Hobson's Choice

free script provided by

Where, if anywhere, did we go wrong? The IMF in Argentina

The IMF in Argentina-1

Think back a few years to 2001, when gazillions of Argentinean citizens gathered outside the Casa Rosada banging pots and pans. Then-president Fernando de la Rua was obligated to leave office in a helicopter, and a period of political upheaval hit that unhappy country. The currency plummeted against the US dollar (chart), triggered not by spectacular Argentine profligacy, but by capital flight largely beyond the control of the state (see a few charts here). In the face of the massive inferno on global capital markets, the IMF had to act quickly. But by summer of 2002 the government had defaulted on over $120 billion in external debt, the economy was plunged into depression, and it looked like a resume blot for the IMF.

Was it? The "Evaluation of the role of the IMF in Argentina, 1991–2001" prepared by the Independent Evaluator’s Office (IEO), has just been released, and some might be interested in this probe into the financial furor of the century. Readers interested in deciding how transparent the IEO really is, are urged to check IEO guidelines; also, this critique by Environmental Media Services (EMS). My own concern is that the IMF's IEO is more like a management consultant, not an economic analyst; the criticism is directed chiefly at administrative practices within the organization, not its underpinning philosophy.

For that, let's listen to Joseph Stiglitz, Nobel laureate and frequent IMF critic:

IDEAS: To slash inflation, expectations needed changing; 'anchoring' the currency to the dollar was meant to do this...If inflationary expectations were changed, then disinflation could occur without costly unemployment. This prescription worked for a time in a few countries, but was risky, as Argentina was to show.
The IMF encouraged this exchange-rate system. Now it is less enthusiastic, though Argentina is the one paying the price. The peg lowered inflation but did not promote sustained growth. Argentina should have been encouraged to fix a more flexible exchange rate, or at least a rate more reflective of its trading patterns.

An account of the hard pegs and their effects in Argentina may be found in "Living and Dying by the Hard Peg" (PDF), especially pp.10-16. Stiglitz's view that it would indeed have been possible to sustain a more flexible regime (without moving to a pure float) is supported by the linked paper.

The world had hardly recovered from the 1997-1998 financial crisis when the 2000 2001 global slowdown started, worsening Argentina's situation. Here the IMF made its fatal mistake: It encouraged a concretionary fiscal policy, the same mistake it had made in East Asia.
Fiscal austerity was supposed to restore confidence. But the numbers in the IMF programme were fiction; any economist would have predicted that concretionary policies would incite slow-down, and that budget targets would not be met.
Needless to say, the IMF programme did not fulfill its commitments. Confidence is seldom restored as an economy goes into a deep recession and double-digit unemployment.

The Stiglitz article is well-worth reading, but I think these are the most important points.

The IEO Report on IMF Conduct in Argentina

A summary of IMF relations with Argentina begins with Appendix II, p. 127. The IMF rejected lending to the Argentine government in the closing days of the Alfonsín Administration ('88) because that president's Plan Primavera did not establish fiscal control over the provinces; the World Bank Group went ahead and lent $1.25 billion to Argentina anyway, but the Plan Primavera collapsed anyway.¹ Subsequently, Carlos Menem came to office and introduced a new plan, which did meet IMF demands for balancing the budget. So the IMF lent the Argentine government 5 billion SDRs (about US$7 billion), and capital flows reversed. However, the fiscal targets were never reached and, according to the IMF's own audits, the outcome was "unsatisfactory at best" (p. 129); stabilization of capital markets was not achieved, rendering it extremely dangerous to invest in Argentina.

Incidentally, taxes in Argentina are highly regressive; while aggregate governmental expenditures in that country were around 17.38% of GDP [1991-2000; see p.130], about half comes from G&S taxes and a quarter from social security taxes. This is a standard feature of countries whose fiscal systems are highly decentralized.)

If not for financing of past public debt, Argentina's fiscal discipline would have been admirable; six of the ten years 1992-2001, the primary balance was a surplus. While most reports imply that Menem's government was a model of fiscal discipline, while de la Rua (Partido Radical) was not, the fact remains that the privatization of state assets occurred during Menem's tenure; anyone can run a surplus while selling off the national airline and the national telecom network (in a telecom bubble, no less). However, the IMF audit I'm reading repeatedly makes the point that off-book financing of government expenditures was endemic; what this means in plain English is that the government spent a lot, then borrowed money to pay for it, without actually documenting it as a deficit. So while the decade reflected European levels of fiscal restraint, Argentina's public sector debt doubled.

