IBRD borrowers pay for global public goods, G-7 agendas

by Chakravarthi Raghavan

Geneva, 2 June -- The World Bank is increasingly spending money and allocating funds to programs for global public goods and foreign policy interests of major share- holders, but the costs are borne by the borrowers from the bank's normal IBRD window, according to a critique by Davesh Kapur of the Bank's sources and allocation of net income and resources.

Among "global public goods" are agricultural and health research programmes; and those serving G-7 political and foreign policy interests or the personal agendas of Bank staff include the HIPC initiative (being pushed by some G-7 countries, international institutions like the UN and UNCTAD, and many national and international NGOs), the MIGA, the technical assistance programmes to assist the former Soviet Union, and jump-starting reconstruction in Bosnia-Herzegovina.

The costs of these programmes, says Kapur, are increasingly cutting into the Bank's net income and reserves. The Bank management's response has been to propose an increase in borrowing costs, through front-loaded fees, with adverse consequences for the Bank's governance.

Rather these costs should be met by more equitable burden-sharing among non-borrowing share-holders, IBRD borrowers, IDA borrowers, and staff and management.

The critique by Devesh Kapur, at the Harvard University Center for International Affairs, is in Volume X of the International Monetary and Financial Issues for the 1990s, published by UNCTAD, containing analytical studies for the G-24 developing country group at the IMF/World Bank.

For more than a year, says Kapur says, the Bank management has been flagging attention to the looming problems facing the Bank's net income, and proposing several steps to address the problems of falling income and rising demands. The Bank management has proposed raising net revenues ny augmenting loan charges and modifying use of net income.

But several factors other than the financial "subsidy" in current loan pricing explain the Bank's predicament.

These factors include: structure of loan pricing that gives little weight to incentives and adverse selection and moral hazard; a common pool problem since net income has been deployed in a manner concentrating benefits, especially political benefits to non-borrowing major share-holders, while the costs are spread among all share-holders, especially the IBRD (the International Bank for Reconstruction and Development, the hard loan window) borrowers, and excessive attention to the revenue side and little to the expenditure side.

The IBRD's equity (or reserves), the 'free' money available to the Bank, can increase either by injection of paid-in capital or by adding to retained earnings. The costs of additions to paid-in capital are borne principally by the large, non-borrowing shareholders. Retained earnings can come only from annual allocations from net income. A higher level of net income increases the bank's ability to absorb higher lending risks; higher reserves means better financial health. Both benefit all members, and high reserves benefit borrowers over the long-term by reducing the Bank's overall borrowing costs, and thus lending costs.

But funding higher reserves from net income through higher loan charges has been a contentious issue among shareholders - with fault lines not only along North-South divide, but among borrowing countries too. Major shareholders have pushed for higher reserves which reduce risk of their contingent liabilities and reduce paid-in capital increases in future replenishment.

Another rationale, during the 1980s, for higher reserves was need for adequate provision for non-accruals. But the burden was shouldered by one group of borrowers - those who serviced their debts on time. Another use for net income favoured by non- borrowing shareholders has been transfers to IDA, which reduced the need for their budgetary allocations.

Borrowers on the other hand are interested in reducing borrowing costs, and hence opposed increases in loan charges. Higher charges have no effect on loan demands from the less credit- worthy borrowers. And while altered loan charges have effects over time, increased front-end fees affect new borrowers.

There has been no unity among borrowers on these issues. IBRD- only borrowers press for net income to be used to lower future loan charges rather than supplementing IDA to which they have no recourse. China and India, blend countries getting IDA and regular loans, have been fence-sitters so far, but their attitude may soon change as they are eased out of IDA, says Kapur.

In 1991, with high projections of future net income, the Executive Directors approved a framework to guide annual net income allocation: first priority for adding to net reserves; second on reducing borrower costs by pre-funding upto 25 basis points for the next year for all borrowers servicing loans within 30 days of due date; third support to high-priority development activity; and four, accumulating funds temporarily in a "surplus account". The rationale for creating the surplus account was the new uncertainty over risk scenarios and a compromise between strongly divergent views in the Board.

The insistence of most major shareholders for high reserves was due to pressures on the Bank from the G-7 to lend more to eastern Europe and Russia and loosen the negative pledge clause for same end; and the perception of major shareholders and Bank management that increasing the paid-in capital would be extremely difficult in the near future.

The debates about net income also brought to the fore long- simmering differences on what should be included in or funded out of the administrative budget and what out of net income.

From 1964, the Bank had been making grants from net income, beginning with IDA. The auditors later argued that grants to organizations not affiliated to the Bank were "expenses" and should be dealt with as part of doing business and included in the administrative budget. So from 1982, some "special grants programmes" (SGP) were included in the administrative budget and the others came from net income.

