Of course, crisis prevention is the main policy challenge during the vulnerability phase. Historically, the corporate sector has not appeared on the radar screen of national policymakers, but recent crises are giving corporate health the attention it deserves. The best way to prevent crises is to look for signs of vulnerability through careful monitoring of the balance sheets of the corporate and financial sectors. Traditionally, aggregate external debt data have been used as indicators of vulnerability.3 However, debt data would not have served as reliable crisis indicators for the nine corporate crisis episodes analyzed here. The ratio of total external debt to GDP did rise during the four years prior to seven of these crises, but debt rose through most of the past 25 years for these countries as well (based on data from the IMF's World Economic Outlook database). Monitoring of the share of short term external debt also would have given numerous false positive signals. These debt data are not reliable leading indicators of corporate sector problems because much of the debt is an obligation of the government, rather than the private sector. The share of total external debt accounted for by private borrowers and not guaranteed by the government does increase ahead of eight of the nine crises and gives fewer false positives, but is still far from reliable. Stock market indices are another useful indicator of corporate vulnerability. In general, stock market indices (here measured in U.S. dollar terms and normalized by the U.S. The burgeoning "early warning system" literature that aims at identification of the macroeconomic developments that presage currency and banking crises is reviewed in Berg and Patillo, (1999). Stock market index) fell sharply during the two to three years before the crisis trough (Figure 3). Stock markets bottomed out several months ahead of the trough of industrial production during the 1990s crises with the exception of Mexico. It should be kept in mind that the limited period of data availability makes it difficult to determine with confidence whether or not these declines could have served as signals of crisis. Still, it is clear that policymakers should pay careful attention to the domestic stock market.
Aggregate corporate leverage, which is typically proxied by the debt equity ratio for the corporate sector is a good summary indicator of corporate vulnerability. Unfortunately, comparable cross country data on corporate debt are generally not produced by official sources, especially for the middle income emerging market countries prone to financial crisis. The importance of this data shortfall is worth emphasizing. It may reflect the expense of collecting data on a large number of corporations, as well as confidentiality concerns on the part of corporations themselves. In addition, the need for such data has not been urgent in the past because episodes of extreme corporate distress have been few in number and moderate in scale. However, the shortage of these data hindered the ability of the government to act early to reduce vulnerability to the crises of the 1990s where the corporate sector played a key role. The emerging market corporate leverage data reported in Claessens (1998s; Figure 4) indicate that as of 1996 debt equity ratios varied from 61 percent for Peru to 355 percent for Korea.
Thus, it seems fair to conclude that cross-country differences in corporate leverage are sizable. Moreover, countries with a higher degree of corporate leverage are more likely to experience corporate sector crisis dynamics, judging by the high leverage for the east Asian countries that experienced corporate crises and Mexico. Leverage increased during the mid 1990s in Korea, Malaysia and Thailand (Claessens, et al., 2000).
Detailed data on the composition of corporate debt would improve monitoring of corporate vulnerability. Ideally, governments would be able to regularly monitor not just the overall level of corporate leverage, but also the maturity structure of balance sheets, the share of debt accounted for by nonbank capital markets, and the extent to which balance sheet risks are hedged. Such monitoring could help governments discern when a level of growth financed by a high degree of corporate leverage is at risk of being unsustainable. The development of complete and timely macroeconomic flow of funds data is an important and practical step that could be taken in this direction.
Corporate balance sheet data can be fed into analytical tools of aggregate corporate risk. Corporate profit simulations, which measure the impact on current profits of changes in also interesting are the countries in east Asia and Latin America with high leverage that avoided the crisis: in the Philippines corporations enjoyed relatively high rates of return (Claessens et al, 1998) and the economy may have been less vulnerable due to relatively weak linkages with the region (Roubini et al, 1998); Singaporean and Brazilian corporations reportedly have relatively low external and short term debt. Domestic and foreign interest rates and the exchange rate are a practical tool. Simulations of the impact of changes in interest rates and exchange rates on corporate cash flow for east Asian countries in Claessens et al. (1999) indicates which countries were more vulnerable. The estimated equity value (EEV) framework has been used for many years for analysis of individual corporations, and is now being applied to the aggregate corporate sector. The EEV framework links corporate balance sheets and macroeconomic policy in a way that accounts not just for the current period, but also for future periods (Gray, 1999). Currently, very few countries systematically assess corporate risk. The Bank of England, which utilizes yield spreads, equity prices, and profitability to assess corporate risk, is a notable exception.