Chapter

9 Corporate Governance and Corporate Debt in Asian Crisis Countries

Author(s):
David Coe, and Se-Jik Kim
Published Date:
September 2002
Share
  • ShareShare
Show Summary Details

Eric Friedman, Simon Johnson, and Todd Mitton*

High levels of corporate debt contributed to the severity of the Asian financial crisis in 1997-98. Many Asian companies were more highly indebted and had a greater proportion of short-term debt than their counterparts elsewhere in the world. As a result, even a small shock to the macroeconomy was enough to push them into insolvency. Companies with more debt suffered larger falls in stock prices (Mitton 2001), and countries with more corporate debt suffered larger falls in output during the crisis (Kim and Stone, 1999). Why did the Asian private sector choose to have so much debt?

There are two plausible explanations. The first is that the debt levels of Asian firms were the result of rational choices based on standard financial decision-making. For example, the size, profitability, and growth prospects of a firm may affect its level of debt (see, for example, Titman and Wessels, 1988, and Myers, 1977). In this view, Asian firms were highly leveraged because management and investors were optimistic about the future. Such optimism is hard to avoid in a fast growing country.

The second explanation is that debt levels were high because the country-level institutions protecting investors were weak. La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1997, 1998, 2000) find that weaker investor protection leads to less financial development, particularly to smaller equity markets. As a result, companies have to place greater reliance on debt finance (Rajan and Zingales, 1998a). Rajan and Zingales (1998b, 2001) also emphasize institutions, but place more emphasis on the political allocation of capital. In this view, Asian firms had higher levels of debt because this was how politicians channeled resources to chosen firms. While the institutions-based view is usually presented at the country level, we show below that it has strong testable implications at the firm level. In particular, firms with weaker corporate governance will tend to be more indebted and this correlation should be stronger when country-level institutions protecting investors are weaker.1

This paper assesses the financial and institutional views of Asian pre-crisis corporate debt. We find strong evidence at the firm level that standard financial considerations, such as size, profitability, and growth were important determinants of debt levels. But controlling for these variables, we also find that debt was higher when firm-level governance was weaker. Across Asian countries open to capital flows, this result is stronger where country-level institutions are weaker.

Our findings are consistent with and extend the recent literature on corporate debt in Asia. Lee, Lee, and Lee (1999) examine changes in leverage of Korean firms from 1981 to 1997. They find that while debt levels in Korea can be explained partly by standard corporate finance models, even after controlling for factors such as size, growth, profitability, and fixed assets, chaebol firms have higher leverage than non-chaebol firms. Bongini, Ferri, and Hahm (1999) also show that chaebol firms have greater leverage. Kim and Lee (2000) argue that leverage, especially short-term leverage, explains the performance of Korean firms after the Asian financial crisis. Alba, Claessens, and Djankov (1998) argue that weaknesses in corporate governance and capital structure contributed to the crisis in Thailand. Claessens, Djankov, and Lang (1998) show that while vulnerabilities in the financial structure of East Asian corporations existed in the early 1990s, short-term borrowing increased during the 1990s.

At the same time, our findings fit within the broader recent literature on the cross-country effects of institutions. Countries with weaker legal protection for minority shareholders have smaller equity markets, other things being equal, and use less outside finance (La Porta, Lopez-de-Silanes, Shleifer, and Vishny (LLSV), 1997a, 1998, 2000a). Across the world (including Asia), the quality of legal institutions is strongly correlated with “legal origin,” meaning whether the country’s institutions derive from a common law or civil law tradition (LLSV 1998). Protection for minority shareholders is weaker in countries with a civil law tradition.

Our theoretical link between weak institutions, firm-level governance, and debt builds on the model of expropriation by managers in Jensen and Meckling (1976). Burkart, Gromb, and Panunzi (1998) introduce the assumption that most diversion by management is costly, for example, because it involves legal maneuvers. LLSV (1999b) model the comparative cost of expropriation across countries in a simple static framework. This approach has been developed also by Johnson, Boone, Breach, and Friedman (2000).

Johnson, La Porta, Lopez-de-Silanes, and Shleifer (2000) define tunneling as the legal expropriation of investors by controlling shareholders. We use this term throughout our analysis. We also introduce the term “propping” to indicate the reverse process, i.e., the discretionary but legal transfer of resources into a firm that benefits its minority shareholders and creditors.

The next section provides a simple framework within which we can develop testable hypotheses about the correlation between firm-level corporate governance arrangements and debt. We then describe the corporate governance and financial variables that we use in our analysis, present country- and firm-level regression evidence, and discuss the macroeconomic implications of our findings.

Debt with Tunneling and Propping2

A Simple Model

Consider a family or “group” that has one publicly traded firm. This firm is controlled by the group but has separate legal status. The group owns share α of the firm and outsiders own share 1 − α. Retained earnings are denoted I. In period t, the group tunnels (or expropriates) St of retained earnings from the firm and obtains utility of St The expected cost of tunneling is (St2/2k). We assume that kI and k is set outside the control of the firm (i.e., it is the country-level of investor protection). A higher level of k indicates less investor protection and a lower cost of tunneling.

Tunneling is wasteful in the sense that it reduces the amount invested. The group invests what it does not tunnel in a project that earns gross rate of return Rt, in period t, and from which it obtains share α. To simplify the analysis, we assume that the support of Rt, is contained on [0, 1/α].3 Also, we assume that the stochastic variable Rt, is persistent: i.e., [Rt+1 | Rt] first order stochastically dominates [Rt+1 | Rt] when Rt>Rt. For example, one reasonable model could be Brownian motion with mean reversion. We assume that the group observes Rt before choosing St.

The publicly traded firm needs to make a debt payment, D, each period.4 In our simple model this payment does not vary over time. It can be considered as the regular payment due on a long-term bond. The firm’s profit in period t is therefore:

We assume that if in any period the firm’s profit (including debt payment) would be negative then bankruptcy is declared and the firm ceases to operate.5 This means that there are no future profits or debt payments or opportunities to tunnel these assets.

Intuitively, the group equates the marginal cost and marginal benefit of tunneling. Because the group owns α of the firm, it has an incentive to invest at least some of the firm’s assets rather than to tunnel them all. As α rises, the amount of tunneling in equilibrium falls. As k rises, the amount of tunneling in equilibrium grows.

We now solve for the group’s optimal behavior by solving the stochastic dynamic program given below.

The group’s expected payoff in any period is:

This can be conveniently written as

where H(x)=0 if x〈0, and H(x)= 1 otherwise. Let δ denote the discount factor. Then the Bellman equation for the group’s value function (expected discounted present and future earnings) can be written as:

and thus the group’s expected payoff is V(R0).

Now we solve this problem. It is easy to see that the value function V(Rt) is strictly positive and non-decreasing in R, as is its conditional expectation W(Rt)=E[V(Rt′] | Rt], by persistence. Let

Solving the first-order condition gives:

which is the solution for the static model without debt.

