The Russell Sage Foundation
"IMF2" 2 March 98 Comments welcome:, or 212-750-6084, Veneroso 908-522-8815
Citation: New Left Review, March-April 1998, pp.3-23.


Robert Wade and Frank Veneroso

How could the widely acknowledged real estate problems of Thailand’s banks in 1996 and 1997 have triggered such a far-reaching debt-and-development crisis? The devaluation of the Thai baht in July 1997 was followed by currency crises or financial instability in Indonesia, Malaysia, Philippines, Taiwan, Hong Kong, Korea, Estonia, Russia, Brazil, Australia and New Zealand. There were few signs of impending crisis, such as rising interest rates in the G-7 countries or a sudden suspension of capital flows to developing countries after the baht devaluation. On the contrary, bank lending to Asia rose to a record level in the third quarter of 1997. The Japanese government’s de facto credit rating agency, the Japan Center for International Finance, gave Korea one of its highest credit ratings for any developing country in June 1997. The IMF and the World Bank lavished praise upon the governments of the region through 1997, including on the Korean authorities as recently as September 1997.

What began as a debt crisis has become a fully-fledged development crisis. Throughout this most successful of developing regions living standards are falling as unemployment rises and the effects of huge devaluations work through into higher import prices. Many millions of poor people are at risk, and many millions of people who were confident of middle class status feel robbed of their lifetime savings and security. It is not a humanitarian tragedy on the scale of North Korea, but the loss of security and productivity is a tragedy nonetheless, almost as cruel as war. South Korea, Thailand and Indonesia are sliding into depressions comparable to those that gripped Europe and North America in the 1930s.

Financial crises—speculative bubbles followed by collapse—have recurred throughout the history of capitalism. In the early 1980s Latin America, another fast-growing developing region, suddenly went into debt-and-developoment crisis and stopped its rise up the world economic hierarchy. The Latin American crisis was due, said the consensus of analysts, to the combination of bad macroeconomic policies plus foreign borrowing by governments, that was wasteful and corrupt because done by governments rather than by private firms operating in competitive markets. But the Asian crisis has occured in the opposite conditions. In East and Southeast Asia today most debt is private debt. And prior to the crisis, the macroeconomic "fundamentals" looked fine. The countries in question have had low inflation, budget surpluses or only small deficits, and until recently stable or rising foreign exchange reserves. They have been growing fast. East and Southeast Asia accounted for a quarter of world output, but half of world growth over the 1990s and almost two thirds of world capital spending. Firms throughout the region make products that sell in the most demanding markets—if the exchange rate is right.

There is little agreement on why the magnitude of the crisis has been so large, what can be done to get out of it, who will gain and who will lose, and what changes need to be made in international regimes to reduce the likelihood of repeats. These matters should be the subject of an international debate as important as the Bretton Woods conference at the end of the Second World War.

According to the IMF, the solution has two thrusts: on the one hand, domestic austerity in order to restore the capacity to repay foreign debts; on the other, radical institutional reforms, including further financial liberalization. Many analysts have come forward to disagree. Indeed, a new "conventional wisdom" among the IMF ’s critics has emerged, which goes like this.

The crisis is a crisis of liquidity more than solvency. Creditors have "run" on the currency and on domestic assets, leaving the borrowers unable to continue to finance their loans. It happened partly because of excessive financial deregulation, including, above all, allowing banks and firms to borrow abroad without any government controls or coordination. And international bankers, also largely unregulated, have been only too happy to provide Asian corporations with large amounts of short-term, unhedged dollar loans. Once some of them tried to call in their loans the crisis began to build upon itself as other lenders tried to call in loans and firms cut operating costs and sold assets, causing unemployment to rise and asset values to crash.

"Instead of dousing the fire the IMF in effect screamed fire in the theater", says Jeffrey Sachs. The Fund’s insistence on shuttering many banks despite the fact that systems of deposit insurance hardly exist led panicky depositors to withdraw their deposits even from sound banks, and hold cash instead. Its insistence on cutting demand and liquidity accelerated the bankruptcy or radical devaluation of the value of firms that were efficient and profitable, as well as those that were not. Its push for institutional reforms in finance, corporate governance and labor markets convinced lenders that the economies were fundamentally unsound.

