Printer Friendly

International Financial Institutions and financial accountability.

Calls for the reform of International Financial Institutions (IFIs) have become both increasingly common and more emphatic. The former International Monetary Fund (IMF) executive director Ariel Buira, for example, concluded that important aspects of the governance system have become dysfunctional because the quota structure no longer adequately reflects changes in the global economy since 1944. He proposed that half of the votes should be given to developing countries to achieve a balance between the interests of all members. (1) Others, such as Ngaire Woods, have called for institutional reforms to make IFIs more accountable. (2) Such reforms are indeed necessary. They are also not infeasible, as the voting majority of regional developing country members at the Inter-American Development Bank (IDB)--which has definitely not performed worse than the International Bank for Reconstruction and Development (IBRD)--proves. (3)

Yet while these useful proposals have been widely discussed, the lack of meaningful financial accountability of IFIs has received surprisingly little attention. The few papers that call for financial accountability have been unable to trigger a wide discussion. (4) Considering the substantial damage done by IFIs this is surprising both from an ethical and an economist's point of view. (5)

In this essay, I argue that IFIs must be made financially accountable for what they do. After presenting the idea of financial accountability, I show how reforms that would make IFIs financially accountable could be implemented without much difficulty. I argue, moreover, that for most IFIs, embracing financial accountability would bring their operations closer to the intentions of their founders.


In a market economy anyone must face the economic consequences of their actions and decisions. If consultants give advice negligently or without obeying minimal professional standards, they call be sued and have to pay compensation for the damage they have caused. National liability and tort laws serve the purpose of compensating those suffering unlawful damages, such as those resulting from negligent or unduly risky conduct or from providing dangerous products. But they also deter such behavior. Knowing that one must pay for the damage done by sloppy work or wrong advice that could be avoided if professional standards were obeyed thus serves as a strong incentive to improve the quality of products and services. The success of market economies is based on linking decisions and risks. The systemic point is not that people receive indemnities--which is necessary and good--but that liability and tort laws make providers of goods and services work more carefully. As a result only a small fraction of all market transactions gives rise to lawsuits. Central planners in former communist countries, by contrast, knew that their institution would never be held financially accountable. The designers of the system, who believed that these central planners would work in the best interest of the party and thus the people, thought this perfectly correct. As could be expected, the lack of financial accountability produced extremely bad results.

Two types of financial accountability can be distinguished: external, accountability of corporations to their customers, or state agencies to people; and internal, between firms or legal entities and their own staffs. Employees may under certain conditions be liable to refund to their employers damages that they have caused and that their employers had to pay for, but this is an internal matter. Even if a corporation indemnifies clients, its employees might not have to refund anything if they conscientiously followed rules, procedures, and orders. Clients have the right to sue the company as the party to the contract. Suing the company's minimum-wage worker who physically caused the damages and would have no money to pay for them anyway is of little use. Similarly, I focus on the possibility of suing IFIs, not their employees, or on the relation between IFIs and their clients as well as people who suffered damages because of unlawful IFI behavior.

The Organisation for Economic Co-operation and Development (OECD) identified accountability as essential for successful development efforts, as "vital and urgent" for any society. (6) Accountability, transparency, and high standards of public-sector management are "basic values in their own right," but also "means of developmental ends." The OECD defined "the need for transparency and accountability, both within the aid agency and among donor agencies and development partners," and good governance as "key operational implications" of its new concept of aid] Advocating both the need and scope for strengthening market mechanisms, the donor community and IFIs have stressed the principles of good governance, democratization, participatory development, transparency, and accountability. These principles include sensible economic and social policies, financial accountability, and the creation of a market-friendly environment.


Given the importance of these principles, it seems reasonable to demand that efficient management also be required of bilateral aid and multilateral development cooperation institutions. To remain consistent with their commitments, one would expect donor countries to have introduced appropriate management and accountability standards at their own national agencies and at all IFIs that they control by voting majorities. However, this has not occurred. Indeed, since the demise of communist economies, IFIs and aid donors are alone in escaping financial accountability. Consequently, developing countries and the poor remain totally unprotected against negligently or even willingly inflicted damage. Even worse, errors and negligently done damage tend to increase the importance of 1Fis, since damages caused by one project or adjustment program call for a new loan to repair them, thus increasing IFI income--in other words, "IFI-flops create IFI-jobs." (8)

Such perverse outcomes are economically unjustifiable. Several telling examples demonstrate the difference in the rules applied to normal market actors and IFIs. In 2003 a German court in Muenster ordered a bank to compensate a client whom it had advised to buy Argentine bonds as high yielding yet safe investments. (9) The court followed the plaintiff's argument that the bank did not explain Argentina's well-known difficulties adequately and ordered the bank to indemnify fully its client because of the advice it had given. In another case, a British couple that borrowed money from Lloyds sued the bank successfully because its manager had advised and encouraged them to renovate and sell a house at a profit. The High Court ruled that the manager should have pointed out the risks clearly and should have advised them against the project. Because the bank had gone beyond mere lending by giving advice, Lloyds had to pay damages when prices in the property market fell and the couple suffered a loss. (10) Orange County sued Merrill Lynch for $2 billion; Bank Austria sued Price Waterhouse for 147 million [pounds sterling], arguing that the firm had not checked Sovereign Leasing, a firm Bank Austria invested in, with sufficient care. The point is not whether courts found in favor of plaintiffs or defendants, but that legal redress is possible. Indeed, the principle that anyone suffering or alleging to suffer damages due to another's fault or because of failures to observe a purely equitable duty must be able to seek redress is firmly established in OECD countries--which are all committed to a market economy. If governments or their agents create damage by negligence--by failing to exercise their duty of care or by not obeying professional standards--or by acting illegally, they can be sued. If consultants fail to respect professional standards or to work properly,

they can be taken to court. Victims have a right to be compensated. The only exception of this generally accepted rule is development cooperation, the last sphere where damage can still be inflicted with impunity and even financial gain. If normal accountability standards applied to Southern debtors there would be no multilateral debt problem. The difference may be illustrated with the incident about which Joseph Stiglitz had heard: an IMF country team had copied large parts of the text for one country's report into another country's report, leaving the original country's name in a few places. (11) If IFIs were German or English banks and debtor countries their clients, a substantial chunk of multilateral debts would in all likelihood be gone because IFIs would be forced to shoulder their part of responsibility for the outcome.