First we consider the case when δ = 0 and future payoffs do not matter. Then the group’s optimization problem deals only with a single-period:

Note that the function to be maximized is continuous with (at most) 2 local maxima. Thus we can show that the optimal policy is S*(Rt)=Su(Rt) for RtRm and S*(Rt)=k for Rt<Rm, where Rm satisfies the equation:

Note that when D=0, then Rm=0 and we get the same result as in the model without debt, but when D>0, Rm>0. In fact, Rm is strictly increasing in D. Thus, for Rt<Rm the presence of debt causes the group to “loot” (i.e., tunnel everything) when rates of return are too low, due to the impending bankruptcy. This “looting” effect of debt is similar to the intuition behind the results in Myers (1977).

Now we return to the case where δ>0 but for simplicity assume that R(t) is distributed i.i.d., in which case W(Rt)>0 is independent of R. Denote this value by w. In this case, the group’s optimization problem is:

Here, the function to be maximized has 2 local maxima and a single downward discontinuity at Sd(Rt)=I−D/Rt. Again if D=0, then the firm never goes bankrupt and S*(Rt)=Su(Rt).

However, in general the optimal decision policy, S*(Rt), can take on 2 forms depending on the relationship between Rm (which was defined in equation 1)and R+, where R+ satisfies R+*(ISu(R+))=D, i.e., R+ is the rate of return at which the firm can just make its debt payment given the amount that the group wants to tunnel. If R+<Rm then the optimal policy is the same as for the case above when δ = 0. However, if R+>Rm, the optimal policy function becomes more interesting as there are 3 regions of behavior.

In the first region, RtR+, S*(Rt)= Su(Rt). In this case the presence of debt does not alter the group’s behavior. In the second region, S*(Rt)= k for RtR, where R satisfies

In this case the group tunnels as much as possible (k) from the firm, and we see the “looting” effect of debt.

Interestingly, in the intermediate region, RRR+, S*(Rt)=Sd(Rt). Note that Sd(Rt) is increasing in Rt. In this region, Sd(Rt)〈Su(Rt) and thus the presence of debt actually reduces expropriation, as the group is trying to protect his future earnings. We call this the “incentive” effect of debt because the debt induces better performance by groups (from the perspective of shareholders), as argued by Jensen (1986).

These three regions are illustrated in the Figure 1 where tunneling St is on the y-axis and Rt is on the x-axis. The dark line is S*(Rt), the optimal amount of tunneling given the value of Rt. The straight line from (0, k) to (1/α,0) is Su (Rt), which would be the optimal policy for D=0. As mentioned above, the presence of debt may reduce tunneling by the group in the intermediate region, R+>Rt>R, in which the group may tunnel less in order for the firm to remain solvent. In this region debt strengthens the group’s incentives to act in the interest of shareholders, as suggested by Jensen and Meckling (1976) and Jensen (1986).

Figure 1.Debt, Propping and Tunneling

Note that in this intermediate region, the group may even tunnel less than 0, i.e., the group puts its own money into the firm to prevent bankruptcy. The explanation is that earnings in the future (both from profit sharing and tunneling) are valuable and the group wants to keep the firm in business in order to have that opportunity. This is “propping.”

There is also a region, when Rt is small, in which the presence of debt increases the amount of expropriation by the group since the firm is going bankrupt and thus there is no gain to be had from firm profits. In this region there can be a “debt overhang” that causes groups to expropriate more and thus, as in Myers (1977), hurt the interests of outside shareholders and bondholders.

It is straightforward to show that the qualitative aspects of this analysis are not changed in the general model, where Rt, is persistent, but not necessarily i.i.d. In this case equation (2) is changed by replacing w with W(Rt). Since this function is nondecreasing, the solution still breaks into three types of regions: normal, in which there is no effect of debt; looting, where the effect of debt causes increased tunneling; and “anti-theft incentives,” where the presence of debt causes the group to tunnel less, in order to protect the firm from bankruptcy. The major difference is simply that the points at which the transition between regions occurs now have to be derived by solving the complete stochastic dynamic program.

In a dynamic context, we have shown the effects of debt on tunneling compared to the 100 percent equity financing case. If the return on investment, R, is sufficiently high, then the presence of debt has no effect. If R is sufficiently low, then the presence of debt can increase the amount of tunneling (a result in line with the intuition regarding “debt overhang” in Myers, 1977). However, if R takes on an intermediate value, then having debt will mean less tunneling, because groups want to keep the firm in business for the future (in this case, the intuition about the effects of debt is similar to that in Jensen and Meckling, 1978 and Jensen, 1986).

In this simple dynamic framework, higher levels of debt can induce better behavior from controlling shareholders (i.e., propping), but it also raises the probability of a collapse with complete looting. The cost of debt is that, in other states, it induces collapses that would not have occurred had the firm been financed just with equity. The optimal choice of debt levels, therefore, depends on the relative costs and benefits of these two considerations.

Extensions

Simple extensions of this model allow us to establish a number of useful results. First, we can allow entrepreneurs to choose the mix of debt and equity when raising capital. Friedman and Johnson (2000) show that under plausible conditions, debt levels will be higher when country-level investor protection (parameter k) is weaker.

We can also allow the entrepreneur to choose his level of firm-specific corporate governance and debt at the same time. An entrepreneur can choose to have stronger corporate governance and less debt or weaker corporate governance and more debt. In this framework, firm-specific corporate governance and debt are substitutes. In cross-sectional firm-level regressions, therefore, we should expect weaker corporate governance arrangements to be correlated with more debt.

A further extension allows debt of different maturities. It seems reasonable to assume that shorter maturity debt is more effective for inducing propping but also more costly in the sense of raising the probability of collapse. The maturity choice is therefore a trade-off between these costs and benefits. It is straightforward to show that maturities will be shorter when country-level investor protection is weaker and when the group chooses weaker firm-specific corporate governance.

Kim (2000) models the broader issue of group formation in a way that is consistent with our approach. In his model, firms borrow from banks and then decide whether or not to form conglomerates in which their debts are cross-guaranteed. After one period, the bank decides whether to liquidate or bailout firms that cannot make an initial payment on the loan. Risk-averse firms have an incentive to join a group, because this will make banks less likely to liquidate them. In this model, groups offer a form of insurance for individual entrepreneurs—in our terminology, they provide the resources necessary for propping, although they also reduce the cost of tunneling.

Macroeconomic Implications

In our model, the distribution of payoffs to outside investors will generally be skewed, with a longer and “fatter” left tail than is usual. Even if the underlying returns, R, are normally distributed, the actual outcomes can have the double hump shape shown in Figure 2. This arises because if returns, R, are slightly low, there is propping, but if R falls below a critical level, then looting occurs.

Figure 2.Corporate Governance and Crises

In this model, weaker legal institutions lead to fewer projects being financed (Friedman and Johnson, 2000). But weak legal institutions can also contribute to economic crises. Having weak protection of investor rights does not make shocks more likely, but it does mean that negative shocks have larger effects on the overall economy. In this view, institutions matter for a particular aspect of volatility - whether countries can suffer large collapses. Reasonable corporate finance arrangements in a weak legal environment can lead to a bimodal distribution of outcomes, i.e., either the economy does well or it collapses.

Testable Predictions

In this framework, debt acts as a substitute for country-level investor protection and firm-level corporate governance. If more frequent monitoring is helpful to the lender, then short-term debt will be more effective as a governance tool.

