Financial Times WEDNESDAY NOVEMBER 4 1998 

MARTIN WOLF: Currency vacuum

There is a gaping hole at the centre of the G7's plan to prevent a recurrence of the recent financial crisis

Frequent repetition of a financial crisis on the present scale would be quite intolerable. So even though they are still trying to put out the fire, policymakers must also do their best to prevent another conflagration.

To their credit, the finance ministers and central bank governors of the Group of Seven leading industrial countries made a serious attempt last week. Their declaration reflected the efforts of Gordon Brown, chairman of the G7 finance ministers this year, and built upon three reports from the Group of 22 "systemically significant economies".

The resulting agenda is sensible. It includes increasing the transparency and openness of the international financial system; enhancing stability, particularly through more effective regulation; and improving ways to respond to crises, particularly through orderly workout arrangements, better insolvency regimes and lending into arrears by the International Monetary Fund. Nevertheless, the recommendations seem both too ambitious and not ambitious enough: they contain a dauntingly wide range of complex and intractable reforms, but they still do not tackle a central question, that of exchange rate regimes.

The questions raised by global financial instability are as many and interwoven as the Gordian knot; leaders must cut through it. The sharpest sword they have is advice on exchange rates. Emerging markets should be encouraged to adopt either of two extremes - either fairly freely floating regimes or robust currency boards, which fix a currency to an anchor, usually the dollar.

Yet all the G7 declaration calls for is "consideration of the elements necessary for the maintenance of sustainable exchange rate regimes in emerging markets, including consistent macroeconomic policies". Even in the G22 report on crisis prevention and management, less than two pages of the 42 are devoted to exchange rate regimes. This is Hamlet without the prince.

Exchange rate regimes matter. To understand why, one must start with the frightening number of banking crises and no less impressive number of currency crises over the last two decades. Bad though these are when separate, maximum devastation occurs when they arrive together, as happened in Chile in the early 1980s, Mexico in the mid-1990s and Thailand, Indonesia and South Korea last year.

The fiscal costs of such banking crises can range from 20 per cent to 40 per cent of gross domestic product. But a currency crisis imposes the additional trauma of a forced external adjustment. Even the IMF's relatively optimistic forecasts suggests that the four most affected Asian economies will regain 1997 levels of GDP only by 2001. Four lost years must be the least bad outcome to be expected.

Contrast that with Japan today or the UK in 1992. Japan has a banking crisis, but there is no constraint on its ability to provide cheap money and fiscal expansion at home. Britain suffered a currency crisis, but devaluation allowed lower interest rates and a swift return to growth. Neither has suffered anything like the trauma of the crisis-hit emerging countries.

The explanation is that the UK and Japan had no significant foreign-currency debt. Large quantities of foreign currency debt, unmatched by foreign currency assets of equivalent size and maturity, guarantee that a currency crisis will cause a banking collapse, and vice versa. Once the exchange rate falls, borrowers, particularly banks, will find their liabilities rising in value against their assets. Similarly, if the financial system begins to look infirm, foreign creditors will take their money out, thereby creating a currency crisis.

There is also a close connection between the size of foreign currency debt, exchange rate regimes and a currency crisis. That connection is a fixed exchange rate. Suppose the authorities have kept their currency pegged to an anchor over a lengthy period. Naturally, some borrowers treat borrowing in foreign currency as no different from borrowing in domestic currency. Similarly, some lenders begin to think their investments are protected against exchange risk.

This is a recipe for huge capital inflows and large current account deficits. Should the government wish to limit the expansionary effects on domestic credit, by raising domestic interest rates, the incentive to borrow in foreign currency will increase. As foreign debt piles up, domestic credit expands and the economy overheats. A crisis of confidence becomes inevitable and the country finds itself buried under the rubble of banking and currency collapses.

There are three ways for a country to minimise the danger: regulatory, prudential and market-based. The regulatory solution is to create a healthy and well-capitalised banking system, while curbing foreign currency mismatches and perhaps imposing permanent curbs or taxes on debt-creating inflows of foreign currency. The prudential solution is to accumulate sizeable foreign exchange reserves or contingent lines of foreign credit. The market solution is a floating exchange rate.

With a floating exchange rate, foreign currency risk is an abiding presence. Floating rates economise on what emerging markets lack - honest, effective and public-spirited regulation. Regulation will still be needed, but its defects should be less disastrous.

Floating exchange rates also economise on international lender-of-last resort facilities. Indeed, the principal G7 suggestion for dealing with contagion, the "contingent short-term line of credit" is, in practice, largely needed to stabilise pegged exchange rate regimes, such as that now in Brazil. Under floating exchange rates, external funds may still occasionally be required to halt collapses in the rate, but these are likely to be more infrequent and less dramatic than when an established peg is broken.

The G7's silence on exchange rate regimes is understandable, since the topic is so controversial. But it is a mistake. The choices should be debated now.

First, countries can continue to operate adjustable-peg or other tightly managed regimes, but they will then need tough regulation of the financial system and, in all probability, control over capital flows as well. There may also need to be large-scale contingent external assistance, in the form of funds available to prevent, cushion, or halt currency collapses.

Second, countries can adopt fully fixed exchange rates, such as Argentina's currency board. This can work, but it demands a liquidity policy for both the banking system and the country, with large reserves, not just to back narrow money, but to support the banking system through panic. Flexible nominal prices and wages will be needed. So may external contingency funds. There are valid arguments for choosing this option: the country may feel politically unable to develop a credible, domestic monetary regime; alternatively, it may fear the destabilising fluctuations in real exchange rates that accompany a float. But it remains a huge gamble.

Finally, countries can float. Some of the need for regulation under adjustable pegs will then be taken care of automatically, via market forces. True, floating rates may also demand foreign currency support in a world crisis, but less than if a peg is to be defended. Floating exchange rates are the worst possible system - except for all the others.

Martin.Wolf@FT.com