Sovereign debt was growing much more quickly than would be expected from the year-to-year deficit figures.... One reason for this was the (often court-ordered) assumption of old debts, including overdue obligations to pensioners, government suppliers, and provincial governments. The authorities were also prone to issue off-budget debt to settle government obligations, through such means as the capitalization of interest payments. While such increases in debt were not given due recognition, the deficit-related performance criteria for the program supported by the 1992–94 extended arrangement included privatization receipts.
[p. 41; emphasis added—JRM]

This forms the crux of the critique. While the IMF had earlier hung tough on fiscal balances, after the inauguration of Pres. Carlos Menem in July '89, afterwards it seemed surprisingly careless. The national and provincial governments were allowed to record the sales of governmental enterprises as revenue, while a host of one-time judgments were treated as write-offs, to be capitalized over many years. With this detail added to the picture, the narrative reverses: instead of Carlos the Thrifty, and Fernando the Sugardaddy, it becomes the Carlos the Shifty, and Fernando the Luckless.

NOTE: 1 The Plan Primavera is described in "Argentina: Major Events Since 1976" (PDF), p.3:

devaluation and re-organization of multi-tiered exchange rate system;

price restraint pact with business;

series of measures to cut the deficit;

general reduction in import tariffs from 35 to 30%.

A run on the austral occurred, and Argentina suffers worst hyperinflation in its history. Under these circumstances, Carlos Menem wins elections and takes office immediately, rather than waiting for inauguration. As in California in 2000-03, the party out of power prevents necessary reforms until it wins the executive position, then accepts them.

Obviously, this was a political problem; the IMF had objected to the failure of the Radical Party administration's plan to curb spending by the provinces, with the result that the deficit was approaching 16% of GDP. This is a compelling reason, but of course the reason why the Radical Party could do no such thing was the "rule-or-ruin" strategy of the Peronist Party, which used strikes to break the government's wage and price control programs. Later, when Menem took office, he imposed a much sterner program of wage and price controls, lowered tariffs more than the PR had proposed to do. The IMF rewarded the Peronists with a $1.5 billion bailout package. It really is far too much to ask that multilaterals like the IMF refrain from rewarding groups like the Peronists, because parties would be eternally stuck in destructive positions after they won. Suppose, in other words, that Menem came to office and sought to implement a left-wing Peronist agenda; in that case, they would be merely being intellectually and ideologically consistent. But Argentina would effectively cease to have any formal economy at all. Likewise, the scope of party differences far exceeds that of economic management. So the Peronists could argue that efforts to blunt the Plan Primavera were motivated by the class shares of the burden. The IMF would have to weigh in on whether this was the case, which it understandably would have no desire to do.

The IMF in Argentina-2

This entry continues my review of the "Evaluation of the role of the IMF in Argentina, 1991–2001," a formal audit by the IMF's IEO. The report is actually very good, much better than I had suspected. First, I was pleasantly surprised by the rigor of the analysis. I had feared the report would basically rubberstamp the "Washington Consensus," but it did not. For example, in the mid-1990's, the WC pretty much insisted that hard pegs (or better still, dollarization) were a panacea for both capital flight and for dodgy political actors.² The peg animated anxiety in the Western Hemispheric Division (WHD), but in real time, exports were surging and the real effective exchange rate, measured in PPP, was stable. As the audit points out repeatedly (e.g., p.32), this points out the failure of real time scrutiny only. "Real time" simply means looking at the immediate outcome of a certain policy.