There are three testable predictions. Countries with weaker country-level legal protection for investors should have on average more debt relative to assets and, potentially, more short-term debt relative to long-term debt.

Second, within a country, firms with weaker firm-specific corporate governance will have more debt and more short-term debt. This correlation should be stronger in countries where institutions are weaker.

Third, when the political allocation of capital is stronger, there will be more debt, assuming that politicians care about the performance of the firms. Within a country, those with stronger political connections should have more debt.

Measuring Corporate Governance

Country-Level Institutions

LaPorta, Lopez-de-Silanes, Shleifer, and Vishny (LLSV 1997a, 1997b, 1998) show that there are systematic differences in the legal rights of investors across countries. LLSV (1998) propose six dimensions—taken from commercial codes or company laws—to evaluate the extent of protection of minority shareholders against expropriation. First, the rules in some countries allow proxy voting by mail, which makes it easier for minority shareholders to exercise their voting rights. Second, the law in some countries blocks the shares for a period prior to a general meeting of shareholders, which makes it harder for shareholders to vote. Third, the law in some countries allows some type of cumulative voting, which makes it easier for a group of minority shareholders to elect at least one director of their choice. Fourth, the law in some countries incorporates a mechanism that gives the minority shareholders who feel oppressed by the board the right to sue or otherwise get relief from the board’s decision. In the United States, this oppressed minority mechanism takes a very effective form of a class action suit, but in other countries there are other ways to petition the company or the courts with a complaint. Fifth, in some countries, the law gives minority shareholders a preemptive right to new issues, which protects them from dilution by the controlling shareholders who could otherwise issue new shares to themselves or to friendly parties. Sixth, the law in some countries requires relatively few shares to call an extraordinary shareholder meeting, at which the board can presumably be challenged or even replaced, whereas in other cases a large equity stake is needed for that purpose. LLSV (1998) aggregate these 6 dimensions of shareholder protection into an anti-director rights index by simply adding a 1 when the law is protective along one of the dimensions and a 0 when it is not.

The highest shareholder rights score in the LLSV (1998) sample of 49 countries is 5. Investor protection is significantly higher in common law countries, with an average score of 4, compared with an French-origin civil law countries, with an average score of 2.33. In the LLSV (1998) data, there is no association between a country’s level of economic development and its antidirector rights score, but a strong association between the score and the size of its stock market relative to GNP.

LLSV (1998 and 1999a) also find that the legal enforcement of contracts is weaker in countries with a civil law tradition. For example, the efficiency of the judicial system is on average 8.15 in English-origin countries (on a scale of 1-10, where 10 means more efficient), but only 6.56 in French-origin countries. Legal origin affects investor protection both through the rights available in the laws and how easy it is to enforce these rights. We use these measures in our cross-country analysis.

Firm-Level Corporate Governance Measures

Mitton (2001) develops a number of measures that capture important dimensions of firm-level governance. He focuses on three aspects of corporate governance that vary among firms within the same country: disclosure quality, ownership structure, and corporate diversification.

Disclosure quality is an important element of corporate governance. LLSV (1998) argue that accounting standards play a critical role in corporate governance by informing investors and by making contracts more verifiable. While LLSV (1998) and Johnson et al. (2000) employ country-specific measures of accounting standards in their studies, Mitton (2001) proposes two ways in which disclosure quality can be measured at the firm level. First, he proposes that a firm will have higher disclosure quality if it has a listed American Depository Receipt (ADR). This higher disclosure quality can emerge formally, through mandated disclosure requirements of the listing exchange (for level II and III ADRs), or informally, through a larger pool of investors spurring increased demand for disclosure and increased scrutiny of the firm’s reports (see Coffee, 1999). Reese and Weisbach (2001) argue that increased protection of minority shareholders is a primary motivation for non-U.S. firms to cross-list in the U.S. (see also Stulz, 1999). Lins, Strickland, and Zenner (2000) show that the sensitivity of investment to cash flow falls when an ADR is issued by a company from a country with a weak legal system and a lessdeveloped capital market (as defined by LLSV, 1997).

Mitton (2001) also proposes that a firm may have higher disclosure quality if its auditor is one of the Big Six international accounting firms. Previous research (e.g. Reed, Trombley, and Dhaliwal, 2000; Titman and Trueman, 1986) has associated Big Six auditors (or Big Eight auditors, for earlier years) with higher audit quality. The Big Six firms may be more likely to ensure transparency and eliminate mistakes in a firm’s financial statements because they have a greater reputation to uphold (Michaely and Shaw, 1995), because they may be more independent than local firms, and because they face greater legal liability for making errors (Dye, 1993). Additionally, even in cases in which actual disclosure quality is not higher, Big Six auditors may offer higher perceived disclosure quality and allay investors’ fears because of their prominent, recognizable names (see Rahman, 1998).

The second aspect of corporate governance studied by Mitton (2001) is ownership structure. Shleifer and Vishny (1997) argue that ownership concentration is, along with legal protection, one of two key determinants of corporate governance. Large shareholders can benefit minority shareholders because they have the power and incentive to prevent expropriation. On the other hand, large shareholders can themselves engage in expropriation. La Porta, Lopezde-Silanes, and Shleifer (1999) find high degrees of ownership concentration in firms from countries with relatively poor shareholder protection, and argue that the conflict between large shareholders and minority shareholders is the primary corporate governance problem in such countries. Morck, Strangeland, and Yeung (2000) and Bebchuk, Kraakman, and Triantis (2000) discuss how controlling shareholders may pursue objectives that are at odds with those of minority shareholders. Concentrated ownership also plays an important role in some European countries. For example, Gorton and Schmid (1999) find that firms are more highly valued when large shareholders own more shares in Germany. In data from 22 emerging markets before the crisis, Lins (2000) shows that large blockholders generally increase firm value, and that divergence between cash flow rights and control rights of controlling management groups and their families corresponds to lower firm values.

The third aspect of corporate governance studied by Mitton (2001) is corporate diversification. While diversification is not a corporate governance mechanism per se, previous research has suggested that agency problems are different within diversified firms. The lower transparency of diversified firms in emerging markets results in a higher level of asymmetric information that may allow managers or controlling shareholders to more easily take advantage of minority shareholders (see Lins and Servaes, 2000; Lins, 2000). If expropriation of minority shareholders increases during a crisis period, then the associated loss in firm value could be particularly pronounced for diversified firms. While diversification can also offer the benefit of improving capital allocation (Stein, 1997), particularly in emerging markets (Khanna and Palepu, 2000), this benefit could disappear in a time of crisis as investment opportunities diminish.

We follow Mitton (2001) in constructing these firm-level corporate governance variables for our sample of East Asian firms. Using data from the Bank of New York, we identify firms that had a listed ADR prior to 1997. Using data from Worldscope, we identify which firms had Big Six auditors prior to 1997. Because of the importance of name recognition, we do not include auditing firms that use local names even if they have affiliations with Big Six firms. (Note that by this definition Korean firms, by and large, are not coded as having Big Six auditors, because the major Korean accounting firms have Korean names, even though some have affiliations with Big Six firms.) Using ownership data from Worldscope, we identify the percentage shareholdings of the largest shareholder in each firm for which ownership data are available. We further classify shareholdings as “management” shareholdings if the owner is listed as an officer or director of the company. Finally, using data from Worldscope, we classify firms as diversified if they operate in more than one industry, where industries are defined at the two-digit standard industrial classification (SIC) level.