The immediate goal must be to restore confidence, which requires overcoming the collective action problem in which no lender wants to refinance for fear that others will not. Demand and liquidity must be increased, not reduced, in order to keep firms turning over. The IMF should be concentrating its attention on organizing a determined refinancing effort and then in helping to erect the structure of financial regulation, especially at the border, that will help to minimize the risks of such a melt-down occurring again.

We agree with this line of argument as far as it goes. We go further, however, arguing that the long-term damage to Asian economies of the IMF’s prescriptions is likely to be even greater than the critics have recognized. The reason has to do with a neglected dimension of the crisis—the financial strucuture of East and Southeast Asian economies, that differs from the kind of case the IMF usually deals with. Because of this difference, a unit of IMF "austerity" and "financial liberalization" will have higher costs and smaller benefits in Asia than elsewhere. The slowness of the IMF’s bail out packages for Thailand, Indonesia and Korea to revive confidence, despite being the biggest in the organization’s history, reflects not only their imposition of impossibly ambitious institutional reforms but also their inappropriateness for Asian financial structures.


In a western, including Latin American, financial system companies normally carry an amount of debt that is no bigger than and generally less than the value of their equity capital; and banks will not, or should not according to standard prudential limits, lend to companies with higher levels of debt. In East and Southeast Asia, and especially in Japan and Korea, corporate debt/equity ratios of the bigger firms are commonly two to one or more.

Why are corporate debt/equity ratios so much higher than in western systems? First, because savings are much higher. Gross domestic savings to GDP ratios in Asia are one third of GDP or more, compared to 15-20 percent in western systems. The savings are done in large part by households. Households hold their savings mostly in bank deposits, bank deposits being much less risky than equities in the noisy stockmarket environments of developing countries. Banks have to lend. When neither households nor government are significant net borrowers, the system is biassed towards borrowing by firms. (Lending or investing abroad is only a very partial alternative, given the amount of savings to be absorbed.)

Second, firms that aim to make an assault on major world industries (as especially in Japan, Korea, Taiwan) must get their hands on very large amounts of resources, which they can do only by borrowing. Neither equity markets nor corporate retained earnings are feasible alternatives for mobilizing resources on the scale required to compete in these export markets and continually upgrade.

High ratios of bank deposits and loan intermediation to GDP, and of corporate debt to equity, make the financial structure vulnerable to shocks that depress cash flows or the supply of bank or portfolio capital. The deeper the intermediation of debt (that is, the higher the ratio of bank deposits to GDP and the higher the ratio of corporate debt to equity), the more likely that any depressive shock will cause illiquidity, default, and bankruptcy. Debt-intensive strategies should be labelled, "This product can be harmful to your wealth".

Such a financial structure requires cooperation between banks and firms, and considerable government support. The trick is to buffer firms’ cash flow and supply of capital against "systemic" shocks, while not protecting firms from the consequences of bad judgement or malfeasance. Restrictions on the freedom of firms and banks to borrow abroad, and coordination of foreign borrowing by government, are a necessary part of this system.

Western commentators often dismiss the system as "crony capitalism", seeing only its corruption and favoritism. They miss the financial rationale for cooperative, long-term, reciprocal relations between firms, banks and government in a system which intermediates high savings into high corporate debt/equity ratios. (They also miss crony capitalism US-style, generated by the electoral finance regime.)

The need for state support allows the state to influence the decisions of banks and firms in line with a national industrial strategy, by withholding support from banks and firms that operate against the strategy. The whole system can be disciplined by making investment incentives conditional on export performance or on reductions in the gap between the firm’s prices and international prices for the same products.

High household savings, plus high corporate debt/equity ratios, plus bank-firm-state collaboration, plus national industrial strategy, plus investment incentives conditional on international competitiveness, equals the "developmental state". For all the white elephants and corruption ( inevitable when a third of national income is being intermediated), the system that allows firms to borrow multiples of their equity has yielded a quantum leap up the world hierarchy in technology and scale, and rates of improvement in living conditions that surpass virtually all other countries’.