As I pointed out in a footnote below, the RE/RBC analysis that forms the underpinning of the Washington Consensus tends to rely mainly on optimization models that feature a homogenous average population. The solution set is then fit to the data using multivariable regression analysis (the latter determines the coefficients on the model). In my opinion, this is wrong because (a) it tends to shunt the observed data to one side; (b) it mines the data so that virtually any theory will fit; (c) it is ideologically driven, and rests on rhetoric that villifies economic analysis from the previous generation as driven by some invidious motive. But when that school of economic thought is applied to the real world, it tends to make the opposite error: rather than plan ahead and identify future issues before they arise, RE/RBC analysis tends to become complacent when GDP is growing. The IMF had some compulsion to consider the political consequences of its policies, so it had to accommodate some remedial policies of the Argentine government if this or that segment of the population were to suffer extreme hardship; but matters such as the accumulation of unsecured liabilities on the part of provincial governments or fiduciary agencies of the state were largely ignored.

I say largely, because it's not as if there were no voices in the IMF objecting to the strains imposed on the Argentine financial structure by the currency board. The growth potential of the country was known to be small; the country was not a potential South Korea. Sooner or later the influx of foreign finance capital would pause, and perhaps reverse direction; the country was running a substantial trade deficit, which would not have been such a problem were the economy headed for a major expansion (as, for example, if imports were mainly of industrial equipment). Incidentally, I think that the reason the IMF-WHD were uneasy was that they were notionally loyal to the WC as a governing ideology, but as intelligent men and women, skeptical of it.

In 1998, these anxieties did arise. According to the report, now the time to switch to a floating exchange rate regime was past. As Argentina moved to a crisis of capital flight (inflicted by the Asian Currency Crisis), the risks and costs of exchange rate movements would be borne by the Argentinean taxpayer, not foreign investors. Brazil's currency, the real, plummeted against the dollar and Argentine exporters lost market share to Brazil. Finally, as if that were not enough, the Federal Reserve raised rates 175 basis points in order to clear up large amounts of excess liquidity. Still, the IMF-WHD failed to discuss an exit strategy from the peg; although by then, doing so would have been very dangerous. Indeed, the administration of Pres. Fernando de la Rua, determined to avoid the sort of panic it would be so costly to defend against, declared his intention to dollarize the economy entirely.

By 1999, the Argentinean government was under intense pressure to cut the country's gigantic deficits. The public sector of Argentina had been tying up a terrifying share of the capital flowing into the country—another reason why IMF analysts understood that the country was not poised for take-off. While it is certainly needful and prudent to pay pensions and what-have-you, such expenditures were being used to run up large off-book debts.³ Unfortunately, the method of during this was to cut back on the politically weakest beneficiaries of the state, while raising the value-added tax to 21% of GDP; this naturally caused the economy to contract faster. In order to defend the peg, the government was compelled to respond to a recession by reducing the deficit, making the recession far more severe. Capital flight accelerated. In August of '99, the IMF began contemplating an exit strategy; it could not think of one, nor did it ever; the opportunity was now past.

In a country like the USA, a lot of the politically unacceptable options for solving problems like deficits or poor infrastructure become acceptable when the problem becomes sufficiently bad. Pres. Nixon and Ford resisted actually doing anything meaningful to fight inflation because they had no mandate to do so; Carter resisted until '80, when oil prices hit their all-time historic high and he appointed Paul Volcker to chair the Federal Reserve Board of Governors. Later, Reagan and Bush pere ignored the deficit because they had no mandate to cut spending and/or increase taxes; Clinton, on the other hand, did. Other examples abound, but the rule of thumb is that Americans can always wait until the cost of doing nothing is prohibitive. In Argentina, the reverse is true; when Argentine citizens experience the level of anxiety needed to redress an economic problem, the opportunity to do it in an orderly way is long past. Hence, while the USA could eventually bumble into a counter-cyclical fiscal policy to end the Great Depression (viz., thanks to the thoughtful Japanese attack on Pearl Harbor), by such a time in the Argentine political cycle, it's too late to do anything besides depose the leader and brace for utter grief. This has nothing to do with a character flaw on the part of the Argentines, but the extreme mistrust of the rival political factions of each other certainly makes it harder to avoid the economic equivalent of nuclear war.

This makes it incumbent upon organizations like the IMF to provide pressure not to be pro-cyclical, since this only accelerates the capital flight that pegs are intended to prevent, but to anticipate extreme cycles and alter the trajectory towards them. Because of the Mundell-Fleming mode Argentina lives in (highly exposed to capital flows, not exposed at all to goods flows), countercyclical fiscal and monetary policy is not an option no matter what exchange rate regime the country is under, and no matter how virtuously the Argentineans behave. This fact is made very clear in the audit (e.g., p.36), I am pleased to say—which leads me to believe the IMF is making some headway.