The Determinants of Corporate Debt

We are testing two hypotheses. First, there should be higher levels of corporate debt relative to assets and more short-term debt in countries where institutions are weaker. Second, higher levels of debt and higher levels of short-term debt should be correlated with weaker investor protection at the firm level. This relationship should be stronger within countries where there is less legal protection for investors.

Country Averages

Average levels of debt in the Asian corporate sector were not particularly high compared to other countries. Table 1 shows that long-term debt was 12.9 percent of total assets on average across all Asian countries at the end of 1996, less than in the United States (19.8 percent) or Europe (15.6 percent). Even comparing the most heavily indebted decile of firms across countries, we find a lower ratio of long-term debt to assets in Thailand or Korea than in the United States. More unusual was the relatively high average ratio of short-term debt to assets and the high level of short-term debt for the most indebted firms. In all Asian countries open to capital flows (Korea, Indonesia, Malaysia, the Philippines, Singapore, Hong Kong, and Thailand), short-term debt levels were significantly higher than in other financial markets.

Table 1.Debt Levels of Asian Corporations Prior to the Financial Crisis
Number of FirmsLong-Term Debt/Total AssetsShort-Term Debt/Total AssetsGrowth MedianProfitability Median
MeanMedianStd. Dev.90th %ileMeanMedianStd. Dev.90th %ile
China980.0760.0360.1050.2250.1760.1630.1050.32712.7%6.8%
Hong Kong2810.1040.0760.1030.2440.1450.1130.1520.32311.1%5.7%
Indonesia1060.1790.1460.1740.4370.1670.1610.1300.35513.8%8.8%
Japan2.0830.1290.1040.1200.2870.1700.1420.1490.3602.4%2.0%
Korea2110.2160.2040.1560.3410.2850.2810.1310.45015.1%4.8%
Malaysia3160.1110.0700.1300.2690.1440.1120.1430.33517.6%8.0%
Philippines680.1090.0470.1280.3190.1340.1050.1330.33822.0%7.3%
Singapore1740.1100.0560.1280.3020.1200.0860.1160.27510.4%4.3%
Taiwan1930.0910.0640.0960.2100.1500.1190.1200.33514.4%6.8%
Thailand2010.1690.1370.1560.4040.2570.2420.1760.49910.6%6.1%
All East Asian countries3.7310.1290.0940.1300.3040.1750.1520.1360.36013.0%6.1%
Hong Kong, Singapore, and Taiwan6480.1010.0650.1090.2520.1380.1060.1290.31112.0%5.6%
China, Indonesia, Korea, Malaysia, Philippines, and Thailand1,0000.1430.1070.1420.3320.1940.1770.1370.38415.3%7.0%
Comparative statistics
United Kingdom1,2690.1190.0770.2130.2680.0860.0480.1270.2064.0%7.7%
United States3,2730.1980.1570.2170.4610.0470.0150.1480.1119.0%6.9%
Latin America3590.1330.1000.1300.2820.1070.0830.1150.22515.8%8.2%
Europe2,1930.1560.1300.1400.3340.0960.0690.1060.2095.2%5.4%
Other emerging markets6910.1120.0750.1200.2500.1380.1O80.1240.29922.6%11.6%
Notes: Data come from financial statements with fiscal year-ends closest to the end of December 1996, and represent all non-financial firms available in the Worldscope database. Growth refers to the one-year growth rate in total assets. Profitability refers to return on assets.
Notes: Data come from financial statements with fiscal year-ends closest to the end of December 1996, and represent all non-financial firms available in the Worldscope database. Growth refers to the one-year growth rate in total assets. Profitability refers to return on assets.

Demirguc-Kunt and Maksimovic (1999) show that firms in countries with a common-law tradition or with better legal systems generally use more long-term debt and a smaller proportion of short-term debt relative to long-term debt. However, this does not appear to be a completely robust result for Asia before the crisis. Table 2 shows that controlling for log GNP per capita, there is only a weak correlation between legal systems (i.e., British or French legal origin) and debt levels. Indeed, the only variable in Table 2 that debt levels are correlated with is French legal origin.

Table 2.Corporate Debt and Country-Level Institutions
Dependent variable is agregate total debt/aggregate total assets
GDP growth0.0100.0100.0100.010
(1.3)(1.4)(1.1)(1.1)
Log GNP0.0070.0100.0100.010
(0.7)(0.8)(1.1)(1.1)
Rule of law0.0000.0000.0000.000
(0.1)(0.1)(−0.8)(−0.8)
French origin−0.050**−0.040
(−2.1)(−1.2)
German origin0.0000.010
(0.0)(0.1)
Scandinavian origin0.0100.020
(0.4)(0.5)
Creditor rights0.0050.000
(0.5)(−0.3)
Anti-director rights0.0110.006
(1.5)(0.5)
Number of observations40404040
Adjusted R-squared0.020.050.04−0.02
Notes: The table presents results of OLS regressions of levels of corporate debt on country-level measures of legal protection. The dependent variable is aggregate total debt of all firms in the country divided by aggregate total assets of all firms in the country. All countries with 30 or more non-financial firms in the Worldscope database as of the end of 1996 are included. “Rule of Law”, “Creditor Rights”, and “Antidirector Rights” are based on scores defined in La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998). Also estimated but not reported is a constant term. Heteroskedasticity-robust t-statistics are in parentheses, and asterisks denote levels of significance (**=5%).
Notes: The table presents results of OLS regressions of levels of corporate debt on country-level measures of legal protection. The dependent variable is aggregate total debt of all firms in the country divided by aggregate total assets of all firms in the country. All countries with 30 or more non-financial firms in the Worldscope database as of the end of 1996 are included. “Rule of Law”, “Creditor Rights”, and “Antidirector Rights” are based on scores defined in La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998). Also estimated but not reported is a constant term. Heteroskedasticity-robust t-statistics are in parentheses, and asterisks denote levels of significance (**=5%).

There appears to be a more robust relationship using data on short-term debt. Countries with a weaker rule of law have higher levels of short-term debt (Table 3). French and German legal origin are also correlated with higher short-term debt levels. Weaker anti-director rights are associated with higher short-term debt levels.