Shortcomings of aid administrators and IFIs are sometimes acknowledged, but without further consequences. (12) Some twenty years ago the IBRD stated:
 Genuine mistakes and misfortunes cannot
 explain the excessive number of 'white elephants'.
 Too many projects have been selected
 either on the basis of political prestige or on
 the basis of inadequate regard for their likely
 economic and financial rate of return....
 External financial agencies have shared the
 responsibility for this inadequate discipline
 over the use of investment resources. (13)

Unfortunately, the word "shared" is notional at best. External financial agencies have never shared financial consequences of genuine mistakes and misfortunes. Countries have always had to pick up the whole bill alone, while "external financial agencies" have often granted another credit to redress damages done by their first credit, thus earning income from both loans. Nor do IFIs necessarily stop payments after they have learned that part or all of the money will be diverted to private pockets or otherwise misused. Jeffrey Winters, for example, has argued that the IBRD did not take proper measures against corruption in Suharto's Indonesia, and continued its lending when it was widely alleged that about a third of all loans would be embezzled. (14) The IMF's reaction to the so-called Blumenthal Report is another example. In 1982 the German expert Erwin Blumenthal, seconded by the Bretton Woods institutions to Zaire's central bank, warned most outspokenly and in writing that Zaire should not get any further money due to widespread corruption. Nevertheless, in 1983 the IMF allowed Zaire the largest drawing that had ever been effected by an African government. Predictably, the money disappeared. Until 1989 the IMF trebled the volume of Zaire's drawings. Under the existing anti-market system it was good business for the Fund and marvelous for Mobutu's clique, but not for Zairians. Of course, Zairians were expected to pick up the bill.

In short, all IFIs claim the privilege of absolute immunity from any financial responsibility for their own actions, decisions, and omissions. Moreover, in contrast to developing countries that feel the economic consequences of inadequate implementation of projects and programs, external financial agencies have benefited economically from the very disasters they have helped to create or even created themselves. Claiming the status of preferred creditors and benefiting from the total absence of any financial accountability for their decisions, IFIs profit from their own errors at their clients' expense. The Bank "now not only admits its mistakes, but has enshrined learning from them as part of their corporate philosophy." (15) These assertions have, however, not left very perceptible traces in the Bank's operations. Apparently, there are no mechanisms ensuring that failures are corrected adequately. Under its previous managing director, Horst Kohler, the IMF followed suit, establishing a learning culture too. But if these institutions have learned, poor countries and vulnerable groups in particular have paid their tuition.

The Asian financial crisis of 1997 revealed another grave problem. Until the crisis broke, both Bretton Woods institutions encouraged further capital account liberalization. The IMF wanted to change its Articles of Agreement to allow it to do what it had done in open breach of them already: force member countries to liberalize capital accounts. Warning signs of the pending crisis were ignored. (16) IBRD staff trying to ring the alarm bell were disregarded. During a visit to Indonesia in the antumn of 1997, 1BRD president James Wolfensohn himself removed a passage by the resident mission that warned of serious problems, "substituting it by even more fulsome endorsement of Indonesia as an Asian miracle." One typically did not want to hear news going against one's ideological preferences--free private capital markets had to be proved right by Asian countries. (17)

In 1999, though, the IBRD acknowledged having known "the relevant institutional lessons" since the early 1990s. (18) An audit report by its Operations Evaluation Department "on Chile's structural adjustment loans highlighted the lack of prudential supervision of financial institutions in increasing the economy's vulnerability to the point of collapse." (19) The Operations Evaluation Department's "key lesson" that "prudential rules and surveillance are necessary safeguards for the operation of domestic financial markets, rather than unnecessary restrictions," did not make "policy makers and international financial institutions give these weaknesses appropriate weight." ((20) Despite what was already known about quick capital account liberalization, they continued to encourage it in Asia. According to the Bank the policy-makers were "guided" by "the lessons of the general debt crisis" (whatever that might mean), not by the "more relevant" cases of Chile and Mexico in 1994-95. The neglect of proper sequencing and institution building "featured prominently in the Chile and Mexico crises." (21) Briefly, the problem was recognized years before the crash, and the unfolding of the Asian crisis could be watched like a movie whose script is known. But instead of pointing out these risks clearly and warning Asian countries--as the High Court would have demanded from a British bank giving advice--IFIs encouraged their Asian members to carry on policies that would in all likelihood produce crashes.

Keenly preaching human rights and respect for private (especially foreigners') property, IFIs and donors have frequently financed projects that violate these principles. During the project cycle they have also taken decisions that produce such results. The right of victims to make official lenders financially accountable for what they do is needed to improve the lot of the poor, whose human rights--even the most basic right to life--are too often considered unworthy of respect by their governments and their governments' public financiers.


Impunity and the possibility of gaining from crises can be strong incentives to violate one's own statutes. If powerful members acquiesce or even applaud such violations, overriding the membership rights of weaker members has no negative consequences. Pursuant to its presently valid articles of agreement, any IMF member has the right to choose policies that differ from the usual, fairly uniform IMF prescription usually summed up as the Washington Consensus. Article IV(3)(b) states: "These principles [general obligations of members pursuant to Article IV(l)] shall respect the domestic social and political policies of members, and in applying these principles the Fund shall pay due regard to the circumstances of members." (22)

In contrast to the conditionality foisted onto members in distress, the IMF's constitution not only allows capital controls but even explicitly restricts the use of Fund resources to finance outflows. Article VI(3) establishes the right of members to "exercise such controls as are necessary to regulate international capital movements, but no member may exercise these controls in a manner which will restrict payments for current transactions." These are defined in Article XXX(d) as "not for the purpose of transferring capital," including "payments of moderate amount of amortization of loans or for depreciation of direct investments" or "moderate remittances for family living expenses" (emphasis added). Although this definition is somewhat opaque, it follows that even restricting such flows is a member's right.