NOTES: 2 p.31 PDF report; or see "Dollarization and the Conquest of Hyperinflation in Divided Societies" (PDF) for a flavor of the thinking in orthodox economic circles c. summer, 2001. The paper was published by the Minneapolis Federal Reserve and is helpful for recognizing institutional biases that led to the fetish of dollarization. Basically, the matter is analyzed as a constrained optimization problem in which Argentine citizens are treated as homogenous averages (a standard Modeling strategy in RE/RBC theory) and solved to find how they would behave under an inflation tax inflicted by provincial governments. The first order conditions assume that provincial governments can always finance deficits by monetizing the debt through the central bank, which they are presumed to control through the legislature; and that this will motivate Argentinean citizens to buy imports to avoid the inflation tax.

Within Argentina, there is an incentive for each region to inflate since the tax imposed by the government in region k is paid in part by agents in the other region. The government of region k chooses σ^k to maximize the welfare of a representative agent in that region,...The government takes as given the money creation rate of the other government. But each government recognizes its effect on the overall rate of inflation in Argentina and thus internalizes the response of all workers in both regions...
[p.5]

Oddly, no effort is made to confirm if Argentine imports burgeoned in response to the many episodes of hyperinflation; if there had been, it would have been observed that no reliable correlation exists in Argentine history. Big upward movements in either imports or in the current account deficit are caused by several things, one of which was virtual dollarization itself. To be fair, RE/RBC theorists did set up models like then and test them against the historical data; the coefficients assigned to predictive variables were determined by multivariable regression analysis. To me, this is data mining, but evidently it survived for two decades as the predominant economic model.

3 Inflationary economies with republican governments, such as Argentina's or Brazil's are prone to legally-mandated one-off expenses, like massive increases in pension liabilities. The courts of Argentina ruled that inflation imposed certain obligations on the provincial governments of the country to raise pensions retroactively, and the Argentinean government made the decision to finance these "unexpected," "one-off" expenses with bond issues. Under the Menem government that preceded de la Rua's, however, one-off windfalls from the privatization of the country's utilities were counted as revenue. This enabled Menem to finance pork-barrel politics while looking very good on fiscal discipline—even to the IMF.

The IMF in Argentina-3

Fiscal Policy

This section of the report is very straightforward.

Fiscal policy was the single most prominent topic of discussion between the IMF and the Argentine authorities for virtually the entire period of convertibility. While fiscal policy often dominates the IMF’s interactions with member countries, it assumed a particular importance in the case of Argentina. For one thing, there was a history of fiscal irresponsibility that had in the past contributed to repeated cycles of defaults and hyperinflation. Moreover, the choice of the convertibility regime made fiscal policy especially important.
[p.37]

After the adoption of the currency regime, there was no possibility of monetary policy to manage the economy. As I pointed out in Part 2, there wasn't much possibility of either fiscal or monetary policy action, even if Argentina had not adopted a currency board. That's because of the Mundell-Fleming mode that Argentina functions under, in which increases in spending (to stimulate the economy) or reductions in the interest rate (to stimulate the economy) are offset by the spillover effects on Argentina's foreign capital flows or trade balance.

As it happened, Argentina's debts were persistently high, in large measure because off-book expenses mandated by the courts (and not accommodated by spending cuts elsewhere). Also, the Argentine government was burdened by a large and growing pool of debt financing costs that typically exceeded 1.5% of GDP Argentina's provinces are extremely important fiscal actors, more so than the USA (since the federal government of the USA, for example, has such a major influence on state spending).

As I've mentioned earlier, the IMF had an implausibly high notion of the amount of debt the country could support; it also had a very high institutional stake in continuing to help Argentina even after the government failed to meet fiscal guidelines. The error of judgment was that the IMF was focused on flows of tax revenues into the public coffers versus flows out; "one-off" expenses were treated as such, and financed with securities (which the finance ministry later bought and allowed banks to use as hot money), while one-off windfalls from the sales of state enterprises were counted as revenue flows. Curiously, these privatization measures were not obscure; they were much discussed, as when a Spanish firm bought the country's telephone system.