Table 3.Short-Term Corporate Debt and Country-Level Institutions
Dependent variable is agregate total short-term debt/aggregate total assets
GDP growth0.010

(1.3)
0.010

(1.5)
0.010

(1.4)
0.010

(1.4)
0.010

(1.2)
Log GNP0.020

(1.6)
0.020*

(1.7)
0.020*

(1.7)
0.010

(1.1)
0.010

(1.1)
Rule of law−0.020**

(−2.5)
−0.020**

(−2.3)
−0.020**

(−2.3)
−0.020***

(−3.0)
−0.020***

(−2.7)
French origin0.080**

(2.2)
0.070

(1.4)
German origin0.150***

(5.6)
0.130***

(3.5)
Scandinavian origin0.080

(1.5)
0.080

(1.4)
Creditor rights−0.010

(−0.7)
−0.003

−[0.2]
0.010

(0.4)
Anti-director rights−0.030**

(−2.5)
−0.030**

−[2.5]
0.230

(1.5)
−0.010

(−0.8)
Number of observations

Adjusted R-squared
40

0.11
40

0.22
40

0.19
40

0.27
40

0.24
Notes: The table presents results of OLS regressions of levels of short-term corporate debt on country-level measures of legal protection. The dependent variable is aggregate total short-term debt of all firms in the country divided by aggregate total assets of all firms in the country. Short-term means maturity of less than a year. All countries with 30 or more non-financial firms in the Worldscope database as of the end of 1996 are included. “Rule of Law”, “Creditor Rights”, and “Antidirector Rights” are based on scores defined in La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998). Also estimated but not reported is a constant term. Heteroskedasticityrobust t-statistics are in parentheses, and asterisks denote levels of significance (***=1%, **=5%, *= 10%).
Notes: The table presents results of OLS regressions of levels of short-term corporate debt on country-level measures of legal protection. The dependent variable is aggregate total short-term debt of all firms in the country divided by aggregate total assets of all firms in the country. Short-term means maturity of less than a year. All countries with 30 or more non-financial firms in the Worldscope database as of the end of 1996 are included. “Rule of Law”, “Creditor Rights”, and “Antidirector Rights” are based on scores defined in La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998). Also estimated but not reported is a constant term. Heteroskedasticityrobust t-statistics are in parentheses, and asterisks denote levels of significance (***=1%, **=5%, *= 10%).

The cross-country evidence, therefore, is broadly supportive of the idea that weak institutions were correlated with higher debt levels for the Asian corporate sector before the crisis, but it is not conclusive. With country-level data it is very hard to distinguish the financial and institutional hypotheses. We therefore turn to firm-level analysis. However, we continue to differentiate between two sets of Asian countries open to capital flows: those with relatively good country-level governance institutions (Hong Kong, Singapore, and Taiwan) and those with relatively weak institutions (China, Indonesia, Korea, Malaysia, Thailand, and the Philippines.)

Debt and Firm-Level Investor Protection

There is evidence that corporate debt levels in 1996 across Asia were partly determined by reasonable financial considerations. When we pool all 9 countries, Table 4 shows that larger firms had more debt and more profitable firms had significantly less debt. Growth, measured as the one-year percentage increase in assets, is not a significant determinant of debt levels in columns 1-8, but it is significant for Hong Kong, Singapore and Taiwan.

Table 4.Firm-Level Corporate Governance and Debt Ratios in Asia
Nine Asian countriesChina, Indonesia, Korea, Malaysia, Philippines, ThailandHong Kong, Singapore, Taiwan
(1)(2)(3)(4)(5)(6)(7)(8)(9)(10)
Ownership Concentration−0.20***

(−4.7)
Mgt. Ownership Concentration−0.16

(−1.6)
−0.15

(−1.5)
−0.42

(−1.5)
−0.40

(−1.5)
0.05

(0.7)
0.05

(0.8)
Non-Mgt. Ownership Concentration−0.22***

(−5.1)
−0.20***

(−4.7)
−0.32***

(−5.3)
−0.30***

(−4.9)
−0.07

(−1.3)
−0.05

(−1.0)
Diversified0.03*

(1.8)
0.05*

(1.7)
0.06*

(1.7)
0.03

(1.0)
Big 6 Auditor−0.02*

(−1.7)
−0.03

(−1.2)
−0.02

(−0.5)
−0.04

(−1.6)
ADR−0.01

(−0.3)
0.00

(0.2)
−0.05

(−1.1)
0.03

(1.2)
Size (Log total assets)0.06**

(2.3)
0.06**

(2.1)
0.07***

(3.5)
0.07***

(3.8)
0.08***

(3.8)
0.05*

(1.7)
0.06

(1.4)
0.06

(1.2)
0.04**

(2.5)
0.03**

(2.0)
Profitability (ROA)−0.96***

(−3.2)
−0.96***

(−3.2)
−0.98***

(−4.0)
−0.99***

(−4.1)
−0.99***

(−4.2)
−0.95***

(−3.1)
−1.29**

(−2.3)
−1.28**

(−2.3)
−0.63***

(−5.9)
−0.62***

(−5.7)
Growth (1-yr. % increase assets)0.03

(1.0)
0.03

(1.0)
0.05

(1.4)
0.05

(1.4)
0.05

(1.6)
0.03

(1.0)
0.02

(0.6)
0.02

(0.5)
0.09***

(3.0)
0.09***

(3.0)
Number of observations986986142714271525986612612374374
Adjusted R-squared0.250.250.270.270.270.250.240.240.160.17
Notes: The table reports coefficients from OLS regressions of debt ratios (total debt/total capital) on variables related to corporate governance. All non-financial firms from the listed Asian countries with available data in the Worldscope database are included. Data are reported for fiscal year-ends closest to year-end 1996. Definitions: Ownership Concentration is the total stock holdings of all shareholders owning 5 percent or more of the firm; Mgt. (Non-Mgt.) Ownership Concentration is the total holdings of these shareholders who are (are not) listed as officers or directors of the firm; Diversified means the firm operates in more than one two-digit SIC industry; Big Six Auditor means the firm’s financial statements are audited by a Big Six accounting firm; ADR means the firm has a listed American Depository Receipt. Also estimated but not reported is a constant, country dummies, and dummies for 11 of 12 industry categories. Heteroskedasticity-robust t-statistics are in parentheses, and asterisks denote significance levels (*=10%,**=5%, ***=1%).
Notes: The table reports coefficients from OLS regressions of debt ratios (total debt/total capital) on variables related to corporate governance. All non-financial firms from the listed Asian countries with available data in the Worldscope database are included. Data are reported for fiscal year-ends closest to year-end 1996. Definitions: Ownership Concentration is the total stock holdings of all shareholders owning 5 percent or more of the firm; Mgt. (Non-Mgt.) Ownership Concentration is the total holdings of these shareholders who are (are not) listed as officers or directors of the firm; Diversified means the firm operates in more than one two-digit SIC industry; Big Six Auditor means the firm’s financial statements are audited by a Big Six accounting firm; ADR means the firm has a listed American Depository Receipt. Also estimated but not reported is a constant, country dummies, and dummies for 11 of 12 industry categories. Heteroskedasticity-robust t-statistics are in parentheses, and asterisks denote significance levels (*=10%,**=5%, ***=1%).

At the same time, our corporate governance variables are also significant. Firms with more concentrated ownership, specifically non-management ownership had lower levels of debt. In column 8, the coefficient on non-management ownership concentration is minus 0.30, which indicates that an increase of 10 percent in ownership holdings of a blockholder not involved with management is associated with a lower debt ratio of three percent. This result is consistent with the hypothesis that a large outside blockholder has the power and incentive to monitor actions of the firm on the behalf of shareholders. The coefficient on Diversified is 0.06, indicating that diversified firms, on average, had higher debt ratios of six percent. This result is consistent with the hypothesis that agency problems are more severe within diversified firms. While not significant in column 8, the coefficients on Big 6 Auditor and ADR are negative, indicating that firms with higher disclosure quality had lower levels of debt.

