Asia: The financial intermediation path to a vicious circle
The Bank for International Settlements (BIS) has traced the roots of the Asian crisis to fundamental weaknesses in financial intermediation, namely excessive credit growth and an over-reliance on volatile short-term funds. The effects of the crisis, in turn, are likely to be felt for some time to come owing to the vicious circle of banking crisis, asset-price collapse and recession it has generated.
by Chakravarthi Raghavan
GENEVA: A fragile financial sector, weak supervision and prudential regulation, and a corporate sector burdened with high levels of short-term debt were at the heart of the series of crises in Asia, according to the Bank for International Settlements (BIS).
But two of the macroeconomic factors that, ironically, received much praise for fuelling the dynamism of the Asian economies in the 1980s and the first half of the 1990s, played a crucial role in provoking the sudden loss of confidence. These factors were the heavy buildup of capacity in a number of sectors and the impact of the chosen exchange rate regime on both trade competitiveness and the stance of monetary policy.
Indications of excessive investment in particular sectors, BIS notes, had already emerged in 1996 when massive investment in Asia's electronics industry contributed to conditions of oversupply and a resulting price collapse in world markets. But investment had sharply increased in other areas as well (such as automobile construction, household appliances and electricity generation), at the risk of flooding local and foreign markets.
Overinvestment in particular sectors tended to erode the rates of return on new capital. In many of the crisis-hit countries, growth rates associated with a given investment rate - the so-called incremental capital/output ratios - fell markedly in the first half of the 1990s.
Referring to financial intermediation and the crisis, BIS says that while there were weaknesses specific to each, two were common to the Asian countries engulfed by the crisis.
The first weakness lay in an excessive expansion of bank credit, fuelling overinvestment, and leading to the creation of unprofitable industrial capacity and asset-price boom-and- bust cycles. The underlying fragility of the financial systems in Asia was often overlooked because a high degree of monetary and exchange rate stability, allied with rapid development of local banking systems, facilitated a long period of investment-led growth.
The second weakness was a reliance on potentially volatile forms of external finance, notably short-term bank borrowing, which made domestic economies increasingly vulnerable to swings of sentiment in the international financial markets.
For much of the 1990s, several countries have had to cope with heavy capital inflows. But investors' confidence was not at first weakened, and there were few downgradings of credit ratings before the crisis. And official surveillance of country performance also failed to identify fully the dangers many Asian economies faced.
But once the crises broke, markets panicked: exchange rates and equity markets overshot; volatility rose dramatically, with liquidity drying up in some markets; and credit-rating agencies downgraded the countries most affected.
Policy-makers faced many difficult dilemmas:
How best to deal with sudden and disruptive reversals of private capital flows was a particularly thorny problem. And as the scale of international official assistance set new records, the issue of how to hold private investors responsible for their decisions and ensure they bear a share of the costs of emergency assistance to countries in trouble received much attention.
Another question, and one that aroused some controversy, was how to set monetary policy in the immediate aftermath of a collapse of confidence in the domestic currency.
BIS underlines that the buildup of substantial global liquidity during the last few years contributed to the development of large financial exposures in Asia.
Easy global liquidity
Easy global liquidity influenced different markets in different ways. With the exception of China and Hong Kong, stock markets in most Asian economies remained relatively weak during the worldwide equity boom, with price/earnings ratios peaking towards the end of 1993, at levels well above those in US markets. There were significant declines thereafter in most markets, suggesting investors had already begun to expect lower profit growth from Asian companies, an indication of early awareness of vulnerabilities in Asia.
But the pattern of international bank lending, however, closely followed global liquidity developments, with lending to non-banks in Asia rising sharply during 1995 and much of 1996. Flows to Indonesia, Korea, Malaysia, the Philippines and Thailand reached an annual rate of $15 billion. In addition, BIS estimates suggest international inter-bank borrowing by banks in these five countries was running at an annual rate of about $43 billion - with about 40% denominated in yen and the remainder in dollars.
Government guarantees or encouragement doubtless played a part in the expansion and some foreign banks may have believed that Asian banks enjoyed implicit guarantees for foreign borrowing from their governments. And foreign transactions of the long-regulated domestic banks had been liberalized before bank managers had acquired the skills for managing foreign exchange risks or before the supervisory framework had been strengthened.
For example, Thailand's Bangkok International Banking Facility (BIBF) had allowed banks to borrow in dollars - leading authorities ultimately to tighten the rules. Also, foreign banks were led to believe that the scale of their BIBF operations would affect the chances of their receiving a licence to operate in domestic markets.
Long-standing policies of fixed or quasi-fixed exchange rates probably nurtured a misperception of exchange rate risk, and a type of "real interest rate illusion" (that is, dollar or yen interest rates deflated by local inflation rates) led to over-borrowing in foreign currency.
A very high proportion of international bank lending was either of short-term maturity or, if long-term, carried floating rates.
And lending banks regard short-term lending as safer than the long-term vicinity, since it mirrors the maturity of much of their funding - a view enshrined in supervisory and risk- weighting practice.
[There is now a growing view that banking supervisors should require a higher risk-weighting for short-term loans, and thus higher capital adequacy provisions.]