Article VI(1)(a) goes further: a "member may not use the Fund's general resources to meet a large and sustained outflow of capital except as provided in Section 2 of this Article [this refers exclusively to reserve tranche purchases] and the Fund may request a member to exercise controls to prevent such use of the general resources of the Fund." Current transfers can be restricted with the Fund's approval. Although the IMF may, but is not obliged to, request controls, these regulations clearly show that it is not supposed to press for liberalization of capital movements in the way it actually did, let alone forbid members to control capital flows. However, when it comes to protecting the rights of non-OECD members, legal regulations and obligations are apparently insignificant. Asian countries not only had the right to control capital outflows in 1997--as the IMF had to admit when Malaysia exercised it (23)--but by forcing members to finance large and sustained outflows by speculators the IMF clearly violated its own constitution.

Corrective measures affecting the balance of payments should be done "without resorting to measures destructive of national or international prosperity" (Article I(v)). This would have been easily possible in Asia if speculators had not been bailed out by having their losses socialized. Article IV(1)(ii) requests the IMF to foster stability and a monetary system that does not produce "erratic disruptions." However, the policy of high real interest rates forced on clients did precisely the opposite. Real interest rates beyond 40 percent are doubtlessly erratic disruptions. Few if any activities apart from drug pushing produce returns high enough to enable entrepreneurs to pay such interest charges. Bankruptcies of domestic corporations and entrepreneurs are the logical result. It is even possible to calculate the cost of such policies: According to Standard & Poor's, nonperforming loans would have surpassed 30 percent of total loans, computed on a three-month basis, if Malaysia had not cut interest rates sharply. (24)

A cynic might even view such crises as in the institutional self-interest of the Bretton Woods institutions. During the Asian crisis the IMF's first deputy managing director Stanley Fischer still argued--using Thailand and Mexico as supporting evidence--that the prospect of larger crises caused by capital account liberalization would call for more resources for the IMF to cope with the very crises the IMF's proposal would create in the future. (25) This demand is easily explained by the perverse incentives created by the present lack of financial accountability. From the narrow point of view of institutional self-interest--which one, of course, hopes to be irrelevant--such crises are better than the use of contractual rights to capital controls, which would not require increased resources from, nor generate additional income for, the IMF.

The IBRD violates its own constitution to the detriment of Southern members. By simply refusing to acknowledge default, even when countries have not paid anything for six or seven years, (26) it creates damages by delaying relief'. In 1992, when the end of the debt crisis was proclaimed and one could argue that insolvency relief was no longer necessary, the Bank itself stated that insolvency was the problem: "In a solvency crisis, early recognition of solvency as the root cause and the need for a final settlement are important for minimizing the damage.... protracted renegotiations and uncertainty damaged economic activity in debtor countries for several years." (27) The IMF's first deputy managing director Anne Krueger clearly acknowledged damages caused by countries waiting too long to opt for insolvency procedures. (28) It was conveniently forgotten that the Bretton Woods institutions themselves had ardently lobbied against debt reductions, arguing that countries would grow out of debts, repeatedly supporting this claim with highly optimistic forecasts of future export earnings.

Economically, this is easy to understand. Obeying one's own constitution by recognizing losses and by using one's loan loss reserves for the purpose for which they have been created would mean losing money and probably defusing further crises. Violating one's statutes means further crisis management, additional income and importance, and new jobs. Creditor countries have repeatedly proved their eagerness to reward this behavior. The Baker Plan for debt management, the Miyazawa-Brady Plan for debt reduction, and all Paris Club debt reduction schemes have reinforced the role of the Bretton Woods institutions by increasing their importance as debt managers. The International Finance Facility recently proposed by the U.K. Treasury would make payments conditional on not having prolonged arrears to the IMF, thus further strengthening its privileged creditor position. Economically and ethically incorrect behavior is rewarded by the powerful. This makes reestablishing the rule of law, empowering economic sense, and showing simple decency all the more necessary.


All IFIs have charged the costs of loan loss provisions to their clients. This is a normal, economically sound, and commendable business practice among lenders. Lenders routinely face a certain amount of losses--just as grocers must cope with the fact that some apples rot before they can be sold. Prices or tees charged to clients must include margins to cover such losses, since they are part and parcel of doing business. IFIs have charged such margins and built up reserves, but have refused to use them for their intended purpose. In other words, their clients have been made to pay insurance pre-mia without getting benefits once damage has occurred. This was not the intention of the founders of multilateral development banks. Indeed, the IBRD's very statute asserts that financial accountability is necessary and possible, its Articles of Agreement recognize default as a fact of life. Article IV.6 demands a special reserve to cover what Article IV.7 calls "Methods of Meeting Liabilities of the Bank in Case of Defaults." If this reserve proves to be insufficient, "other reserves, surplus, and capital available to the Bank" can be used. Finally, appropriate amounts of unpaid subscriptions of members can be called. As the Bank is only allowed to lend either to members or to other borrowers if member states fully guarantee repayment (Article III.4), the logical conclusion is that default of member states was definitely considered a possible, and maybe even an occasionally necessary, solution. The IBRD's founders understandably wanted lending to be subject to some market discipline, and designed mechanisms that would allow the Bank to shoulder its fair share of the risks involved. Contrary to their founders' intentions, however, the IBRD has refused to use these mechanisms, wrongly claiming that doing so would make development finance inoperational. This is clearly false. The European Bank for Reconstruction and Development (EBRD) writes off losses and submits to arbitration (also foreseen for the IBRD)--which proves that multilateral development banks, if properly managed, can survive financial accountability and market risk.

The IBRD has also argued that it should always enjoy preferred creditor status because its loans at near-market terms have lower interest rates than countries would be able to find elsewhere. This argument has some merit, but slightly better financial terms of loans--even the low fees of the International Development Association (IDA) credits--do not necessarily make them cheaper. If countries must pay for wrong decisions by the IBRD, loans might in the end turn out to be much more expensive than borrowing at market terms, a fact also recognized by the IBRD's own evaluation method. Unlike commercial banks, though, the IBRD is strongly involved in shaping projects and programs. The IBRD has massively influenced the use of loans and the adoption of policies it thought appropriate. (29) IFIs have repeatedly acknowledged that they have taken important decisions--they must not be allowed to refuse shouldering the risks connected with these decisions.