Table 5 shows similar results for short-term debt relative to total assets. In this case, firm size is significantly correlated with short-term debt only in columns 9 and 10: larger firms in Hong Kong, Singapore, and Taiwan had less short-term debt. More profitable firms in all countries had lower short-term debt levels. There is no relationship between growth and short-term debt.

Table 5.Firm-Level Corporate Governance and Debt Maturity in Asia
Nine Asian countriesChina, Indonesia, Korea, Malaysia, Philippines, ThailandHong Kong, Singapore, Taiwan
(1)(2)(3)(4)(5)(6)(7)(8)(9)(10)
Ownership Concentration−0.12***

(−5.1)
Mgt. Ownership Concentration−0.10***

(−2.7)
−0.09***

(−2.6)
−0.12

(−1.6)
−0.11

(−1.5)
−0.05

(−1.2)
−0.05

(−1.2)
Non-Met. Ownership Concentration−0.13***

(−5.3)
−0.12***

(−5.0)
−0.18***

(−5.4)
−0.17***

(−4.9)
−0.06**

(−2.0)
−0.06*

(−1.8)
Diversified0.01

(1.3)
0.02*

(1.8)
0.02*

(1.8)
0.02

(0.7)
Big 6 Auditor−0.02*

(−1.8)
−0.02

(−1.6)
−0.03

(−1.5)
−0.01

(−0.3)
ADR−0.01

(−1.0)
−0.00

(−0.5)
−0.01

(−0.7)
0.01

(0.4)
Size (Log total assets)−0.01

(−1.5)
−0.01

(−1.6)
−0.01*

(−1.9)
−0.01*

(−1.7)
−0.01

(−1.4)
−0.01

(−1.6)
0.00

(−0.2)
0.00

(−0.1)
−0.03**

(−2.3)
−0.03***

(−2.6)
Profitability (ROA)−0.36***

(−6.6)
−0.36***

(−6.6)
−0.39***

(−7.3)
−0.40***

(−7.4)
−0.40***

(−7.4)
−0.36***

(−6.4)
−0.39***

(−4.8)
−0.39***

(−4.7)
−0.30***

(−4.0)
−0.29***

(−3.9)
Growth (1-yr. % increase assets)0.01

(1.0)
0.01

(1.0)
0.02

(1.5)
0.02

(1.4)
0.02

(1.6)
0.01

(1.0)
0.01

(0.9)
0.01

(0.8)
0.00

(0.1)
0.00

(0.1)
Number of observations985985142314231520985611611374374
Adjusted R-squared0.270.270.250.250.240.280.270.270.190.19
Notes: The table reports coefficients from OLS regressions of a short-term debt ratio (short-term debt/total assets) on variables related to corporate governance. All non-financial firms from the listed Asian countries with available data in the Worldscope database are included. Data are reported for fiscal year-ends closest to yearend 1996. Definitions: Ownership Concentration is the total stock holdings of all shareholders owning 5 percent or more of the firm; Mgt. (Non-Mgt.) Ownership Concentration is the total holdings of these shareholders who are (are not) listed as officers or directors of the firm; Diversified means the firm operates in more than one two-digit SIC industry; Big Six Auditor means the firm’s financial statements are audited by a Big Six accounting firm; ADR means the firm has a listed American Depository Receipt. Also estimated but not reported are a constant, country dummies, and dummies for 11 of 12 industry categories. Heteroskedasticity-robust t-statistics are in parentheses, and asterisks denote significance levels (*=10%, **=5%, ***=1%).
Notes: The table reports coefficients from OLS regressions of a short-term debt ratio (short-term debt/total assets) on variables related to corporate governance. All non-financial firms from the listed Asian countries with available data in the Worldscope database are included. Data are reported for fiscal year-ends closest to yearend 1996. Definitions: Ownership Concentration is the total stock holdings of all shareholders owning 5 percent or more of the firm; Mgt. (Non-Mgt.) Ownership Concentration is the total holdings of these shareholders who are (are not) listed as officers or directors of the firm; Diversified means the firm operates in more than one two-digit SIC industry; Big Six Auditor means the firm’s financial statements are audited by a Big Six accounting firm; ADR means the firm has a listed American Depository Receipt. Also estimated but not reported are a constant, country dummies, and dummies for 11 of 12 industry categories. Heteroskedasticity-robust t-statistics are in parentheses, and asterisks denote significance levels (*=10%, **=5%, ***=1%).

The governance results in Table 5 are again consistent with the institutions hypothesis. Firms with more non-management ownership concentration have higher levels of short-term debt. This is a robust result across all columns. More diversified companies have more short-term debt, but only in the countries with relatively weak institutions. Management ownership concentration is correlated with debt but only in the regressions with all nine countries; it is not significant in either of the sub-samples. The magnitude of the coefficient on management ownership concentration in the subsamples fits with the result in the full sample, but the smaller number of observations means that the result is not significant.

Macroeconomic Implications

Firms and countries with weaker corporate governance had higher debt levels before the onset of the Asian financial crisis. How important was debt per se in determining the severity of the crisis across countries and within countries?

We answer this question by assessing the evidence that debt is correlated with adverse outcomes at the firm level during the Asian crisis. Assessing the precise impact of higher debt is made more difficult by the fact that weaker corporate governance likely affects performance both indirectly (via higher debt) and directly (via investor confidence).

Without a valid instrument for corporate debt levels it is hard to determine precise coefficients. In particular, we cannot determine if corporate governance acts on performance directly or just through increased debt levels. This is the subject of our current research, and the results reported below should be interpreted as only preliminary.

Country-Level Outcomes

Why would corporate governance matter directly for the severity of crises? Consider the model presented above with pure equity financing. In this case, the amount of tunneling will be increasing in k, the legal protection of investors. Intuitively, this implies that there will be more tunneling in a downturn when investors have weaker enforceable rights. Johnson, Boone, Breach, and Friedman (2000) discuss this possibility in more detail. They show that if expropriation by managers increases when the expected rate of return on investment falls, then an adverse shock to investor confidence will lead to increased tunneling, and to lower capital inflows and greater attempted capital outflows for a country. These, in turn, will translate into lower stock prices and a more depreciated exchange rate.

Table 6 summarizes some of the accusations regarding expropriation of investors during the Asian crisis. For comparative purposes we also include some notable Russian cases. Clearly there is strong anecdotal evidence that tunneling occurred during the Asian crisis.