The Western Hemisphere Division (WHD) that managed IMF's relations with Argentina was not blind to the fact that, even ignoring off-book accounting methods employed by the Argentine Ministry of Finance, the country's debts were far too high:

When growth was robust, staff did sometimes try to argue for tightening the fiscal targets, but to no avail. In March 1993, for example, the staff advised the authorities “that a strengthening of the public finances, perhaps even beyond the programmed level, would restrain absorption and reduce the risks to the program.” The authorities responded that, in their view, demand pressures were subsiding and additional restraint was not necessary. In the end, the targets that had been set at the beginning of the year were not adjusted and were met only with a small margin. Likewise in April 1998, senior staff wrote a letter to the authorities stressing the need to tighten fiscal policy in view of a large current account deficit. It should be noted, however, that the staff’s approach to fiscal policy in these instances was motivated by cyclical demand management considerations, and not by debt sustainability concerns.
[note, p.40]

The problem was not thatthe IMF was either too loose or too tight in its advice to the Argentine government; it was that it continued to enable the government to accumulate both SBA's and hot money when the economy was expanding.⁴

And that expansion was driven almost wholly by consumption spending.

Did the IMF learn anything about fiscal policy management as a result of the Argentina crisis?

It is relevant to ask whether diagnostic tools developed in the IMF since the Argentine crisis would have generated stronger warning signals, had they been available earlier. Our analysis shows that debt sustainability analysis would have consistently projected the external debt-to-GDP ratio to exceed the suggested benchmark of 40 percent during 1998-2000 ...In this sense, external debt sustainability would clearly have been questioned by 1998. The public debt-to-GDP ratio, however, would have been projected to exceed 50 percent only in 2001 even under the most extreme scenario. It is not clear if staff would have taken this as a sufficiently alarming
signal... Even with the best available methodology, debt sustainability analysis remains “fundamentally a matter of judgment.” To trace what actually happened, debt sustainability analysis would have required unusually adverse assumptions on the exchange rate.
[p.44-45]

With regard to the actual size of the Argentine state sector:

We are not making a judgment on the relative size of the public sector compared with other countries, but on the country’s willingness to generate tax revenues on a sustainable basis to support choices on the level of public expenditures. However, ...the average Argentine federal employee was paid much more than the average Argentine private sector employee (as much as 45 percent in 1998) and that Argentina’s size of the public sector was large by international standards, with its public sector employment comparable to that of some industrial countries.
[note, p.45]

In response to this debacle, the IMF's Independent Evaluation Office (IEO) issued a recommendation (p.122) to the effect that the IMF should make its continued support more explicitly conditional on the target state meeting guidelines. This followed the first recommendation, that for stop-loss measures (basically, a strategy to withdraw from an intervention and absolve itself of involvement in a poorly managed national economy).

I thought the debt surveillance recommendation was a bit weak, since they already acknowledged the paucity of evidence available to the IMF that Argentina's debt load was too high (p.123)

Other recommendations failed to reference specific economic lessons of the Argentine experience; they did mention the possibility that the IMF abstain from involvement in situations where the current account balance was not an issue, and a strengthening of the role of the executive board. Since directors of the executive board are appointed by the executives of top-ranked member states, such as the US president, the UK and Japanese PMs, and so on. This seems an odd suggestion to make in view of the close (and morally challenged) relationship between Carlos Menem and the Clinton Administration. In the past I'd found this relationship rather disturbing, mainly because Argentina was vulnerable to its relationship with the USA and Carlos Menem was a horrible—if charming—leader. It's hard to imagine the POTUS rebuffing a genuinely friendly foreign head of state, even if he's making a hash of his country's finances. It really looks as if the increased involvement of the Executive Board (Rec 6) would make Rec 2 (more explicit conditionality) harder to observe.

NOTE: 4 SBA = stand-by arrangement; basically, a loan of funds before they are likely to be needed by the monetary authorities. Different from an extended arrangement in that SBA funds are either used or they lapse; the other are "outstanding" until they are entirely lent out.