Notice the parallels with Keynesian theory, which identifies savings rates in excess of investment rates as a cause of depressions and insecurity and even higher savings rates. Keynesian theory, however, posits government deficit spending as the solution. We see in East Asia a model of private debt based on high corporate debt/equity ratios, which give rise to the need for government protection against periodic slippages that would otherwise lead to widespread bankruptcy.


The other side of the equation, however, is very high levels of corporate debt. It is likely that Korea’s corporate debt/GDP ratio is of the order of 30 to 50 percent higher than the US’s corresponding ratio. This represents a debt mountain that sits at the heart of the Korean problem. The mountain may be less high in other Asian countries, but it remains much higher than is normal in western systems.

To see the dangers of debt, compare systems with low and high corporate debt/equity ratios. Low corporate debt/equity systems, as are found in Latin America and North

America, are not able to invest as heavily as the others but are also less vulnerable to shocks. They can sustain a sharp rise in real interest rates for some time. Corporate gross profits before interest and taxes are more likely to be high enough relative to the higher interest charges that some degree of debt repayment out of cash flow remains possible. Therefore the tendency for real debt to grow as a result of higher real interest charges is less than when debt/equity ratios are high. If the interest rate rises to the point where the firm cannot repay any of the extra cost out of cash flow or reserves and therefore must recapitalize it (that is, add it to its stock of debt), the balance sheet still has room for a higher debt/equity ratio without threatening the firm’s viability by wiping out its equity.

The risk that an interest rate above the rate of gross profit has disastrous consequences increases with the debt/equity ratio. In higher debt/equity systems firms have to use more of their gross profits on interest charges. A significant rise in interest costs may not be able to be met out of profits, in which case it has to be recapitalized into debt. But the balance sheet may not have room for more debt without threatening the firm’s viability. A rise from 10 to 20 percent lifts a debt/equity ratio of 80:20 (or 4:1) to 88:12 in the first year (if the interest due is all recapitalized and if the corporation is just breaking even at the start). Replicated across many firms, the country’s overall debt to GDP ratio rises. If, in addition, the high real interest rate policy also depresses aggregate demand, it will further depress cash flow relative to interest charges, accelerating the indebtedness of the corporate sector. And all the worse if, as is true in Asia, a substantial share of the debt is foreign debt and the domestic currency is devalued, raising the fixed interest payments on the foreign debt in domestic currency.

A higher debt/equity ratio not only makes for higher borrower’s risk; the lender’s risk equally rises with the ratio- unless the IMF bails them out.

The box illustrates the contrast between high and low debt/equity systems with a simple example. It shows how a "Latin American" firm with low debt to equity is able to survive a devaluation, interest rate hike and austerity program much more easily than an "Asian" high debt to equity firm.


What made for the high-growth performance of Asian systems in the past has led to or at least amplified the present crisis. Over the 1990s western and Japanese banks and investment houses lent heavily to Asian companies. They assumed, contrary to all historical knowledge about growth rates, that fast growth (four times the OECD average) would continue, and consequently that exchange rates would remain stable. They also each ignored their own prudential limits on lending to companies with high debt/equity ratios, because everyone else was ignoring the limits and they each wanted to win business. International bankers have a powerful incentive to follow the herd, because the banker who does not make money where others are making it risks being seen as incompetent but does not suffer by making losses when everyone else is making losses too.

Meanwhile, Asian governments undertook radical financial deregulation, encouraged by the IMF, the OECD, and by western governments, banks and firms. They removed or loosened controls on companies’ foreign borrowings, abandonned coordination of borrowings and investments, and failed to strengthen bank supervision. By doing so they violated one of the stability conditions of the Asian high debt model, helping to set up the crisis.

The rush to capital liberalization in the early to mid 1990s without serious institutional support stands out as the single most irresponsible act in the whole crisis, for which the blame falls equally on national governments and international organizations. The Bank of Thailand, for example, undertook capital liberalization just as it and other relevant government regulatory agencies were overwhelmed with other complex issues and political strife. The subsequent capital inflows were entirely unmonitored, let alone controlled. In Korea key people were bribed by Japanese and western financial institutions, thanks to which they did something that was counter to the whole thrust of Korean development policy for decades past. Bribery aside, the government placed great emphasis on joining the OECD, and the OECD made financial openness a condition of membership. As part of the same set of reforms, the government abolished the Economic Planning Board, the main body for making economic strategy since the early 1960s, and made the Finance Ministry dominant.