All regional multilateral development banks provide for default in a similar way. All have the authority to modify the terms of loans other than the currency of repayment. The Agreement Establishing the Inter-American Development Bank provides for "Methods of Meeting Liabilities of the Bank in Case of Defaults" (Article VII(3)). Charges should first be made "against the special reserve provided for in Article III, Section 13," which is to meet the IDB's liabilities in the case of debtor default. (30)

The Agreement Establishing the Asian Development Bank similarly demands a special reserve to meet liabilities in the case of default (Article 17). Article 18 gives the detailed description of how to proceed already known from the IBRD (Article IV) and the IDB (Article VII)). (31)

In the case of the African Development Bank, these rules are apparently enshrined in Articles 20-22. But since the African Development Bank does not grant access to the text of its own statutes--a remarkable fact that suggests the "quality" of its governance--this can only be inferred from headings, which are displayed on its home page. Clicking the links, though, brings one straight to the Bank's "visions," rather than the text of the statutes, a special case of "transparency." (32)

IDA's Articles of Agreement are somewhat vague. Pursuant to Article V(3), titled "Modifications of Terms of Financing," IDA may "agree to a relaxation or other modification of the terms on which any of its financing shall have been provided." In the case of maturities of thirty-five, forty, or even twenty years with ten-year grace periods and "no interest charge," (33) this leaves little realistic alternatives but outright grants. Meanwhile, grants are estimated to amount to somewhere between 10 percent and 25 percent of the IDA 13 portfolio depending on countries' absorptive capacity and country performance. The argument that amortizations are needed to refill IDA funds, which would preclude debt relief, is thus no longer valid. Grants to finance basic education or health (34) do not create reflows, just like canceled IDA debts. In both cases no money flows back. Since grants to finance IDA activities are possible without harming the functioning of IDA, debt relief is logically possible as well.

As conditionality was initially not foreseen for IMF drawings, loan loss provisions were unnecessary. Initially lending without any conditions attached to its money, the IMF acted as an emergency source of finance providing short-term liquidity support on a comparatively small scale. Under these circumstances unconditional repayment can be justified. Nevertheless, no preferred creditor status was enshrined in its statutes. When conditionality was introduced, no appropriate changes in order to make the IMF financially accountable for its decisions were made. This led, however, to repayment problems that induced the IMF to introduce a "burden-sharing mechanism" in 1986 to distribute losses of income equally among its members. Under this mechanism, lenders get lower and borrowers pay higher interest rates. Amounts "collected in this way are refunded when overdue interest charges are settled." (35) lust as in centrally planned economies, the IMF's profits are thus determined in advance and its rates set accordingly: the executive board "sets the basic rate of interest charged on regular lending so as to achieve an agreed net income target for the year ahead." (36)

As of January 31, 2004, Liberia had been in arrears with the IMF since April 1986, Iraq since November 1990, Somalia and Sudan since 1991. Between 1978 and 1989 nineteen countries were in arrears with the Fund at least once, excluding cases that were settled without formal complaints to the executive directors or where complaints were outstanding for less than thirty days. The 1986 audit raised the possibility that the next one might have to be qualified--that is, the audit would have to warn about the real value of some debts still booked at face value, with negative effects on the IMF's standing as a creditor because it had failed to provide against losses--if the Fund did not take clear steps to recognize the poor quality of some assets and claims. (37) Forced by external auditors, the IMF started to provide for non-payment by building up loan loss provisions. Of course, these resources are not called loan loss provisions, presumably because that might lead people to conclude that the IMF thinks that losses are unavoidable. Rather, they are called "precautionary reserves." The Fund's precautionary reserves consist of General and Special Reserves and the SCA-1 (Special Contingent Account). (38) The IMF's margin includes a surcharge to "generate resources for a SCA-1, established specifically to protect the IMF against the risk of loss of principal resulting from arrears." (39) What all this complicated wording means is simple: the IMF has already charged its clients the costs for providing against loan losses. This surcharge was 0.1 percent on average in fiscal year 2003. While charging members the costs of defaults, the IMF refuses to use this money when members become incapable of paying--this is wholly unjustified. It is like an insurance company charging necessary fees but refusing to cover damages. Unlike IFls, no insurance company would be allowed such behavior.

This is not to criticize the Fund for establishing loan loss provisions. On the contrary, provisioning is economically sound and financially very prudent. Although the IMF's Articles of Agreement do not demand it, it is not prohibited--in contrast to making capital account liberalization a loan condition. The IMF's "precautionary balances" were about SDR 6 billion, or 8.5 percent of credit outstanding, as of the end of October 2003. (40) It had decided to bring them to SDR 10 billion. All other IFIs have much higher "precautionary balances," ranging from slightly more than 20 percent (IBRD) to over 30 percent (Asian Development Bank) of credit outstanding. Under HIPC II the IMF has already, though reluctantly, agreed to reduce its claims via grants.


Economically viable projects earning their amortization and interest payments pose no problem. But if projects go wrong, there is a need to allocate costs unless IFIs have worked with all appropriate care and obeyed all requirements set by professional standards. Borrower and lender(s) may immediately agree on sharing costs fairly, which would settle the issue quickly and simply. If they do not, the solution used between business partners or transnational firms and countries in cases of disagreement could be applied: the decision of a court of arbitration. (41) Institutions of this kind have worked effectively in the field of international investments. If disagreements between transnational firms and host countries can be resolved through arbitration there is no reason why it would not also be effective for resolving disputes between IFIs and borrowing countries.

A permanent international court of arbitration would be ideal, for which developing countries and IFIs would be permitted to nominate the same number of members. These members would then elect one further member to reach an uneven number to assure majorities in order to avoid deadlocks. If necessary, this court might consist of more than one panel established in the way proposed above. It would determine whether IFIs are at fault and have to pay for (part of) the damage. If so, the simplest way would be to waive a percentage of the loan to cover damages for which the IFI is responsible. The right to file complaints would be conferred to NGOs, governments, and international organizations. Because NGOs are less trader pressure from IFIs and member governments, their right to file complaints on behalf of affected people is particularly important. Giving nonofficial entities the right to start cases is not new. NGOs can already file complaints at the IBRD's Inspection Panel and the IDB's investigation mechanism. The court of arbitrators would, of course, have the right and the duty to refuse to hear poorly prepared or obviously ill-founded cases.