"Hot money"—reserves employed as the base for currency creation.

The IMF in Argentina-4

Structural Policy

The usual picture of the IMF as an enemy of the 3rd world and friend to finance capital is usually validated best here, in the pressure on sovereign nations to alter labor laws to favor exports—the assumption invariably being that depressing wages is the path to competitiveness. Much of the time this picture is a cartoon; if the critic favors urban development, then the IMF is usually faced with poor options. In Argentina, for example, the IMF was blamed for the wrong reasons—while stung by charges that it was unduly parsimonious, preventing stimulative fiscal policy and social investment, in reality it erred on the side of letting the country accumulate excessive debts.

The first point that the IEO audit points to is the unrealized need for Argentinean reform of the federal-provincial finances. As I've mentioned in previous installments, Argentina is even more devolved than the USA, which is—in turn—the most devolved fiscal entity in the OECD. Fiscal devolution is an inherently libertarian/conservative concept; it allows affluent regions to avoid transferring wealth to poorer ones, even if they are industrially dependent upon those regions. In Argentina, there exists an extremely complicated system of gathering and sharing tax revenue that varies with the type of tax gathered. However, the provinces are famous for being able to create fiscal crises that the central government must resolve (see, for example, p. 48 and p.132 of the IEO audit linked above). Part of the problem is that regions differ from each other so much in the sorts of needs they have, so there's no unified philosophy of what sorts of subsidies are legitimate and which aren't.

While the USA at least has most of the revenue paid out by state and local governments collected by those same governments—so that debates on public spending at least include concomitant debates on the taxes required to pay for those programs—in Argentina, such debates are lopsided; the tax angle is ignored since the "coparticipation" scheme doesn't require politicians to devise taxation schemes, it merely requires them to grab a slice of the pie for their specific province. The IEO report criticizes the governments of Argentina for failing to reform this system, relying instead on a temporary provision that essentially "borrows" from the provinces in order to mop up the fiscal disaster they orchestrated.

Subsequently, the fiscal compulsions of the federal government necessitated temporary changes in the coparticipation scheme. The 1998 tax reform, which increased shared taxes, compensated the federal government for the lost revenue share by allowing a fixed deduction (of up to Arg$2,154 million a year) from the collected revenue until the end of 2000. The fiscal pact of December 2000 extended the validity of this deduction for the federal government until 2005. At the same time, it replaced revenue-based transfers by a fixed transfer of Arg$1,364 million a month for 2001–02 and, for 2003–05, by a predetermined but increasing amount of transfers. Additional changes were introduced during the crisis in early 2002, but a permanent reform of the revenue-sharing scheme was not made. In view of the federal government’s inability to pay the guaranteed levels of transfers to the provinces, the pact of February 2002 abolished the deduction of Arg$2,154 million and made 30 percent of revenues collected from the financial transactions tax subject to revenue sharing.
[p.48-49]

As for the taxes themselves, the IMF sent a mission to the country to research and design tax reform concepts; these were mostly implemented in the '98 Tax Reform bill. The object was broaden the base of taxation and reduce reliance on payroll taxes. The IEO is mum on the merit of these reforms Presumably most economists would agree that tax policies have significant microeconomic effects; most political scientists, though, would argue that the scope of action for any multilateral lending agency to affect these was slight. Argentina, like most countries with a strong "libertarian" political tradition—the USA and Colombia are other examples—have endemic problems with tax evasion, and this tends to favor regressive taxes and user fees over progressive and corporate taxes. The IMF took the unprecedented step of sending technical advisors to Buenos Aires Province to administer a new tax compliance program (1996; p.50); to little avail.

Successive FAD missions noted that weak revenue administration was associated with frequent changes in tax law and senior management in tax administration, politicization of the tax administration, lack of a computer- based accounting system that consolidates different payments and tax liabilities of each taxpayer into a single account, insufficient audit coverage, numerous payment facilitation schemes, frequent use of tax amnesties, and lengthy and inefficient appeals procedures. As a tangible reflection of these weaknesses, from 1993-96 to 1997-2000, total net tax collection remained essentially unchanged at 21 percent of GDP. Notably, there was no change in net receipts from the VAT (at 6.8 percent of GDP), despite the fact that the VAT rate was raised to 21 percent from 18 percent in 1995.