Starting with the allocation for international agricultural research in 1971, the SGP has steadily increased in scope and size over the next 25 years, but remaining concentrated in two areas of agriculture (three-fifths) and health (one-fifth). In the 1990s, two others were added - the Institutional Development Fund (IDF) and the Consultative Group to Assist the Poorest (CGAP). Together they accounted for $110 million of the administrative budget in 1998 fiscal year.

Much of these expenditures were for global public goods "above the line". But by including them in administrative expenses, the Bank's costs of doing business have been inflated.

And borrowers have to pay for these "above the line" public goods. They also are picking up the tab for the "below the line", ie expenses out of "net income".

In the 1990s, the Bank began funding humanitarian efforts from net income and activities in non-members. Trust funds for technical assistance to the former Soviet Union, for investments in Gaza strip, and jump start reconstruction in Bosnia- Herzogovina, without waiting for financial normalization and membership.

The Heavily Indebted Poor Countries initiative is another making new and significant claims on net income. At least some of the earlier IBRD loans now to be written off under HIPC were "the result of projects undertaken due to political pressures of major powers and/or personal agendas of the Bank staff," says Kapur. "For some years other borrowers and bank staff had privately expressed scepticism on this lending, though none has ever gone on record against it."

"And while there is no doubt about the need for a new approach to the debt problem, the HIPC Initiative does not have any repercussions for non-borrowers," Kapur points out. "With the Bank's contribution coming out of the net income, the burden falls substantially on IBRD borrowers, who have only themselves to blame for weakly expressing their responsibility as share- holders." While allocations from net income have been ostensibly for the benefit of the Bank's membership, there is little doubt that the particular foreign policy interests of some major shareholders, rather than intrinsic merits or benefits to the membership as a whole, have played an increasingly important role.

Despite the managements dire predictions about trends in net income, the issue needs to be seen in perspective, says Kapur. While the management argues that the decline in net income is an aberration to be rectified by increasing borrowing charges, it could be turned on its head by pointing to the increase in net income between 1987-1997 as due to an aberration. The Bank's net income was artificially buoyed through the 1980s by purportedly high returns on its investments - due largely however to transferring the currency risk to borrowers. Another source of buoyancy was high-interest fixed-rate loans in the 1990s, when interest rates were declining.

Except for the 1988 General Capital increase which moderately boosted the Bank's equity, "the borrowers disproportionately bore the costs of the Bank's financial policies, while the benefits in net income and reserves were shared by all."

Another reason for the Bank's predicament is political. Unlike the IMF, which moved to the Special Drawing Rights more than two decades ago, the Bank persists in maintaining accounts in US dollars, leading to net dollar income to decrease as the dollar appreciates or vice versa. The dollar's appreciation over the last two years has thus affected its income, and to that extent is temporary.

The Bank management's inability to manage growing demand on net income has also been shown by the fact that even as it was lamenting the trends in net income, it sanguinely transferred from net income to partially pay for capital increase of the Multilateral Investment Guarantee Agency (MIGA).

And since the onset of the Asian crisis and the sharp increase in Bank's lending, the projections have turned bleaker. Even if the dollar income improves marginally, the reserves to loan ratio is set to decline to 11.8 percent by 2002.

The Bank management's proposals to increase net income, complains Kapur, have focused exclusively on the revenue side, rejecting categorically cutting expenditures. The management claims there is an increase in the implicit subsidies to borrowers as the spread on loans covers a decreasing share of administrative expenses, and assumes the administrative costs as a given.

But unlike commercial banks, the World Bank is a price setter in lending, not a price taker, thus sharply reducing incentives to cut costs. For a long time the Bank has had a "soft budget".

This is shown by the fact that the administrative costs per project has doubled in real dollars between 1970s and 1990s -- from $2.3 million in 1975-77 to $3.7 million in 1985-87 and $4.4 million in 1995-97. Over the same period, the share of satisfactory projects as a percent of all projects declined from 85 to 68 (in 1985-87), and marginally increasing to 69 in 1995- 97.

Despite substantial rhetoric, the Bank's administrative costs (including that on IDA) increased during much of 1990s, while lending stagnated. Between 1986 and 1996, the administrative costs increased by 95% while lending rose only by 28 percent. The rise in costs was most apparent in areas far removed for lending - by 142 percent or from $56.5 million to $136.5 million on corporate management expenses which are weakly linked to direct lending, but closely linked to management and major share-holder prerogatives.

The Bank's standard answers that the costs increase because projects have become more complex, Kapur says, are most unconvincing - if one compared the Bank's administrative costs with that of the European Investment Bank (EIB). As against outstanding loans of $110 billion, the Bank's administrative expenses rose in Fiscal 1996 to $733 million or $6.7 million per $ billion in loans outstanding. The EIB's in comparison was $181 million administrative expenditure against $132 billion outstanding loans or $1.4 million per $ billion outstanding loans.