The possibility of being held financially accountable would act as an incentive for IFIs to perform better. If IFIs were well managed, damages would largely be avoided. Thus the costs of damage compensation would remain quite low, but the benefits of projects would be increased. Even if a small surcharge were to be added to lending, its costs would definitely be lower than the costs of catastrophic failures financed at a few basic points less and the costs of the additional loan needed to repair damages done by the first. If, on the other hand, it is argued that this surcharge would have to be high because of the high probability that IFIs would continue to cause a lot of damage, this would be a further convincing reason to reform them rather than let them go on inflicting damages on the poor.

The IBRD's "General Conditions" (Section l0.04) foresee arbitration as the means to settle disagreements with borrowers (whether members or not) for "any controversy between the parties to the loan agreement" or "any claim by either party against the other" not settled by agreement. Bank and country (guarantor country and borrower if these are not identical) appoint one person each. Both sides agree on a third arbitrator. (42) The president of the International Court of Justice or the UN secretary-general appoints this umpire if parties fail to agree. Thus, arbitration could be applied if governments or other borrowers claim damages suffered due to the Bank's or IDA's negligence--the procedures for making the Bank financially accountable need not be invented. Naturally the International Finance Corporation should also be subject to arbitration.

The IBRD's Articles of Agreement (Article VII.3) allow actions against the Bank except by members or persons acting for or deriving claims from members. Property and assets are "immune from all forms of seizure, attachment or execution before the delivery of final judgment against the Bank" (emphasis added). Actions may be brought against the Bank in courts of competent jurisdiction in the territories of members in which the Bank has offices, appointed agents for the purpose of accepting service or notice of process, or issued or guaranteed securities. Accountability was not initially meant to be removed. Suing the Bank before national courts was therefore considered technically feasible.

All other IFIs' statutes contain similar rules. (43) One might argue, though, that a specialized panel, more familiar with development work, would be a better choice. (44) From a purely technical perspective any independent entity--courts or arbitration panels--could perform this task. Ad hoc panels for each case would, of course, be feasible, as dispute settlement under NAFTA or within the WTO framework illustrates. But as cases will continue to come up in the future a standing institution would save the effort of establishing many ad hoc panels. Its continuity could also be helpful in forming legal standards. (45)

If national courts were used as suggested by the IBRD's Articles of Agreement, the national laws of the jurisdiction chosen would determine procedural questions. If cases were taken to an already existing international institution, such as the International Court of Justice, or to a new panel affiliated with it, its procedural norms would be applied. A totally new institution could draw from existing procedures and blueprints provided, for instance, by the UN Commission on International Trade Law. Unlike the IBRD's in-house Inspection Panel this court or panel must not be restricted to investigating conformity with an IFI's operating norms, but must be free to examine these norms too. While presenting facts and proofs should be the duty of the parties, the court/panel should also have investigative powers, exercising them if this is considered necessary. It should be able to invite outside opinion, for instance, in the form of amicus curiae briefs.

Cases would be initiated when a complaint is filed by a person or group alleging to have suffered damages unlawfully inflicted by an IFI. (46) Naturally, the mere occurrence of damage or errors would not be sufficient to receive compensation. Even the most conscientious and cautious work cannot always avoid negative outcomes. It would have to be shown that the IFI did not act with appropriate care or that it failed to observe the professional standards that are applicable to comparable services. As in tort theory, some form of negligence or irresponsible behavior would be necessary.

Risks arising from events beyond the parties' control would remain with the client. A conscientiously planned and executed project that goes wrong would not give rise to compensation. On the other hand, if projects were pushed through too swiftly (perhaps to meet country targets) or without proper checks (perhaps to grant political favors), IFIs would have to pay damage compensation. Normally, compensation should be paid directly to the injured parties. If the borrowing country itself files complaints, the IFI could simply waive a percentage of the loan to cover damages for which it is responsible. Indemnifying individuals, one might also think of arrangements where the IFI converts part of its claims against the country into domestic currency to pay indemnities to domestic injured parties.

Claimants would have to demonstrate that their rights and interests have been or are likely to be directly affected by a negligent or tortious action of the IFI that has had, or threatens to have, a material adverse effect. Naturally open breaches of contract by IFIs would equally establish a right to compensation. The necessity to prepare cases meticulously would serve as a deterrent against ill-founded claims.


Most IFIs were established to finance projects, such as building dams, harbors, or factories. The IBRD increased the percentage of "programs" by starting "structural adjustment lending" against its constitution, apparently to increase its influence. Clearly, adjustment programs are not for any specific project but target whole sectors or the whole economy. Yet the IBRD's Article III.4.vii.b still stipulates that loans "shall, except in special circumstances, be for the purpose of specific projects." Article V, Section 1(b) of IDA's Articles of Agreement repeats this restriction literally. The IMF was established in 1944 to provide unconditional emergency resources, not to finance any programs. IFIs were not meant to engage strongly in program lending and to intervene in members' economies--rightly so, as their record proves.

Debt sustainability analysis highlights the inefficiency of IFI programs. For decades, overly optimistic forecasts have inflicted damages on member countries, rendering strategies based on such forecasts, especially debt reductions, useless. Recently, the IMF and IDA themselves conceded:
 past experience suggesting a systematic tendency
 toward excessive optimism ... a common
 theme behind the historical rise in
 low-income countries' debt ratios was that
 borrowing decisions were predicated on
 growth projections that never materialized ...
 analysis of projections made by Fund staff
 over the period 199-2001 suggests a bias
 toward over optimism of about 1 percentage
 point a year in forecasts of low-income country
 real GDP growth. The bias in projecting
 GDP growth in U.S. dollar terms, however, was
 considerably larger, at almost 5 percentage
 points a year. (47)

There is doubtlessly a "need for well-disciplined projections, including by laying bare the assumptions on which they are predicated and by subjecting them to rigorous stress tests that explicitly incorporate the impact of exogenous shocks." (48) For some three decades--the first adjustment measures started in Africa in the mid-1970s--the Bretton Woods institutions have consistently made undisciplined projections whose assumptions were not explained and that tailed to take a country's vulnerability to exogenous shocks--such as abrupt changes in export earnings or exchange rates--properly into account. Any normal clients could successfully sue such consultants and get financial compensation. By contrast, those affected by IFI negligence, including many of the world's poorest people, must shut up, pay up, and suffer the consequences.