This conforms to my preconceived notion that the VAT, however suitable it might be to countries like Canada and Sweden, is not at all useful in countries like Argentina; the latter is both opaque and ideologically libertarian, making evasion both technically easy and socially approved.

In response to this, the IEO's audit is a disappointment: despite a gigantic body of literature on administration reform, the IEO has only a couple of sentences on the matter, and is largely exculpatory. This, despite the fact that Argentina is famous for being opaque. It's routine for analysts to simply write off the country as a basket case, too "corrupt" to ever get out of its perennial stagflation, and yet generations of these same analysts—Rüdiger Dornbusch springs to mind—have professed shock when that same corruption undermined their plans for the country. If it was universally recognized that corruption was the problem, why wasn't more exhaustive effort applied to tackle corruption? The technology exists.

I want to address this in a subsequent post on the IMF and multilateral entities like it, but here is the crux of my point: the IEO does appear to have succumbed to a standard neo-colonial fallacy of exculpating the physician (the IMF) by blaming the patient (Argentina) for a problem the physician explicitly resolved to cure.

What about reactionary social policy prescribed by the IMF? From the first, it is clear that the IMF saw labor market "flexibility" as meaning increased inequality. The IEO and IMF were unembarrassed about the fact that their version of reform meant downward movement of wages. Here the disappointment of the IEO at political resistance is risible. Imagine! The vast majority of the population of the country was reluctant to have their incomes slashed by huge rates! Not surprisingly, the labor market reforms mentioned (p.52-53) were measures that IMF governing nations had largely outlawed in their own countries. If you want to see a country with a really inflexible labor market, pay a visit to the ones that appoint directors, such as France or Germany...ones that run trade surpluses, as opposed to the less inflexible ones that run gigantic trade deficits as their labor markets become more "flexible."

Had the IMF gotten what it wanted, it would probably have (a) required a falangist junta, such as the "Process" junta that governed the country between 1976-83; (b) the "reforms" would scarcely have affected the general fiscal soundness of Argentina, since the real problem was the banking system and currency speculation by a tiny minority of the superrich; tax evasion, also by the superrich; and plundering of the treasury by Menem's privatizations, which benefited mostly the superrich, foreigners, and state employees; (c) Argentina would be ripe for another cycle of imperialist exploitation thanks to a dramatic long term reduction in its real exchange rate below parity.

The IMF in Argentina-ADDITIONAL OBSERVATIONS

Readers in OECD countries will perhaps be interested in the segment on social security reform (away from pay-as-you-go, or PAYG) to asset-based programs, p.54 and following:

The pay-as-you-go (PAYG) social security system of Argentina was reformed in 1994 with a partial “privatization” that created a fully funded pillar in the system. Younger workers were allowed to choose between the state-run system and approved private pension funds. IMF staff had long recognized that the existing PAYG system was headed for insolvency and that a serious reform of some kind was needed. Social security reform was made a structural performance criterion for the program supported by the extended arrangement approved in March 1992. The policy memorandum specified that the reform would involve “[f]inancial equilibrium of the existing pay-as-you-go system on both a cost and accrual basis” as well as “a new mandatory, capitalized, privately administered system,
and a voluntary private supplementary system.”
...In principle, the switch from a PAYG system to one that is fully funded can lead to a higher level of national savings and investment, higher capital accumulation, and higher long-run per capita income. ...Contributors invest in a mix of public and private assets,... while retirees draw on the income from the assets they had accumulated during their working lives to pay for their retirement. If the population is growing and the pension funds are well run, this creates a pool of savings entrusted to private managers who compete in search of high returns, a setup which should improve the efficiency of capital allocation.

For these long-run benefits to obtain, however, the transition costs from one regime to the other must be financed through taxation rather than public borrowing. [A] tax on the “old” generation ...would seem unfair, since this generation already has made contributions under the old system, which went to support the previous generation of retirees. ...[If by a] tax on the current “young” generation...., the young will be taxed twice: once for their contributions to the new regime and once for the transition payments to the current beneficiaries ...Some countries have tried to smooth the transition costs by issuing debt. But debt-financed privatization is no different from taxing the young.

In the case of Argentina, the transition was financed through an increase in debt.

In retrospect, most observers (the IMF, the World Bank, local commentators, and the administrators of the new private funds) overemphasized the potential benefits of the new system and failed fully to anticipate its severe fiscal consequences. Part of the problem was that it overestimated the self-financing component of the reform, without recognizing the imperfections of capital markets that would create an immediate burden on the government’s borrowing requirements. The increase in fiscal deficits arising from the reform was considered simply as an explicit recognition of already existing implicit debt, which the markets should be willing to finance.

In view of the fact that the IMF and its affiliated schools of thought remain champions of asset-based retirement plans, as opposed to PAYG ones, it follows that more critical examination of the IMF's role in such a major decision is required.

Financial sector liabilities associated with dollarization

In view of the proximate role of bank failures and the peg to the US dollar in Argentina's financial collapse, this seems to belong front and center in the IEO's audit. In a few words, if banks hold dollar-denominated savings (an option that was available to Argentinean depositors under the currency board) but assets in pesos, this meant that the banks held the risks associated with dollarization. Inflation domestically (in excess of US inflation) meant liabilities rose, since the REER of the peso was increasing, while the US Federal Reserve set interest rates based on inflation in the USA; hence, the spread between inflation and interest rates in Argentina was slightly—but critically—smaller than in the USA. Banks tended to favor the dollar peg, however, since it was the sternest possible measure against inflation.

This means that the banks were in favor of a measure that tended to increase the risk to themselves—since they tended to systematically underestimate the risk of an unsecured run on the peso. This reflects a fundamental flaw in the efficient markets hypothesis, viz., a hidden assumption that risks and risk management by financial intermediaries follow a normal distribution, accompanied by a continuous relative risk aversion. This assumption is crucial because it allows actors to make different guesses as to the risk involved in a certain position, risks that tend to be randomly distributed around the most probable outcome. It further assumes actors can, based on their guesses, hedge (or insure against) quantifiable risk. However, if the risk is not quantifiable, and if the markets involved have multiple equilibria (a run on the peso, versus no run) for which no probability can possibly be assigned, then of course actors cannot make rational prudential responses to risk. While the IEO office throws a few softballs at the IMF's WHD for failing to anticipate the risks of an unsecured run on the peso (p.60 & following), this was in fact precisely the sort of error the IMF was institutionally designed to analyze well. And it flunked.

After p.62, the IEO report (less appendices and recommendations) tends to be exculpatory of the IMF and its Western Hemisphere Division (WHD); readers unfamiliar with the mission and institutional knowledge of the IMF are unlikely to realize how brazen this is. Argentina's leaders operated under extreme political constraints, and without the detailed knowledge of finance for which the IMF is famous. And yet the IEO largely hands off blame for poor judgments to its ward, rather than the master. Needless to say, by the time the crisis broke there was next to nothing the IMF could have done without massive intervention from a combination of powerful treasury departments.

At this time, two diagnoses were possible regarding Argentina’s protracted recession
and loss of market access. One was to view them primarily as a liquidity crisis resulting from adverse but temporary shocks. According to this interpretation, growth could return shortly, if some confidence-enhancing policy adjustments were implemented...
An alternative diagnosis would have been to view the slowdown in economic activity as resulting from an exchange rate that had become significantly overvalued because of a series of adverse shocks. According to this interpretation, adjustment would call for either a nominal devaluation or a substantial price deflation, each with adverse implications for debt sustainability.

The appropriate response to Argentina’s request for IMF support depended critically on which diagnosis was correct. If the country were indeed facing a liquidity crisis, and had good prospects for regaining market access on appropriate terms in the near future, the provision of large IMF financing, combined with some adjustment, was warranted...

Of course, it was also possible that the country needed some orderly distribution of liability for what was actually a colossal error by financial intermediaries, not entirely the Argentine government's fault. In the event, the IMF did organize a gigantic bailout package that was backed by the World Bank Group and the government of Spain. By this time, one of the barriers to a shift away from the dollar peg was the pervasive use of US currency and US-denominated assets as quasi-money in the economy. Unfortunately, commerce with the USA was insignificant relative to the dollar exposure of the country.