework.
Reflecting this progress, the IMF's Executive Board agreed on July 15 to extend by one year the expected repayment of $985m due from Argentina under the Supplemental Reserve Facility.
Our discussions with the authorities will now focus on two key areas:
Rapid progress in these and other areas is essential to create a strong and comprehensive program with the capacity to restore confidence, underpin a sustained economic recovery, and win international support. Time is not on Argentina's side. We will remain fully engaged with the authorities in working towards that goal.
Argentina: Personnel Expenditure
|Wage bill, in percent of GDP|
|Public sector workers, in thousands|
|Public sector workers, in percent of total labor force|
|Average annual wage, in Arg$|
|Average annual wage, in Arg$ constant prices (1999=100)|
|Ratio average wage to GDP per capita|
| Labor force
(39% of pop)
|GDP per capita||4,636||5,814||6,983||7,501||7,421||7,727||8,210||8,277||7,744||7,695||7,177|
1Excludes public enterprises.
| Public Sector Employment
and Wages (1996-2000)1,2
|Average Wage||Public sector employment|
|(% GDP per capita)||(% of labor force)|
|Middle Income countries|
|Brazil||. . .||7.3|
|Thailand||. . .||6.01,2|
|Russia||. . .||8.3|
|South Africa||. . .||9.21|
|Sample of European countries|
|World Bank: Cross National Data on Employment and Wages (for middle income countries)|
|OECD: Public Governance and Management Statistics, 1997-98 (for European countries)|
|Fund staff estimates.|
|Public Sector coverage: (unless otherwise specified)|
|World Bank and OECD data includes central government, local government,education and health employees, and armed forces.|
| Middle income countries: 1996-2000,
with the exception of: Argentina (1999-2000), Bolivia
Brazil (1991-95), South Africa (1991-95).
|European counties: 1997-98.|
|1Includes State Owned Enterprises.|
|2Exclude local governments.|
|3Excludes armed forces.|