If programs to restore a country's economic viability are successful then no problem would arise. If they are not or if the crisis gets even worse, something must apparently be wrong. A program that is unable to improve the country's economic performance is obviously not the right remedy. It is practically impossible to determine the fair share of one or more IFIs in failed programs, but there is an easy way of holding IFls financially accountable. If a debt-distressed country becomes insolvent, all creditors should be treated equally. Instead of attempting to determine precise shares of blame in failed programs, all IFIs should lose the same percentage of claims as other creditors. (49) Having at least codetermined the country's policies, IFIs must also shoulder the risk. Since IFIs are more involved in a country's debt strategies than private creditors, one may say that even equal treatment would let them off the hook too easily. Clear cases of lacking professional responsibility, such as Zaire's drawings or Stiglitz's copying incident, must be treated differently even though they happened in connection with programs. In such obvious cases damage compensation should have to be paid if damage ensues, irrespective of whether the country becomes insolvent. They would thus be settled like damages done by projects and by the same court or panel.

The present de facto status of "preferred creditor" that multilaterals enjoy must be abolished. As the shares of multilateral debts are sometimes quite high in the poorest countries, protecting IFIs from losses is done at the expense of particularly poor clients, usually highly dependent on solutions elaborated by IFI staff. Meanwhile IFIs' control over their clients has considerably increased: In 200l the IMF adopted a strengthened framework of measures to supplement conditionality. Its staff now proposes remedies. To prevent new arrears the IMF "imposes conditions on the use of resources," assesses the members' "external viability and ability to repay," and provides technical assistance to "help formulate and implement reforms." (50) However, when its strategy goes wrong the IMF does not walk away. It stays, condemning the policies its former model pupil had to implement as inefficient and economically ill-advised, selling new advice, and helping with another program the country has to pay for.

IFIs argue that forgoing their "preferred creditor status" would deteriorate their own excellent rating as borrowers, which would increase the costs of lending--but this argument is flawed. The IBRD's practice of not acknowledging default of countries that have not paid anything over several years as long as such countries stay "in mutual respectful contact" (51) mocks all acceptable accounting rules. If publicly known (though officially denied) defaults have not reduced the IBRD's rating, properly acknowledged and handled default is unlikely to do so. Apparently, guarantees by OECD governments rather than their own lending records account for IFI ratings. Finally, a certain amount of lost loans is simply part and parcel of lending money. It is economically a necessary part: the possibility of losing money acts as an incentive for appropriate care in lending. All IFIs have loan loss reserves and should use them. If the rating argument were true no commercial bank would have excellent ratings as none gets everything repaid as stipulated. Banking centers such as London's City or New York would be viewed as assemblies of credit risks.

Legally and even according to their own statutes IFIs have no preferred creditor status. This technical concept does not formally apply to the IBRD, (52) nor to regional multilateral banks or the IMF. When preferring the IMF was discussed around 1988, the fact that the 1MF had no legal or contractual status as a preferred creditor could not be denied. (53) "Executive Directors stressed the ... need ... in practice ... to treat the Fund as a preferred creditor." (54) In September 1988, the Interim Committee endorsed this position and "urged all members, within the limits of their laws, to treat the Fund as a preferred creditor." (55) In contrast to the impression IFIs (and especially the IMF) try to create, there is no legal hindrance to equal treatment. Therefore, obviously, the IMF's sovereign insolvency proposal attempted to obtain de jure preferred status for IFIs in an extremely self-serving way. (56)

The lack of financial accountability and the existence of privilege have produced an unhealthy concentration of the Fund's general resources. The IMF stated: "The five largest users of credits as of January 31, 2004, were Brazil, Turkey, Argentina, Indonesia and the Russian Federation." (57) These five countries accounted for 86.4 percent of outstanding credits, a slight increase over the 85.1 percent they had owed oil April 30, 2003. Of all outstanding credits, 70.7 percent was owed by the largest three users of IMF resources--Brazil, Turkey, Argentina--slightly tip from 68.8 percent in April 2003. (58) Unlike all other IFIs, whose shares declined, the IMF increased its share of the public external debt (and debt service) of the top three borrowers significantly after 2000. Moreover, all three countries were "clients" for over five years-which raises questions regarding the effectiveness of IMF problem solving. All five were once model pupils of the IMF, paraded as examples of countries with particularly sound and successful economic policies. Similar exposure would not be allowed in the case of any commercial bank for good economic reason. But most if not all commercial banks would be too cautious to act in this way.

The IMF envisages further concentration as "inevitable," which "does not embody the same degree of risk" as for other financial institutions. (59) "Diversification is not an objective per se"--that is, the IMF will go on with past unhealthy practice. The Fund fails to give any convincing reasons. It simply refers to the responsibility of Fund members to pay back and mentions past repayments. Continuing such high concentration can only be explained by the absolute exemption enjoyed, the complete absence of economic risk, which in turn allows political influence that exposure to risk would simply preclude.

The high concentration of the IMF's claims has repeatedly been used as an argument, especially by Fund employees, for why financial accountability could not be introduced. Doing so would destabilize the Fund. Logically, this means that once a grave error has been committed, this fact itself would bar any correction. The reason for the high concentration of IMF resources is the lack of financial accountability, which has increased the weight of political pressure. Assured that it would not suffer losses but gain importance, the IMF was prepared to lend unreasonable amounts of money. Its current attempt to gain legally exempt status for its own and other multilateral claims within its proposed SDRM framework must also be seen with this background in mind. (60) Although the Fund has accumulated the money to cover losses, it attempts to avoid necessary debt reductions at great cost to debt-distressed clients.