Table 6.Alleged Incidents of Expropriation in the Asian Financial Crisis
CompanyCountryDateAlleged Incident
In Asia
Bangkok Bank of CommerceThailand1996-97Bank managers moved money to offshore companies under their control.
United Engineers (Malaysia) Bhd.Malaysia1997-98United Engineers bailed out its financially troubled parent, Renong Bhd, by acquiring a 33 percent stake at an artificially high price.
Malaysia Air System Bhd.Malaysia1998The chairman used company funds to retire personal debts.
PT Bank BaliIndonesia1997-98Managers diverted funds in order to finance a political party.
Sinar Mas CroupIndonesia1997-98Group managers transferred foreign exchange losses from a manufacturing company to a group-controlled bank, effectively expropriating the bank’s creditors and minority shareholders.
Guangdong International Trust & Investment CoHong Kong/China1998-99Assets that had been pledged as collateral disappeared from the company when it went bankrupt.
Siu-Fung Ceramics CoHong Kong/China1998-99Assets that had been pledged as collateral disappeared from the company when it went bankrupt.
Samsung Electronics Co.Korea1997-98Managers used cash from Samsung Electronics to support other members of the Samsung group (notably Samsung Motors) that were losing money.
HyundaiKorea1998-99Managers of a Hyundai-controlled investment fund channelled money from retail investors to loss-making firms in the Hyundai group.
Outside Asia
TokobankRussia1998-99Creditors who may have been linked to bank managers took control of the bank and its remaining assets following default. Foreign creditors got nothing.
MenatepRussia1998Following Menatep’s bankruptcy, managers transferred a large number of regional branches to another bank they controlled.
AO YukosRussia1998-99Managers transferred Yukos’s most valuable petroleum-producing properties to offshore companies they controlled.
UneximbankRussia1999Following Uneximbank’s bankruptcy, managers moved profitable credit-card processing and custodial operations to another bank.
Notes: Reproduced from Johnson, Boone, Breach and Friedman (2000). Sources for the table are: Wall Street Journal, May 7, 1999, p. A1; April 17, 1998, p. A12; September 21, 1999, p. A1; August 25, 1999, p. A14; April 4, 1999, p. A1; April 8, 1999, p. A14; The Economist, March 27, 1999 and September 11, 1999.
Notes: Reproduced from Johnson, Boone, Breach and Friedman (2000). Sources for the table are: Wall Street Journal, May 7, 1999, p. A1; April 17, 1998, p. A12; September 21, 1999, p. A1; August 25, 1999, p. A14; April 4, 1999, p. A1; April 8, 1999, p. A14; The Economist, March 27, 1999 and September 11, 1999.

There is considerable evidence that countries with weaker investor protection suffered greater adverse effects when hit by the Asian crisis. Johnson, Boone, Breach, and Friedman (2000) present evidence that the weakness of legal institutions for corporate governance had an adverse effect on the extent of exchange rate depreciations and stock market declines in the Asian crisis. Table 7 shows that in simple OLS regressions the extent of legal protection for investors is strongly correlated with the extent of exchange rate depreciation during the Asian crisis. Table 8 shows a similar result with the extent of stock market decline as the dependent variable. Johnson, Boone, Breach, and Friedman (2000) show that corporate governance provides at least as convincing an explanation for the extent of exchange rate depreciation and stock market decline as any or all of the usual macroeconomic arguments.

Table 7.Legal Protection of Shareholders and Exchange Rate Depreciation
(i)(ii)(iii)(iv)(v)(vi)(vii)(viii)(ix)(x)(xi)(xii)(xiii)(xiv)(xv)(xvi)(xvii)(xviii)
A. Enforceability of contracts and shareholder rights
East Asia Dummy−0.06−0.1−0.005−0.09−0.05−0.13−0.04−0.01
(0.07)(0.09)(0.07)(0.09)(0.07)(0.09)(0.09)(0.12)
Judicial Efficiency0.05**0.05**0.05**
(0.02)(0.02)(0.02)
Corruption0.05**0.05**0.04*
(0.02)(0.02)(0.02)
Rule of Law0.04**0.04**0.03
(0.02)(0.02)(0.02)
Corporate Governance0.14**0.14**0.13**
(0.05)(0.05)(0.06)
Reserves0.0020.0020.0030.001
(0.002)(0.002)(0.002)(0.002)
R-squared0.340.370.390.280.280.360.190.20.270.30.310.31
Adjusted R-Squared0.310.290.280.20.210.250.150.120.140.260.220.18
Observations202020232323232323202020
B. Shareholder protection, creditor rights, and accounting standards
East Asia dummy−0.06−0.13−0.1−0.11−0.06−0.11−0.1−0.11−0.06−0.12−0.03−0.02
(0.08)(0.10)(0.10)(0.10)(0.08)(0.10)(0.07)−0.10(0.10)(0.12)(0.10)(0.12)
Antidirector Rights0.06*0.06*0.05*
(0.03)(0.03)(0.03)
Antidirector Rights x Judicial Efficiency0.007**0.007**0.007**
(0.003)(0.003)(0.003)
Antidirector Rights x Corruption0.008**0.008**0.007**
(0.003)(0.003)(0.003)
Antidirector Rights x Rule of Law0.01**0.01**0.01**
(0.003)(0.003)(0.004)
Creditor Rights−0.0070.007−0.003
(0.03)(0.03)(0.04)
Accounting Standards−0.0008−0.002−0.005
(0.003)(0.01)(0.01)
Reserves0.0030.0020.0020.0010.0030.002
(0.002)(0.002)(0.002)(0.002)(0.002)(0.002)
R-Squared0.170.200.290.260.300.350.250.280.330.290.370.380.000.020.160.000.010.06
Adjusted R-Squared0.130.110.160.220.220.230.210.200.210.250.300.26−0.06−0.10−0.02−0.05−0.12−0.13
Observations202020202020202020202020191919191919
Notes: This table is reproduced from Johnson, Boone, Breach, and Friedman (2000). All results are for OLS regressions, with the dependent variable as the purchasing power of local currency vis-à-vis the U.S. dollar in January 1999 (end 1996=100). Definitions of indices: Judicial Efficiency (0-10, from 1980-83) is higher for more efficient legal systems from perspective of foreign business people; Corruption (0-10, from 1982-95) is higher if less bribery among government officials; Rule of Law (0-10, from 1982-95) is higher for stronger tradition of law and order; Corporate Governance (1-5, from 1998) is higher for better treatment of minority shareholders; Antidirector Rights (0-6, from 1996-97), is higher for better protection of minority shareholders; Creditor Rights (0-4, from 1996-97), is higher for better protection of creditors; Accounting Standards (0-90, from 1990) is higher for more disclosure in company reports. Total central bank reserves in billions of $U.S. at end 1996 is included as a control variable. Number of observations varies if data are missing. Standard errors are in brackets, and asterisks denote significance levels (**=5%, *=10%).
Notes: This table is reproduced from Johnson, Boone, Breach, and Friedman (2000). All results are for OLS regressions, with the dependent variable as the purchasing power of local currency vis-à-vis the U.S. dollar in January 1999 (end 1996=100). Definitions of indices: Judicial Efficiency (0-10, from 1980-83) is higher for more efficient legal systems from perspective of foreign business people; Corruption (0-10, from 1982-95) is higher if less bribery among government officials; Rule of Law (0-10, from 1982-95) is higher for stronger tradition of law and order; Corporate Governance (1-5, from 1998) is higher for better treatment of minority shareholders; Antidirector Rights (0-6, from 1996-97), is higher for better protection of minority shareholders; Creditor Rights (0-4, from 1996-97), is higher for better protection of creditors; Accounting Standards (0-90, from 1990) is higher for more disclosure in company reports. Total central bank reserves in billions of $U.S. at end 1996 is included as a control variable. Number of observations varies if data are missing. Standard errors are in brackets, and asterisks denote significance levels (**=5%, *=10%).
Table 8.Legal Institutions and Change in Stock Market Value
(i)(ii)(iii)(iv)(v)(vi)(vii)(viii)(ix)(x)(xi)(xii)(xiii)(xiv)(xv)(xvi)
Dependent variable is the stock market value at lowest point in 1998 with end 1996= 100
East Asia dummy−50.1**−63.6**−48.2**−55.0**−53.8**−56.0**−41.3**−53.4**
(13.6)(15.9)(11.3)(15.0)(10.5)(13.5)(8.4)(10.1)
Efficiency of Judiciary2.02.81.80.003
(4.2)(4.5)(3.2)(3.3)
Corruption7.6*9.4**5.9*4.9
(4.3)(4.2)(3.2)(3.5)
Rule of Law610.0**7.5*7.1**
(4.2)(4.3)(2.8)(3.3)
Corporate Governance12.9*15.0*6.31.7
(7.50)(7.60)(5.00)(5.3)
Macroeconomic Control Variable
Reserves−0.30.5−0.5*−0.2−0.60.07−0.3−0.4*
(0.4)(0.3)(0.3)(0.3)(0.3)(0.3)(0.3)(0.2)
Obs20202020232323232323232319191919
R-Squared0.010.040.450.520.130.240.540.560.090.260.610.610.150.220.660.73
Adjusted R-Squared−004−0.080.390.430.090.170.50.490.050.180.570.550.10.120.620.67
Countries missingChinaChinaChinaChinaCzechCzechCzechCzechCzechCzechCzechCzechChinaChinaChinaChina
CzechCzechCzechCzechRussiaRussiaRussiaRussiaRussiaRussiaRussiaRussiaColombiaColombiaColombiaColombia
HungaryHungaryHungaryHungaryGreeceGreeceGreeceGreece
PolandPolandPolandPolandPortugalPortugalPortugalPortugal
RussiaRussiaRussiaRussiaVenezuelaVenezuelaVenezuelaVenezuela
RussiaRussiaRussiaRussia
Notes: Reproduced from Johnson, Boone, Breach, and Friedman (2000). All results are for OLS regressions, with the dependent variable being the stock market value (IFC Investable Index) at its lowest point in 1998 (end 1996=100). Definitions: Judicial Efficiency is an index from 0 to 10, for the period 1980-83, with a higher score meaning a more efficient legal system; Corruption is an index from 0 to 10, for the period 1982-95, with a higher score meaning there is less bribery among government officials; Rule of Law is an index from 0 to 10, for the period 1982-95, with a higher score meaning a stronger tradition of law and order; Corporate Governance is an index from 1 to 5, for early 1998, with a higher score indicating better treatment for minority shareholders. The East Asia dummy equals one for China, Hong Kong, Indonesia, Korea, Malaysia, Philippines, Singapore, Thailand, and Taiwan. Standard errors are in parentheses. Asterisks denote significance levels (**=5%, *=10%).
Notes: Reproduced from Johnson, Boone, Breach, and Friedman (2000). All results are for OLS regressions, with the dependent variable being the stock market value (IFC Investable Index) at its lowest point in 1998 (end 1996=100). Definitions: Judicial Efficiency is an index from 0 to 10, for the period 1980-83, with a higher score meaning a more efficient legal system; Corruption is an index from 0 to 10, for the period 1982-95, with a higher score meaning there is less bribery among government officials; Rule of Law is an index from 0 to 10, for the period 1982-95, with a higher score meaning a stronger tradition of law and order; Corporate Governance is an index from 1 to 5, for early 1998, with a higher score indicating better treatment for minority shareholders. The East Asia dummy equals one for China, Hong Kong, Indonesia, Korea, Malaysia, Philippines, Singapore, Thailand, and Taiwan. Standard errors are in parentheses. Asterisks denote significance levels (**=5%, *=10%).