Domestic corporate borrowers discovered that they could borrow abroad half as cheaply as they could borrow at home. Foreign debt escalated, most of it private and short-term (maturing in 12 months or less). In Korea, foreign debt incurred by its banks and the companies that borrowed from them exploded from very little in the early 1990s to roughly $160 billion by late 1997.

China’s devaluations of 1990 and 1994, together with its lower inflation and faster productivity growth, made the yuan the most undervalued major currency in the world, worsening the export competitiveness of other East and Southeast Asian economies. The appreciation of the US dollar against the yen after 1995 (the result of an agreement between the US Treasury and the Japanese Finance Ministry to help Japan export its way out of trouble and use the resulting surpluses to buy US Treasury bills, thereby allowing US interest rates to be kept at politically desirable levels and assisting the re-election of President Clinton) worsened their competiveness still further, because their own currencies were pegged to the dollar and rose with it. Thailand in 1996 experienced zero growth of exports, the slowest rate of growth of GDP in a decade, and a ballooning current account deficit. The Thai stock market lost a fifth of its value in the first nine months of 1996, and growth of direct foreign investment fell sharply. In Korea, manufacturing production started to fall in 1996, the current account went strongly negative in the same year, industrial bankruptcies occured.

Once foreign lenders began to worry about currency falls, they "discovered" their heavy exposure to companies with debt/equity ratios far above their prudential limits. More exactly, they discovered the possibility that others might make a similar "discovery", the aggregation of which would precipitate falls in the exchange rate—multipling the loan burden and the risks of default. Hence they have tried in every way to call in their loans and not make new ones. The Japanese banks that lent heavily to firms in East and Southeast Asia have been especially anxious to call in these loans as their domestic position deteriorated.

This is why the run has been so surprisingly big. International banks have slashed credit lines to all borrowers, including the export-oriented firms that should be benefiting from currency depreciation. Even the big Korean chaebol, with world-wide brand names, are finding it difficult to get even trade credit (letters of credit to cover the import of inputs into export production). In Latin America of the 1980s, where companies had much lower debt/equity ratios, foreign lenders found that companies continued to meet their test of prudence, even if countries did not. Latin American companies therefore did not suffer such a big withdrawal of bank credit, and the problem of refinancing the Latin American debt was solved more quickly than looks like happening in the Asian case—though the Latin American countries were far less creditworthy. As of this writing, the IMF, the United Nations and other international forecasters continue to chase the whole world economy downhill, at least for 1998.


The IMF is designed to provide bridging finance while a country gets its balance of payments in order. IMF programs normally seek to reduce current account deficits, keep inflation in check, and keep domestic demand constrained. Such objectives are set out in the opening lines of the IMF stand-by agreement with South Korea, dated December 5, 1997:

1. Objectives The program is intended to narrow the external current account deficit to below 1 percent of GDP in 1998 and 1999, contain inflation at or below 5 percent, and—hoping for an early return to confidence—limit the deceleration in real GDP growth to about 3 percent in 1998, followed by a recovery toward potential in 1999." From then on the IMF program for Korea goes well beyond standard IMF programs, calling for structural and institutional reforms even though they are not needed to resolve the current crisis. It requires major financial restructuring to make the financial system operate like a western one, though without actually saying so. This includes closing down or recapitalizing troubled financial institutions; letting foreign financial institutions freely buy up domestic ones; requiring banks to follow western ("Basle") prudential standards; requiring "international" (read "western") accounting standards to be followed and international accounting firms to be used for the auditing of financial institutions. It requires the government to undertake not to intervene in the lending decisions of commercial banks, and to eliminate all government-directed lending; and to give up measures to assist individual corporations avoid bankruptcy, including subsidized credit and tax privileges.

The Fund also requires wider opening of Korea’s capital account, to enable even freer inflow and outflow of capital, both portfolio capital and direct investment. All restrictions on foreign borrowings by corporations are to be eliminated. The trade regime, too, will be further liberalized, to remove trade-related subsidies and restrictive import licensing. Labor market institutions and legislation will be reformed "to facilitate redeployment of labor".