The Bank's administrative expenses have also increased due to the non-borrowing shareholders insistence on introducing mounting safeguards, increases in reviews, consultations and conditions. First introduced through the backdoor of IDA replenishment, they have now become Bank-wide policies. And over time, a vested bureaucracy has grown in the Bank, insisting on the virtues of ever more safeguards, with little attention paid to costs which are passed through in loan pricing.

As the Bank's own advice to borrowers has stressed, ever mounting safeguards and regulations create their own political economy dynamics, skewing bureaucratic incentives within the Bank.

The Bank's internal administrative practices have also been cost- driven. Personnel policies have made it extremely difficult to fire staff, and on the rare occasions when the Bank has done so, the handshake was golden. "It is regrettable that the Bank's shareholders have not chosen to seek any equivalence between personnel policies of the institution, and what the institution regularly asks its borrowers to do in its lending advice and conditions."

Sheer administrative incompetence, manifest in the myriad of reorganizations, plays a not unimportant role in ratcheting up the costs. Shareholders reluctance to shield Bank from outside carping, built-in structural features of the Board - ranging from frequency of rotation of Executive Directors to widely varying agendas - have made oversight difficult. The problems of asymmetric information as between the agent (management) and the principals (Board) has made the problem particularly acute in the case of the Bank because of the differing interests of the principals.

Perpetual reorganizations in the Bank have engendered distrust and cynicism and lowered morale. And because of the substantially higher transaction costs involved in Bank loans, borrowers with access to markets, have themselves "rationed" their intake of Bank loans.

Borrowing countries have usually been opposed or at best reluctant in their support for tighter budgets at the Multilateral Development Banks (MDBs), and there is a collective action problem. Private reservations aside, no developing country is individually willing to publicly cross swords with management of an MDB on the budget, fearing that its programmes will be singled out to bear the burden of cuts.

Another way of meeting the problem of net income and need for reserves would be to raise equity through additions to paid-in capital. At the 1988 General Capital increase, the major shareholders, pointing to difficulties at home, insisted on reducing the paid-in share to three percent. Yet two years later, the OECD countries had no problem in a 30% contribution to a new MDB (the European Bank for Reconstruction and Development, EBRD) - and agreeing to a larger budget outlay of $3.45 billion for the EBRD, compared to the $2.25 billion for the IBRD, even though EBRD's functions could have been replicated by the IBRD at a smaller cash outlay.

Kapur argues that the IBRD still has considerable headroom, but additional paid-in capital would boost equity and result in more equitable burden-sharing. A small increase of 15 to 25 billion dollars, but with a 20% paid-in (the maximum under IBRD statutes) would "not only ensure that the burden of providing global public goods is shared in a more equitable manner, but strengthen the fraying links between power and financial burden."

Unlike in the UN system, the BWIs had unequal distribution of power, but this was initially linked expressly to unequal financial burden - both direct in the form of paid-in capital and indirect, in liabilities for callable capital. "Both these burdens have waned, but the distribution of unequal power has remained constant."

For instance, in real dollar terms, the capital contribution of the United States (the Bank's largest single shareholder) to the IBRD was greater before 1949 than now. The contingent liabilities are minuscule - both the IBRD's historical track record and rising reserves and substantial loan-loss provisions make a call on capital even more improbable.

While the Bank's financial strength grew, the cost of "ownership" has fallen: easier borrowing and comfortable equity has reduced the need for additional paid-in capital; higher reserves and track record on defaults have diminished the risks to callable part of subscribed capital. One consequence has been that "the influence that came with ownership has become less expensive, and almost costless - and therefore more attractive."

"Hence the greater intensity of disputes on even slight changes in capital shares and use of net income by major shareholders for expressly partisan purposes. The transfers of net income from IBRD to IDA allow major shareholders to retain major voting shares over IDA while reducing their financial outlays.

"And using net income to augment MIGA's capital increase serves a similar, but even more questionable purpose: the IBRD's borrowers are paying for non-borrowers to retain their voting power in MIGA.

"And it stretches credulity to argue that the much larger transfers from the Bank's net income for Bosnia-Herzogovina and Gaza compared to those for countries like Liberia, Somalia or Rwanda, are driven by humanitarian or developmental rather than political concerns." (SUNS4447)

The above article first appeared in the South-North Development Monitor (SUNS) of which Chakravarthi Raghavan is the Chief Editor.

[c] 1999, SUNS - All rights reserved. May not be reproduced, reprinted or posted to any system or service without specific permission from SUNS. This limitation includes incorporation into a database, distribution via Usenet News, bulletin board systems, mailing lists, print media or broadcast. For information about reproduction or multi-user subscriptions please contact < >