The perverted incentive system that rewards errors and negligence of IFIs must be brought to an end. The mechanism of financial accountability that I have sketched in this essay has numerous benefits. Projects and programs financed under such a system would have a much better rate of success and much more positive impacts on development. Overall, loans would be cheaper to developing countries than present loans because of the many costly errors that would be avoided. It would also finally establish the right of victims of development cooperation to be compensated. Financial accountability would soften problems caused by the perceived undemocratic nature of IFIs. Potential economic costs would considerably constrain political influence still wielded by shareholders. Financial accountability would also be beneficial to IFIs themselves, as sound economic reasoning would more easily prevail over politically motivated interventions. It would provide staff a good argument against pouring money into regions just because of lending targets, political interference by important shareholders, or demands for bailouts.

(1) Ariel Buira, "Reforming the Governance of the Bretton Woods Institutions, "in OPEC Fund for International Development, ed., Financing for Development, Pamphlet Series no. 33 (Vienna: OPEC Fund, 2002), pp. 213-57. Over the years the democratic representation component of the Bretton Woods institutions--the basic votes given to each member equally--has been perceptibly diminished by capital increases. The basic votes, which were, e.g., 12.4% of the IMF's total voting power at Bretton Woods, are now 2.1%. In response, changes of voting rights and the representation of members by executive directors were demanded as both strongly disadvantage developing countries. The search for a new managing director after Kohler resignation brought the IMF's selection procedures under unprecedented critique.

(2) Ngaire Woods, in "The Challenges of Good Governance for the IMF and the World Bank Themselves," World Development 28, no. 5 (2000), pp. 823-41, has proposed special majorities or double majorities differentiating between the stakes countries hold in IFIs. More recently, in "Holding Intergovernmental Institutions to Account," Ethics & International Affairs 17, no. 1 (2003), pp. 69-80, she has advocated greater constitutional, internal, and political accountability of IFIs.

(3) On the performance of multilateral development banks, see Kunibert Rafter and H. W. Singer, The Foreign Aid Business, Economic Assistance and Development Co-operation (Brooldield, Vt.: Edward Elgar, 1996), pp. 175ff.; on the IDB, see Task Force on Portfolio Management, Managing for Effective Development: Report of the Task Force on Portfolio Management for the Inter-American Development Bank (Washington, D.C.: IDB, 1993).

(4) Kunibert Rafter, "Applying Chapter 9 Insolvency to Inter national Debts: An Economically Efficient Solution with a Human Face," World Development 18, no. 2 (1990), pp. 301ff.; K. Rafter, "International Financial Institutions and Accountability: The Need for Drastic Change" in S. M. Murshed and K. Rafter, eds., Trade, Transfers and Development, Problems and Prospects for the Twenty-First Century (Brookfield, Vt.: Edward Elgar, 1993), pp. 151ff.; available at /Kunibert.Raffer; and K. Rafter, "Good Governance, Accountability, and Official Development Co-operation: Analyzing OECD-Demands at the Example of the IBRD," in K. Ginther, E. Denters, and P.J.I.M. de Waart, eds., Sustainable Development and Good Governance (Dordrecht: Martinus Nijhoff/Kluwer, 1995), pp. 343fl: No other author has taken up this problem. Woods (2003) uses the term "financial accountability" but refers to the political accountability vis-a-vis members that finance IFIs. In my terminology that would be political accountability.

(5) Documenting failures of the Bretton Woods institutions has developed into a cottage industry. Among the more widely known books are Catherine Caufield, Masters of lllusion: The World Bank and the Poverty of Nations (London: Pan, 1998); Susan George and Fabrizio Sabelli, Faith and Credit: The World Bank's Secular Empire (Harmondsworth: Penguin, 1995); and Bruce Rich, Mortgaging the Earth: The World Bank, Environmental Impoverishment ant the Crisis of Development (Boston: Earthscan/Beacon, 1994). Arguably even more damaging are internal documents of which the Wapenhans Report is still the best known but by no means only example. See World Bank, "Effective Implementation: Key to Development Impact," Report of the Portfolio Management Task Force (Washington, D.C.: World Bank, 1992).

(6) Organisation for Economic Co-operation and Development, "DAC Orientations on Participatory Development and Good Governance" (Paris: OECD, 1993), para. 3, p. 2; available at INET/IMAGES.NSF/vLUImages/HRDG/$file/Dace. pdf.

(7) 0rganisation for Economic Co-operation and Development, Development Co-operation, Efforts and Policies of the Members of the Development Assistance Committee: 1997 Report (Paris: OECD, 1998), p. 22.

(8) Rafter, "International Financial institutions and Accountability," p. 158.

(9) "En Alemania, un fallo judicial apunta a un banco por los bonos, Un tribunal dispuso que un cliente sea indemnizado por haber sido mal asesorado,", October 29, 2003; available at diario/2003/10/29/p-00601.htm.

(10) John Gapper and Richard Gourlay, "Damages Win Leads Lloyds to Rethink Advice," Financial Times, September 5, 1995, P. 9.

(11) Joseph Stiglitz, "What I Learned at the World Economic Crisis: The Insider," New Republic April 17, 2000, pp. 56-59.

(12) See, e.g., OECD, Development Co-operation, Efforts and Policies of the Members of the Development Assistance Committee, pp. 18ff, or pp. 67ff.

(13) International Bank for Reconstruction and Development, "Toward Sustained Development in Sub Saharan Africa: A Program of Action" (Washington, D.C.: IBRD, 1984), p. 24.

(14) Jeffrey A. Winters, "Criminal Debt," Written statement, "Combating Corruption in the Multilateral Development Banks," Hearing before the Committee on Foreign Relations, U.S. Senate, 108th Congress, and session, May 13, 2004; available at WintersTestimony040513.pdf. At a conference at Northwestern University in 1999, IBRD staff objected vigorously, claiming that losses to corruption were lower, 20-25% at most.

(15) Paul Mosley, Jane Harrigan, and John Toye, Aid and Power: The World Bank and Policy Based Lending, vol. 1 (London: Routledge, 1991), p. 24.

(16) Robert Wade, "From Miracle to Meltdown: Vulnerabilities, Moral Hazard, Panic and Debt Deflation in the Asian Crisis" (paper presented at East Asia Crisis Workshop, Institute of Development Studies, Brighton, Sussex, June 1, 1998); available at

(17) Ibid.