And while comprehensive data is lacking, it would appear that long-term investments in real domestic assets like property were financed by short-term bank loans.
Movements in global liquidity also were reflected in spreads on emerging-market debt instruments, with spreads narrowing significantly early in 1995. It was only in the fourth quarter of 1997, after the onset of the crisis, that spreads on emerging-market debt widened sharply - rising from almost 130 basis points during June 1997 to 375 basis points by January 1998, but falling back moderately in subsequent months. The spreads on Indonesian bonds, though, continued to widen, touching an average of 750 basis points in March.
Such large movements in spreads demonstrate the difficulties financial markets had in pricing risk.
Prudential supervisory mechanisms
Along with the rise in spreads, the credit-rating agencies also downgraded loans to Indonesia, Korea and Thailand, taking their ratings down to non-investment grade - mainly reflecting concerns about the extremely precarious international liquidity situation of the countries.
Decisions were typically taken after a substantial tightening of macroeconomic policy under IMF programmes had probably improved their long-term ability to service foreign debts.
The crisis, BIS comments, was exacerbated by two shortcomings in the management of external assets and liabilities: insufficient external liquidity and inadequate diversification into foreign financial assets.
A key issue raised is how to ensure that proper prudential safeguards apply to private-sector borrowing in foreign currency.
One approach is to use controls or taxes in borrowing countries in order to limit foreign currency borrowing or to lengthen its maturity. In the past, this has been done in several ways: by imposing quantitative restrictions, allowing only those corporations and banks with high credit ratings to borrow abroad, or requiring borrowers to maintain unremunerated accounts at the central bank equal to a percentage of their borrowing. The authorities have sometimes designed such mechanisms to discriminate against short-term and potentially volatile inflows.
A second approach is to put in place insurance mechanisms against liquidity crisis - such as that in Argentina (where the peso is convertible at par into the dollar) where commercial banks are required to hold 20% of their liabilities in liquid assets, and are also required to take steps to ensure they have access to adequate foreign currency liquidity through pre-negotiated credit lines. A third approach is to tighten regulatory constraints on creditors, particularly on creditor banks.
[And, given the Basle core principles for banking supervision, this last would mean that bank supervisors in the industrial countries should undertake better supervision of the activities of the branches of their banks in the developing world, and in offshore centres. There is already some criticism that these supervisors failed in these duties in relation to the Asian crisis, since mid-1996 at least.]
Official liquidity assistance
Analyzing the various stages of the Asian crisis, BIS comments: "It has been a complicated and in many ways unprecedented crisis. It is not yet over."
Economic policy in most Asian countries has had to address a difficult external financing environment in recent months. Between 1996 and the second half of 1997, capital movements have swung from inflows at an annual rate of almost $100 billion to outflows of about the same size.
In terms of official liquidity assistance, the scale of assistance has set new records, with the rescue packages sharing some common features with the Mexican one. A total of $117 billion was offered to Thailand, Indonesia and Korea, with the size of commitments growing with each successive package.
A reason for this is to have a psychological impact on markets. How large a package needs to be to do this is difficult to judge, says BIS, adding "international official support offered to Asian countries was not large enough to cover all their short-term foreign obligations as had the Mexican package, a difference that did not pass unnoticed in the markets."
The relative contribution of official and private creditors, BIS points out, has also been a matter of much debate in the wake of the Mexican and Asian crises.
A central aspect of the response to the 1980s debt crises was that final arrangements, at official instigation, were concluded between creditor banks and sovereign borrowers who could no longer service those debts. The negotiations were long and difficult, and took several months. But liquidity assistance from official sources was contingent on the agreement of credit or banks to roll over bank debt or provide new funds.
It did not happen that way after the Mexican crisis.
But the massive foreign official assistance allowed holders of short-term dollar-linked Mexican government debt to escape without any loss, and the scale of official assistance covering all of the country's short-term external liabilities may have set a standard by which financial markets could later judge the adequacy of subsequent packages for Asian countries.
And some observers felt that the Mexican bailout weakened the investors' sense of responsibility for their own actions. And because holders of other forms of Mexican paper (equities, long-term bonds or peso-denominated debt) suffered heavy losses, it may have distorted the pattern of capital flows from equity to debt.
Referring too to the asset-price boom caused by foreign bank lending, BIS adds: "The deflation of an asset price bubble and the necessary contraction of the banking industry are likely to depress demand in the countries affected for a significant period of time. The collapse of equity values in many centres has saddled banks with large unrealized losses that are a potential claim on already weak bank capital.
"As some property developers default on their bank loans, banks could be left holding real estate that may not be salable at its collateral value.
"The effects of economic slowdowns, asset price collapses and banking crises tend to be mutually reinforcing as the curtailment of bank credit depresses asset prices and further deepens recessions. This in turn creates additional problems for banks that are forced to retrench still further. 'Vicious circle' has been an overworked term, but it describes the banking crisis/asset price collapse/recession interaction too well." (Third World Economics No. 187/189, 16 June-15 July 1998)
The above article was originally published in the South-North Development Monitor (SUNS) of which Chakravarthi Raghavan is the Chief Editor.