The IMF programs for the other Asian cases differ from case to case. But they too include the traditional IMF measures to improve the balance of payments, including high real interest rates; and they also go beyond the IMF’s mandate into areas of institutional reform, including capital account opening and financial sector and labor market deregulation.

In terms of the first set of objectives (to remove the current account deficits and achieve macroeconomic balance), we are already seeing, at current heavily depreciated exchange rates, big trade surpluses from several Asian countries. Korea ran a giant current account surplus of $3.7 billion dollars in December 1997, equivalent to something like 15 percent of Korean GDP when annualized at the post-devaluation exchange rate of 1,600 won to the US dollar. Thailand ran a current account surplus for the last several months of 1997. So did Malaysia. So far the surplus is due more to falls in imports than to rises in exports. But the majority of imports are capital goods and industrial materials and fuel rather than consumer goods, and their cutback hurts exports. Only truly heroic improvements in the trade balance could garner enough foreign exchange to cover interest payments falling due in the next several years.

The difficulties of doing so are compounded by the costs associated with the Fund’s second set of objectives, those to do with radical institutional reform, and in partcular with liberalizing the financial sector both domestically and externally. Radical financial liberalization will face very large "transitional" costs; and in any case, even if a "western" look-alike system is established it would not be stable given the high flow of savings. It would also sacrifice the developmental advantages of a high debt system.

The transitional difficulties relate to the implications of the existing debt. Before western prudential limits can be viable, before the financial system can be made to work like a western one, the debt mountain must be brought down. The IMF seems not to have thought through the consequences of doing this.

Historically, debt mountains have been reduced in one of four ways. One is inflation: the debt is vaporized by means of a domestic inflation that causes real interest rates to turn negative. The second is bankruptcy: existing creditors lose some of the value of their assets as the debt is written down, the new creditors reorganize the assets and (hopefully) make the company viable again at the lower level of debt. The third is repayment of the debt out of cash flow. The fourth is by debt-to-equity swaps.

All of these have social costs, but some more than others. A country that goes the bankruptcy route will suffer major social disruption and loss of output while "resources" (including people) are reallocated. The principal lenders are banks, which are always highly leveraged (have high debt/equity). When banks write down the debt of the companies to whom they have lent they lower their own asset base, and jeopardize their own ability to meet their principal and interest payment obligations on deposits. They may have to stop refinancing sound companies that then become insolvent, in turn transmitting insolvency pressures to their customers and suppliers. Asset prices may begin to collapse as foreclosing creditors sell at firesale prices, causing further problems for holders of existing assets who see their value knocked down. Fears of bank deposit failures increase the demand for currency relative to bank deposits. The banking system may undergo a multiple contraction of deposits and loans. Layoffs proliferate. Consumers cut back. The disinflationary impetus is reinforced.

The bankruptcy route has been an integral part of all great depressions. Irving Fisher observed that the central propagating mechanism of the Great Depression of the 1930s was the rising real value of dollar liabilities (a rising real interest rate). As the price level fell, the real value of the principal of the debt rose. Firms found themselves facing higher and higher levels of indebtedness and repayment obligations, and banks called their loans and refused to lend. The resulting bankruptcies deepened and propagated the deflationary dynamic just described.

The repayment-out-of-cash flow route is likely to be protracted (if the existing levels of debt and interest rates are not too high for it to be workable from the start). As firms use most of their return on assets to repay principal and interest, their investment falls. This route is associated with prolonged stagnation. It takes many years to pay down the debt to the point where western prudential standards can be met. Japan has tried to follow this route during the 1990s. We can see its costs in the very slow growth of the Japanese economy since the stock market and property market bubbles burst in 1990.

The inflation route also has social costs, but historically the costs have tended to be lower than those associated with the others. Provided the inflation is kept at 40 percent or less the social costs are small. This approach requires the price level to be rising fast enough to make real interest rates low or negative, and it requires a semi-closed capital account in order to check capital flight.

The IMF prescriptions preclude the inflation route. They call for high real interest rates (in order not only to curb demand but also to encourage a reversal of the capital outflow). And they emphatically call for the capital account to be opened wide.

Debt-to-equity swaps were used to help reduce the Latin American debt crisis of the 1980s. In the Asian context, where debt to equity ratios are much higher, it is not clear that they could be used on a scale sufficient to make a big difference. In any case, given the lack of equity resources now held by post-crisis Asian nationals, a significant reduction of debt by this method requires massive foreign ownership positions in Asian firms and banks. Vast swathes of the corporate sector would end up in foreign hands. Already we are seeing Japanese and American companies jumping from minority to majority owners of Southeast Asian firms in return for a writing down of the debt. And debt-to-equity swaps aside, the devaluations enable foreign companies to buy up Asian companies at fire-sale prices (or in the current Korean phrase, "IMF cold wave prices"). We are already seeing a political backlash against the sudden jump in foreign ownership. Korean and Southeast Asian editorialists have started to write about "The Second Opium War" and US/IMF imperialism.

If the debt claims were swaped for equity claims held by domestic banks, the banks would end up as huge equity holders, something that banks are not supposed to be. If the debt claims were transferred to the state and the state in return acquired a voting interest in the company, this would amount to nationalization¾ not something the IMF or the US Treasury wish to encourage. Morgan Bank has suggested government guarantees for private foreign borrowing, in the hope that this would allow foreign lenders to resume lending to highly indebted firms on the grounds that the guarantee makes the debt look like sovereign debt. But the debt mountain would remain.

If the IMF prescriptions for reshaping Asian financial systems into something more like the western model require, as a condition of viability, a run down of corporate debt, and if the inflation route is ruled out, then the social costs are likely to be huge and long-lasting—especially because of the sheer amount of corporate debt relative to GDP that has to be shrunk to western levels. Inflation is the only way to reduce such a debt mountain without years of stagnation, nationalist backlash, or chaos.

The much higher real interest rates required by the Fund will tip many high debt/equity firms into bankruptcy—and the resulting financial instability and unrest may cause net capital outflow instead of the inflow that the Fund expects. Meeting western standards for the adequacy of banks’ capital entails a rapid fall in banks’ debt/equity ratios and a sharp cut in their lending, causing more company bankruptcies. Opening up the financial sector to foreign banks will result in a large-scale takeover, because after the bankruptcies and liquidations foreign banks and companies will be the only ones with the capital for recapitalizing the domestic ones. But foreign banks may not lend to high debt/equity local companies, and may not participate in the kind of alliances between government, the banks, and companies that a high debt/equity financial structure requires. If Citibank buys up Korean banks and applies its normal prudential limits (by which lending to a company with a debt/equity ratio of 1:1 is getting risky), it will not lend to Daewoo with a debt/equity ratio of 5:1. The amount of restructuring of Daewoo before its debt/equity ratio comes close to 1:1 is hard to imagine.

It seems particularly unwise for the IMF to insist that companies receive even more freedom than before to borrow on international capital markets on their own account, without government coordination, when it was their uncoordinated borrowing that set up the crisis in the first place. This will make the country more, not less, vulnerable to capital flight.

The IMF approach is likely to generate big social costs long before there is any significant amount of debt reduction, all because of a short-term and unforeseeable run by mobile capital. It aims to dismantle the high debt system, its developmental advantages notwithstanding. And it wants to see a western-type financial system in its place that can only work with a huge reduction in levels of corporate debt. The Fund has not properly weighed the economic and social costs of doing this, nor even the question of whether it has any legitimate business in entering the field of structural and institutional reform. Eventually Asian economies will start to grow again, for their "fundamentals" are strong; but by then their fundamentals will not be as strong. There will be an inner source of instability created by the attempt to integrate the massive flow of household deposit savings with a financial structure based on western norms of prudent debt/equity ratios. And by then they will have a rather different pattern of ownership, with foreign firms and banks—in particular, US firms and banks—having much more control than before and receiving much more of the profits. They will have given up the developmental advantages of a high debt system based on government-bank-firm collaboration in return for somewhat lower risks of financial crashes.

Once the crisis is passed, some reneging on IMF agreements may occur. But by that time foreign banks and other financial institutions may be well established, making the high debt/equity system difficult to rebuild.


The high savings of Asian households impart a bias towards high levels of corporate debt. Household saving rates may come down over the next several decades, as overall growth comes to be based more on internal demand than on exports. But saving rates much higher than in the west are likely to be a feature of these economies for many years to come, and the current crisis will only intensify households’ wish to save.

Households’ risk aversion precludes transfer of more than a small part of the savings through equity markets. Equity markets will of course develop over time as the infrastructure for a well-working stock market is gradually built up. But even in the most highly developed equity markets only a few percentage points of GDP or less are transferred. In the United States, where the equity market is a celebrated national institution, net savings transferred to the corporate sector through the equity channel have averaged less than 1 percent of GDP, and have often been negative over the past decade. Such small flows may be meaningful in a country like the US where household savings are only 4 percent of GDP, but are trivial where household savings are more like 20 percent of GDP. Moreover the current crisis has caused huge losses for most of the Asian households that have recently begun to participate in national stock markets. This makes it all the more likely that sizable development of equity markets is a dead issue in Asia for another decade at least. The Japanese experience is cautionary. Since the crash of 1990, over 90 percent of the mutual fund holdings accumulated over the 1980s have been redeemed.

Nor is a regime of internal corporate finance, through retained earnings, likely to work in a system where firms aim to make inroads in major world industries. Such a regime of "internal financing", where the ratio of equity to debt is high because of high rates of retained profits, is found in some developing countries. But internal financing tends to be associated with very unequal distributions of income and wealth, since wealth accumulation is concentrated among corporate owners who earn very high profit rates—often because of high rates of protection and subsidy, or because of monopoly positions. And the regime, though financially stable, is "blind", in the sense that accumulation tends to be confined to established firms, and may be remote from the discipline of either market competition or a national industrial strategy.

If Asia continues to save at anything close to current levels there is an inescapable problem of how to invest the savings productively. It is a fine irony, since Asia’s high savings have been instrumental in its fast growth and the envy of the rest of the world.

We argue that high savings and high corporate debt yield powerful advantages in terms of national development. The high levels of debt can be sustained under normal conditions provided that banks and firms have mutual understandings about the refinancing of the debt, and provided the government supports them. This in turn requires, above all, closing or semi-closing the capital account, so that mobile capital cannot go freely in and out. In such high saving societies foreign savings are not needed; it is already difficult enough to allocate domestic savings to efficient investments at the margin. These arrangements may stop well short of a developmental state a` la Japan, Korea and Taiwan, but they are well along in that direction, and far from the IMF’s model of a desirable financial system.

To resolve the crisis, inflation is the least costly way to reduce debt. Real interest rates have to be kept negative or at least very low, which would also reduce the pressure for bankruptcies and financial instability more generally. Household savers have been content with the low rates they have been getting, so there is no reason to raise the rate above the level of inflation. The government would let the exchange rate float, removing the impetus to raise interest rates in order to keep the currency stable. The IMF assumes that low real interest rates will lead to net capital flight and greater refinancing difficulties. It is not obvious that high real interest rates will not have an even worse effect on capital flight, because of the magnitude of bankruptcies and financial instability caused by high rates in the context of high debt/equity ratios. At the least, the trade-off has to be examined, as the IMF, reading from a script developed in low debt/equity situations, has not.

The government should push weak banks to merge with strong domestic ones. It should use its own strong balance sheet to support existing banks, not close them down or let them be bought by foreign banks. Many of these banks are bankrupt only by western standards and under transitory crisis conditions, not according to the rules of the developmental state in conditions of reasonably fast growth.

Hence the government should step in to reintroduce controls on capital movements, to create credit in order to cover the extra costs of foreign debt incurred by the devaluation (injecting equity into banks, directly buying loans from foreign creditors, and so on), thereby setting off a controlled inflation which will spred the ultimate costs among the whole population of savers and the consumers of imports. At the least, there has to be large-scale rescheduling of the debts of the private sector. Korea and Thailand are already getting some help of this kind, but need much more. Indonesia, facing a greater political crisis and a larger number and nationality of creditors among whom agreement is more difficult, has received virtually none, and