(18) IBRD, 1998 Annual Review of Development Effectiveness, OED (Washington, D.C.: IBRD, 1999), p. 2; available at 11d38e62c269811285256808006a00a2/56124588862d25d88 52567/f200568e93/$FILE/arde98.pdf.

(19) Ibid.

(20) Ibid.

(21) Ibid.

(22) See the IMF's Articles of Agreement at external/pubs/ft/aa/.

(23) Kunibert Raffer and H. W. Singer, The Economic North-South Divide: Six Decades of Unequal Development (Northampton, Mass.: Edward Elgar, 2001), p. 157.

(24) Ibid.

(25) Stanley Fischer, "Capital Account Liberalization and the Role of the IMF" (paper presented at the IMF seminar "Asia and the IMF," Hong Kong, September 19, 1997); available at FISCHER.HTM.

(26) Caufield, Masters of Illusion, p. 319.

(27) IBRD, World Debt Tables 1992-93, vol. 1 (Washington, D.C.: IBRD, 1992), pp. 10ff (emphasis in original.

(28) Anne Krueger, "International Financial Architecture for 2002: A New Approach to Sovereign Debt Restructuring" (address given at the National Economists' Club Annual Members' Dinner, American Enterprise Institute, Washington, D.C., November 26, 2001; available at

(29) See Raffer, "International Financial Institutions and Accountability"; and Raffer, "Good Governance, Accountability and Official Development Co-operation."

(30) See Agreement Establishing the Inter-American Development Bank; available at leg/Documents/Pdf/Convenio-Eng.Pdf.

(31) See Agreement Establishing the Asian Development Bank; available at Charter/charter.pdf.

(32) An e-mail to the AfDB (April 7, 2004, 14:20:06) describing the problem and asking for the legal text was received according to the Bank's computer but has remained unanswered.

(33) International Development Association, "Background"; available at EXTERNAL/EXTABOUTUS/IDA/o,,contenMDK:200 51270~menuPK:83991~pagePK:83988~piPK:84004~the SitePK:73154,00.html.

(34) As recommended by Raffer and Singer, The Foreign Aid Business, pp. 209ff.

(35) International Monetary Fund, "Executive Board Reviews IMF's Income Position," Public Information Notice no. 03/64 (May 22, 2003); available at

(36) Ibid. (emphasis in original).

(37) James M. Boughton, "Digging a Hole, Filling It In: Payments Arrears to the Fund," in James M. Boughton, Silent Revolution: The International Monetary Fund 1979-1989 (Washington, D.C.: IMF, 2001), p. 814; available at history/2001/ch16.pdf.

(38) International Monetary Fund, "Executive Board Reviews IMF's Income Position," p. 3.

(39) Ibid.

(40) International Monetary Fund, "Financial Risk in the Fund and the Level of Precautionary Balances," February 3, 2004, p. 26; available at nal/up/tre/risk/2004/020304.pdf.

(41) I first proposed the idea of a court of arbitration to ensure financial accountability for projects in 1993. See Raffer, "International Financial Institutions and Accountability."

(42) I have proposed nearly the same procedure in my model of sovereign Chapter 9 insolvency. See Raffer. "Applying Chapter 9 Insolvency to International Debts."

(43) The IMF would have to waive its immunity, but is empowered to do so.

(44) For details, see Kunibert Raffer, "Introducing Financial Accountability at the IBRD: An Overdue and Necessary Reform" (paper presented at the conference "Reinventing the World Bank," Northwestern University, Evanston, Ill., May 14-16, 1999); available at home

(45) I recommend ad hoc panels as part of my model of sovereign debt arbitration because in that case there is hope that any standing institution will become severely underemployed once the backlog of cases of debt-distressed sovereigns are solved.

(46) Helping poor people from the South with advice and financial resources in the preparation of complaints would be a useful task for NGOs.

(47) International Monetary Fund and IDA, "Debt Sustainability in Low-Income Countries--Proposal for an Operational Framework and Policy Implications," February 3, 2004, p. 13; available at pdr/sustain/20004/020304.pdf.

(48) Ibid.

(49) Designing a fair and humane solution to the sovereign debt crisis in the late 1980s, I proposed this easy way to hold IFIs financially accountable. To the best of my knowledge it was the first time financial accountability of IFIs was brought up. This equal treatment provision has remained a fundamental feature of my sovereign insolvency model. See Raffer, "Applying Chapter 9 Insolvency to International Debts," pp. 307ff.; and Kunibert Raffer, "The IMF's SDRM: Another Form of Simply Disastrous Rescheduliilg Management?" in Chris Jochnick and Fraser Preston, eds., Sovereign Debt at the Crossroads (Oxford: Oxford University Press, forthcoming).

(50) International Monetary Fund, "Overdue Payments: IMF strategy focuses on prevention, cooperation, remedial measures, and deterrence," IMF Survey Supplement, September 2001, p. 21; available at

(51) Caufield, Masters of Illusion, p. 319.

(52) Ibid., p. 323.

(53) Boughton, "Digging a Hole, Filling It In," p. 820.

(54) Ibid.

(55) Ibid., p. 821.

(56) Raffer, "The IMF's SDRM: Another Form of Simply Disastrous Rescheduling Management?"

(57) International Monetary Fund, "Financial Statements, Quarter Ended January 31, 2004," p. 8; available at

(58) Compared with 1999 this is a steep increase: on April the three largest users "only" accounted for 49 percent, the "big five" for 69 percent. See International Monetary Fund, "Financial Statements, Quarter Ended April 30, 1999," p. 10; available at nal/pubs/ft/quart/1999fy/043099.pdf.

(59) International Monetary Fund, "Financial Risk in the Fund and the level of Precautionary Balances," February 3, 2004, pp. 24ff.

(60) Raffer, "The IMF's SDRM."
COPYRIGHT 2004 Carnegie Council on Ethics and International Affairs
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2004 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Raffer, Kunibert
Publication:Ethics & International Affairs
Date:Oct 1, 2004
Previous Article:Informing the integration debate with recent experience.
Next Article:Models of international economic justice.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |