The International Monetary Fund: Has Congressional Conditionality Made It More Effective?


My name is Ian Vásquez and I direct theProject on Global Economic Liberty at the Cato Institute. I wouldlike to thank Vice Chairman for Oversight Crapo for inviting me totestify before the U.S. Senate. Let me note, in keeping with thetruth in testimony requirements, that the Cato Institute does notreceive government money of any kind.

Dissatisfaction with the way the InternationalMonetary Fund has handled financial crises in Asia and Russia ledCongress to attach conditions to the increased funding it rewardedthe IMF last fall. Those conditions are intended to addresslegitimate concerns about transparency, accountability, IMFconditions and advice, and the overall effectiveness of IMFbailouts. In assessing the extent to which IMF programs in crisiscountries have been consistent with congressional intent, it isimportant to also identify the extent to which we can expectCongress's measures to address the fundamental problems of thelending agency. There is unfortunately little reason to believethat the IMF today does not still suffer from the flaws thatplagued it before Congress attempted to make the Fund moreeffective. The IMF continues to make matters worse in Asia andelsewhere by perpetuating moral hazard, hindering rapid andwidespread reforms, and undercutting superior, less expensivemarket solutions.

Moral Hazard Is Still A Problem

A major and widely recognized problem with IMFintervention -- that it establishes moral hazard -- has not beenresolved or adequately addressed by the new legislation. As long asthe IMF bails out countries, nations will continue to slip intocrises in the future because the bailouts encourage risky behavioron the part of governments and investors who fully expect that ifanything goes wrong, the IMF will come to their rescue.

The Fund has played a large role in creating thefinancial turmoil of recent years. With every election cycle since1976, for example, Mexico has experienced a currency crisis causedby irresponsible monetary and fiscal policy. Each episode has beenaccompanied by U.S. Treasury and IMF bailouts, each time inincreasing amounts. In Mexico, everybody has come to expect afinancial rescue at the end of each presidential term.i And although IMF and U.S.officials had since 1995 claimed the last Mexican bailout asuccess, its legacy has been the Asian crisis of today -- at leastin its degree and severity. Indeed, the bailout of Mexico was asignal to the world that if anything went wrong in emergingeconomies, the IMF would come to investors' rescue. How else can weexplain the near doubling of capital flows to East Asia in 1995alone?

Governments in Asia were not discouraged frommaintaining flawed policies as long as lenders kept the capitalflowing. Lenders, for their part, behaved imprudently with theknowledge that government money would be used in case of financialtroubles. That knowledge by no means meant that investors did notcare if a crisis erupted; but it led to the mispricing of risk anda change in the investment calculations of lenders. Thailand,Indonesia, and South Korea, after all, shared some common factorsthat should have led to more investor caution, but didn't. Thosefactors included borrowing in foreign currencies and lending indomestic currency under pegged exchange rates; extensivelyborrowing in the short term while lending in the long term; lack ofsupervision of borrowers' balance sheets by foreign lenders;government-directed credit; and shaky financialsystems.ii Thefinancial crisis in Asia was created in Asia, but the aggravatingeffect of moral hazard was extensive.

Efforts to "promote policies that aim at appropriateburden-sharing by the private sector so that investors andcreditors bear more fully the consequences of theirdecisions"iii may help chasten investors and promotegreater prudence. However, there is no reason creditors should notbear the full consequence of their decisions or why the IMF shouldbe in the position to determine what those consequences should be.If the IMF continues its bailout function, moral hazard will stillexist and the Fund's efforts to make lenders pay for their badjudgement will merely amount to a sloppy approximation of a penaltythat the market would have imposed in the absence of IMFintervention.

Moral hazard -- which is difficult, if not impossibleto measure before a crisis erupts -- has likely been reducedsomewhat by Russia's debt default. However, that outcome hasresulted from the failure of IMF lending there, not from an IMFpolicy of burden sharing. In the Russian episode, the IMFunwittingly showed investors that it could be an unreliable bailoutagency, thus creating a greater incentive for market participantsto play by market rules. Evidence for this was provided by thedevaluation of the Brazilian real, which also occurred after an IMFbailout but had no dramatic effect on international financialmarkets. International investors may be more cautious today, butthey are still insulated from full exposure to market forces.

The IMF's Credibility Problem Continues ToPlague Economies In Crisis

The second major problem with IMF bailouts is thatthey are expensive, bureaucratic, and fundamentally unjustsolutions to currency crises. The Fund's approach helps explain whythe Fund's conditionality lacks credibility and why reform effortsare unlikely to improve its performance. In the first place, thefinancial aid cuts investors' losses rather than allowing them tobear the full responsibility for their decisions. Just as profitsshould not be socialized when times are good, neither should lossesbe socialized during difficult times. "The $57 billion committed toKorea," Jeffrey Sachs observes, "didn't help anybody but thebanks."ivUnfortunately for the ordinary Asian citizens who had nothing to dowith creating the crisis, they will be forced to pay for the addeddebt burden imposed by IMF loans.

IMF bailouts pose another burden on ordinary citizensbecause they don't work very well. The Fund's money goes togovernments that have created the crisis to begin with and thathave shown themselves to be unwilling or reluctant to introducenecessary reforms. Giving money to such governments does not tendto promote market reforms, it tends to delay them because it takesthe pressure off of governments to change their policies. Rather, asuspension of loans will tend to concentrate the minds ofpolicymakers in the various troubled countries. The reason, afterall, that there is any talk today of market-reform is not becausethe IMF has shown up and suggested it is a good and necessarything. That is fairly obvious. Economic reality is forcing thelong-needed change. To the extent that the IMF steps in andprovides money, those reforms will not be as forthcoming. Thus, thecitizens of recipient nations suffer the added burden of IMFintervention. Not only do they have to pay a greater debt; but theyalso have to suffer prolonged economic agony that is produced bythe Fund's bailouts.

That has certainly been the experience of Mexicosince becoming a recipient of IMF aid in 1995,v and appears to be the case as wellin Korea, another country in which the Fund is claiming success.Although progress has been made there, backtracking in importantareas such as bankruptcies and privatization has occurred. Afterhaving promised to quickly privatize state-owned monopolies, theKorean government has put off such moves for the energy,telecommunications and tobacco industries until As the EconomistIntelligence Unit reported in February 1999, banks are more tightlycontrolled by the government today than when the crisis began andthe chaebol, or Korean conglomerates, dominate the economy evenmore.

"The bottom line is that the chaebol virtually are theeconomy. In 1998 the top seven accounted for more than half of allSouth Korean exports. The biggest are too big to fail; they knowit, and they know the government knows it too. Indeed, chaeboldominance means they can continue to joust with the government overreforms. These basic truths of the balance of power limit KimDae-jung's ability to impose change. In the final analysis, stateand business in Seoul remain as intertwined as they have ever been.Their interdependence will and must adapt to a changing world. Buta truly free market is a chimera." vii

In Thailand the story is much the same. Progress hasoccurred, but more would have likely occurred without IMF lending.Critical bankruptcy legislation has been postponed until the secondhalf of 1999 after the Thai government had agreed with the IMF thatit would be implemented by October of 1998.viii The Fund's money inIndonesia has allowed the government to halt progress on reform,forcing the IMF to delay its flow of credit there. Earlier thismonth the government postponed a plan to restructure the bankingsystem, which would have closed dozens of bankrupt banks, many ofwhich have close political connections to the government. Onemarket-oriented Indonesian economist, Rizal Ramli, noted, "With thefall of Suharto, you might have expected that Suharto-isms would bereduced. But it hasn't worked out that way. The game is the same,only the players have changed."ix

But what about the Fund's "strong conditionality"?Don't the strict conditions of IMF lending ensure that importantpolicy changes will be made? It has become clear that domesticfactors, not outside aid agencies, tend to determine whether acountry reforms or not. In one study on multilateral lending,Oxford University's Paul Collier explains that "some governmentshave chosen to reform, other to regress, but these choices appearto have been largely independent of the aid relationship. Themicro-evidence of this result has been accumulating for some years.It has been suppressed by an unholy alliance of the donors andtheir critics. Obviously, the donors did not wish to admit thattheir conditionality was a charade."x

The IMF's own record -- 86 countries have beenrelying on IMF credit for more than 10 years -- shows that theconditionality of the Fund's "short term" loans has not worked wellin the past. But besides the Fund's poor record, there is goodreason why the IMF has little credibility in imposing itsconditions. As we have seen with Russia over the past severalyears, a country -- especially a highly visible one -- that doesnot stick to IMF conditions risks having its loans suspended. Whenloans are cut off, recipient governments tend to become moreserious about reform. Note that the IMF encourages misbehavinggovernments to introduce reforms by cutting loans off; it is thecut off of credit that induces policy change.

Unfortunately, when policy changes are forthcoming,the IMF resumes lending. Indeed, the IMF has a bureaucraticincentive to lend. It simply cannot afford to watch countriesreform on their own because it would risk making the IMF appearirrelevant. The resumption of financial aid starts the process overagain and prolongs the period of reform. The Fund's pressure tolend money in order to keep borrowers current on previous loans andto be able to ask for more money is well documented.xi The IMF's bureaucraticincentive to lend is also well known by both recipient governmentsand the IMF itself, making the Fund's conditionality that much lesscredible.

Just this week, the IMF agreed to release the secondinstallment of its $41.5 billion package to Brazil despite everyindication that the government will continue to resist long-overduereforms. The IMF bailout package for Brazil, which came after theU.S. Congress passed the Foreign Operations Act of 1999 urgingreforms at the IMF, shows that the Fund's program there has beenflawed from the beginning.

One would have thought that the fall of the Russianruble in August 1998 would have put to rest the notion that the IMFcan prevent financial crises by providing bailouts before turmoilerupts. Nevertheless, the Fund provided just such aid to Brazillast fall with similar consequences. The Brazilian government hashad years since the Mexican peso devaluation to put its house inorder and get its spending under control. Only after investorsbegan to lose confidence in the real in the fall of 1997 didPresident Fernando Henrique Cardoso react by announcing taxincreases and spending cuts worth $20 billion. Tax increases ofabout $9 billion were implemented, but most spending cuts neverwere. Revenues have increased, but spending has gone up at a fasterrate leading to a growing budget deficit of 8 percent of GDP.

Higher spending and higher taxes have been the normin Brazil during the Cardoso years. Public spending has increasedfrom 30 to 40 percent of GDP since 1994, while the tax burden hasgone from 26 to 31 percent of GDP during that time. The IMF's newausterity plan agreed to in the fall of 1998 promised $11 billionin new tax increases and proposed tax cuts amounting to only 6.3percent of the total budget for 1999.xii

With the budget out of control, Brazil has relied ondisciplined monetary policy to maintain stability. But monetarypolicy alone cannot save Brazil. Brazil's pegged currency, highinterest rates, issuance of short-term, dollar-denominated debt andlack of urgency about reform created an environment that inhibitedself-sustaining growth and eventually led to the devaluation of thereal after the IMF came to the rescue. Brazil's central problem oftoo much government spending is essentially a political problem,one that the IMF is ill-suited to address and appears to beenabling. Those problems are severe. For example, the percentage ofstates' revenues spent on salaries in Brazil range from a low of 40percent to a high of 85 percent (for the states of Minas Gerais,Rio de Janeiro and Sao Paulo the figures are 80, 79.7 and 62.3percent respectively). At the heart of the government'suncontrollable spending is the bankrupt state-run pension system.Two thirds of the budget deficit is due to that program, yet theBrazilian Congress has postponed its reform for years and the"reform" agreed to under the IMF package does little to addressthis structural problem. Although the state pension system had a1998 deficit of around $40 billion, the so-called pension reformfinally passed by Brazilian lawmakers will collect $2.6 billiondollars generated from new taxes on pensions. The measure neitherdeals with fundamental flaws in government spending nor qualifiesas reform.xiii

Brazil is a clear example of a country in which IMFcredit is performing as poorly as it traditionally has despite itbeing made available after the U.S. Congress took steps to improvethe Fund's performance. New IMF lending has served as a sort offinancial morphine for the Brazilian political system, allowing itto continue postponing reforms. There is no reason to believe thatthe latest installment of IMF money will increase the Fund'scredibility in terms of its ability to push Brazil towardsignificant policy change. A more effective approach would rely onthe market.

The IMF Should Stand Aside

IMF bailouts, however, undermine superior, lessexpensive market solutions. No amount of IMF reform in that area,short of an end to its bailout function, can change that reality.In the absence of an IMF, creditors and debtors would do whatcreditors and debtors always do in cases of illiquidity orinsolvency: they renegotiate debt or enter into bankruptcyprocedures. In a world without the IMF, both parties would have anincentive to do so because the alternative, to do nothing, wouldmean a complete loss. Direct negotiations between private partiesand bankruptcy procedures are essential if capitalism is to work.As James Glassman has stated, capitalism without bankruptcy is likeChristianity without Hell. IMF bailouts, unfortunately, undermineone of the most important underpinnings of a free economy byoverriding the market mechanism. As Catherine Mann put it

"In the current situation, the more difficult,drawn-out, ad hoc, and therefore costly are the financial disasterworkouts, the greater are the incentives for investors to demandand institutions to offer instruments ex ante that will help togenerate a market-oriented solution to the workout process. Sorather than intervening more frequently, official institutions muststand aside." xiv

There is simply no reason why international creditorsand borrowers should be treated any differently than are lendersand debtors in the domestic market.

Governments would also react differently if no IMFinterventions were forthcoming. There would be little alternativeto widespread and rapid reforms if policymakers were not shieldedfrom economic reality. Lawrence Lindsey, a former governor of theU.S. Federal Reserve opposed to bailouts, has noted, for example,that, "All of the 'conditions' supposedly negotiated by the IMFwill be forced on South Korea by the market."xv Of course, there is always thepossibility that a government would be reluctant to change its waysunder any set of circumstances; but that is a possibility that islarger, and indeed has become a reality, under IMF programs.

Reform Proposals

Because the Fund suffers from an array of inherenttensions in the way it operates, causes more harm than good once acrisis erupts, and undermines superior market solutions, we shouldnot expect current reform efforts to somehow overcome the Fund'sfundamental flaws, much less to make a noticeable difference in theperformance of IMF client countries. The most important reformCongress has advocated -- but one which may only make a marginaldifference in Fund's effectiveness -- is that of increasedtransparency and accountability at the lending agency.xvi The U.S. governmentand the U.S. public are entitled to know what advice the IMF isgiving countries, when it is giving that advice, under whatfinancial and policy conditions is the Fund providing support, andwhat exact criteria the IMF will use to measure whether a countryis successfully reforming or merely delaying comprehensive marketreforms. The Fund should also publicly identify those institutions,public and private, in recipient nations that are receiving itssubsidized finance. In practice, that means that IMF documents,including economic evaluations, policy prescriptions, letters ofintent, and other memoranda be made public. The Fund should alsoallow outside auditors to assess the quality of IMF loans and toconduct full evaluations of the IMF's lending record.

At the very least, more transparency would providefor a more informed public debate and, in theory, would make theIMF more accountable. There is, however, no reason to believe thatthe IMF would noticeably improve its performance or significantlychange its behavior even if much damning evidence came to light.After all, Washington's other major multilateral lending agency,the World Bank, already conforms with many of the stipulations someare suggesting for the IMF. The Bank's Operations and EvaluationsDepartment and other internal reviews have for years consistentlyturned up self-admitted dismal performance. These have resulted inreforms and promises of reform, but no noticeable improvement inperformance. Congress should expect the same experience from theIMF bureaucracy if it attaches demands for transparency in exchangefor continued or increased financial support. And although Section602 of the Foreign Operations Act calls for the U.S. TreasurySecretary to certify that IMF resources to Korea are not being usedto provide financial assistance to certain industries, there is noway such a certification can be reliably made because of thefungibility of money. IMF funds to the Korean government merelyfree up resources that allow the government to support its favoredprojects anyway.

Even though Congress has asked the U.S. TreasuryDepartment to report whether the IMF is complying withCongressional stipulations, it should not rely on the TreasuryDepartment to provide adequate assessments. Because the TreasuryDepartment, through the U.S. executive director at the IMF, hasinfluence over the use of IMF funds, it is in its bureaucraticinterest to maintain and even increase those resources under itsinfluence. Treasury has long supported the prerogatives of the IMF,regardless of whether U.S. administrations have been Republican orDemocratic. Treasury's disingenuous claims that funding the IMFdoes not cost U.S. taxpayers a dime is only the most recent exampleof its virtually uncritical endorsement of IMF lendingactivities.

Congress's efforts to prevent the Fund from financingbankrupt institutions are part of a larger effort to make the IMFfunction as an international lender of last resort. For example,Congress stipulated that emergency finance be made available at"not less than 300 basis points in excess of the average of themarket-based short-term cost of financing of its largestmembers"xvii and that such credit be made available forshorter terms. Unlike a true lender of last resort that providesfunds at a penalty rate to solvent banks that are temporarilythreatened by panic, the IMF has generally provided subsidizedfunds that bail out insolvent financial institutions. Imposingpenalty interest rates alone will not discourage imprudent behaviornor allow the Fund to be an effective lender of last resort. Indeedthe agency cannot act quickly nor create money as can true lendersof last resort. Countries that experience threats to theirfinancial systems can already rely on their own central banks aslenders of last resort. That includes the United States, where theFederal Reserve is charged with such a mission.xviii Turning theIMF into an international lender of last resort is both unnecessaryand politically improbable.

Another reform urged by the U.S. Congress is for theIMF to promote policies aimed at "strengthening crisis preventionand early warning signals through improved and more effectivesurveillance of the national economic policies and financial marketdevelopment of countries."xix Yet it is unclear how a warning mechanismwould work. As economist Raymond Mikesell asks, "Who would bewarned and when? As soon as the financial community receives awarning that a country is facing financial difficulty, a massivecapital outflow is likely to occur, in which case crisis preventionwould be out of the question."

On the other hand, if the IMF perceives seriousfinancial difficulties in a country and does not disclose thatinformation, then it undermines its credibility as a credit-ratingagency for countries. That appears to have been the case inThailand, where the IMF now claims it issued warnings about theeconomy before the crisis erupted but kept those concernsconfidential. The fund's credibility is further undercut byinherent conflicts of interest: in many cases, it would beevaluating countries in which it has its own money at stake; in allcases, it would be evaluating countries that, as member-owners ofthe IMF, have contributed to the fund's pool of resources. Only byceasing to lend could the agency increase its integrity. At thatpoint, however, its evaluations would merely replicate a servicealready available.

Finally, it is worth noting that the global financialsystem could improve in a number of ways. Countries do need toincrease the transparency of their financial institutions anddisclose economic data. Financial deregulation and openness toforeign investment in that and other sectors would also reduceinstability and promote prosperity in developing nations. Thatwould imply an end to policies of directed credit and the cronycapitalism that such policies often engender. The establishment ofbankruptcy procedures and the rule of law are requisite toencourage sustained investment, both foreign and domestic. Freeexchange rates and disciplined monetary and fiscal policy are alsonecessary to help avoid crises. Currency boards or dollarization asalternatives to floating exchange rates have also been consistentwith stability and economic growth in the countries that haveadopted those mechanisms.xx In general, economic liberalization will helppromote prosperity and stability in the global economy. But all ofthose reforms can be unilaterally introduced by developing nations.Such policy change does not require IMF intervention. Reforms atthe IMF, which promise little, are thus a distraction to theimportant tasks developing countries are more likely to undertakewithout IMF money and advice. The conditionality that Congress hasattached to increased funding of the IMF has shown itself to beabout as effective as IMF conditionality itself.


i. W. LeeHoskins and James W. Coons, "Mexico: Policy Failure, Moral Hazard,and Market Solutions," Cato Policy Analysis no. 243, October10, 1995. The authors claim that the result of the Mexican bailout"is a set of perverse incentives for Mexican officials and foreigninvestors that ensures the 'crisis' will reappear on an even largerscale."

ii. Some ofthese points were made by Allan H. Meltzer at a Cato Policy Forum,"Why We Should Say No To the IMF," Washington, D.C., February 12,1998. See also Allan H. Meltzer, "Danger of Moral Hazard,"Financial Times, October 27, 1997.

iii.Section 610 of Title VI of the Foreign Operations, ExportFinancing, and Related Programs Appropriations Act of 1999,Congressional Record, October 19, 1998, p. H11104.

iv. Quotedin Peter Passell, "Economic Scene," New York Times, February 12,1998.

v. See IanVásquez and L. Jacobo Rodríguez, "What To Expect FromIMF? Look At Mexico," Investor's Business Daily, April 1, 1998.

vi. EvelynIritani, "South Korea's Economy Offers Illusion of Reform," LosAngeles Times, September 13, 1998, p. A1.

vii.Economist Intelligence Unit, "South Korea: On the leading edge offree-market reform?" February 17, 1999.

viii."Crisis Highlights Risks To Investors," Bangkok Post, January 27,1999.

ix. MarkLandler, "Bank Reform Issue Hobbles Indonesia," InternationalHerald Tribune, March 5, 1999.

x. PaulCollier, "The Failure of Conditionality," in Catherine Gwin andJoan Nelson, eds., Perspectives On Aid and Development (Washington:Overseas Development Council, 1997), p. 57.

xi. See,for example, John Williamson, The Lending Policies of theInternational Monetary Fund (Washington: Institute forInternational Economics, 1982); Roland Vaubel, "Bureaucracy at theIMF and the World Bank: A Comparison of the Evidence," The WorldEconomy, March 1996; and Peter B. Kenen, Ways to Reform ExchangeRate Arrangements (Princeton, N.J.: Princeton University Press,1994).

xii. SeeL. Jacobo Rodríguez, "Bailout Will Handicap Brazil," Journalof Commerce, Novermber 17, 1998.


xiv.Catherine L. Mann, "Market Mechanisms to Reduce the Need for IMFBailouts," International Economics Policy Briefs no. 99-4, February1999.

xv.Lawrence B. Lindsey, "The Bad News About Bailouts," New York Times,January 6, 1998.

xvi. SeeSection 601 (2), (3) and Section 610 (13), Congressional Record pp.H11102 and H11105.

xvii.Section 601, (4) (A), p. H11102.

xviii. For more on international lenders of lastresort, see Charles Calomiris, "The IMF's Imprudent Role As Lenderof Last Resort," Cato Journal (Winter 1998) vol. 17, no. 3, pp.275-294; Anna J. Schwartz, "Time To Terminate the ESF and theIMF," Cato Institute Foreign Policy Briefing no. 48, August 26,1998; and Jim Saxton, "An International Lender of Last Resort, TheIMF, And the Federal Reserve," Joint Economic Committee of the U.S.Congress, February 1999.

xix.Section 610, p. H11104.

xx. See,for example, Steve Hanke and Kurt Schuler, "A DollarizationBlueprint for Argentina," Cato Institute Foreign Policy Briefing,March 12, 1999.

Ian Vásquez

Subcommittee on International Trade and Finance
Committee on Banking, Housing, and Urban Affairs
United States Senate