Looking at a broader set of countries, Pivovarsky and Thaicharoen (2001) find similar results. For all developing countries over the past 40 years, the size of the largest crisis (measured in terms of exchange rate depreciation) and the average size of the largest three crises are larger where institutions are weaker. They find that higher debt levels are correlated with a larger “worst crisis” in some specifications, but their data do not allow them to differentiate between corporate and government debt.

Firm-Level Outcomes

If corporate governance and debt matter for the severity of crises, we should expect to find correlations between performance, debt, and our governance measures at the firm level.

Mitton (2001) looks at five Asian countries most affected by the 1997-98 crisis, and finds those firms with larger inside ownership and less transparent accounting suffered larger falls in stock price. Table 9 summarizes some of his results. He also finds more diversified firms suffer a greater fall, particularly if they have more uneven investment opportunities (measured in terms of Tobin’s Q). This is consistent with the view that firms with weaker corporate governance faced a larger loss of investor confidence. It may also be the case that more diversified firms are less able to allocate investment properly due to internal politics, as suggested by Scharfstein and Stein (2000), and that these political problems become worse in a downturn. At the same time, Mitton (2001) finds that firms with more debt suffered larger falls in stock price.

Table 9.Firm-Level Corporate Governance and Stock Returns in Asian Crisis Countries
Pre-Crisis

July 1996 to

June 1997
Crisis Period

July 1997 to

August 1998
Later Period

Sept. 1998 to

August 1999
(i)(ii)(i)(ii)(i)(ii)
ADR−0.008

[−0.11]
−0.039

[−0.45]
0.116**

[2.46]
0.104**

[1.99]
−1.205*

[−1.74]
−0.904**

[−2.03]
Big Six auditor0.022

[0.39]
0.066

[0.98]
0.073*

[1.85]
0.045

[1.55]
−2.056

[−1.06]
−0.917

[−1.00]
Largest management blockholder %−0.113

[−0.25]
0.154

[0.92]
2.777

[1.14]
Largest nonmanagement blockholder%−0.161

[−0.73]
0.260***

[3.49]
0.191

[0.15]
Diversification*high variation0.054

[0.98]
0.048

[0.76]
−0.082***

[−3.21]
−0.069***

[−2.77]
0.475

[0.95]
0.373

[0.91]
Diversification*low variation−0.040

[−0.89]
−0.019

[−0.41]
−0.027

[−0.80]
−0.024

[−0.83]
2.818

[1.21]
0.695

[1.25]
Firm size−0.025

[−0.36]
−0.007

[−0.09]
0.026

[0.98]
0.061**

[2.31]
−0.714

[−1.03]
−0.180

[−0.51]
Debt ratio0.000

[0.06]
0.001

[0.74]
−0.0031***

[−5.67]
−0.0024***

[−4.14]
0.036

[1.54]
0.014

[1.39]
Number of observations356289384294370288
R-squared0.2130.1860.2860.3370.0690.078
Notes: Reproduced from Mitton (2001). The table reports coefficients of regressions of stock returns over the periods indicated on corporate governance variables for East Asian sample firms. Stock returns are measured in local currency terms adjusted for local price changes. Firms with missing data on ownership, total assets, or debt ratios are excluded from regressions that include these variables. “ADR” means the firm had an American depository receipt listed in the U.S. Big Six auditor means the firm’s auditor is one of the Big Six accounting firms. Blockholder percentages are classified according to the holder being associated with mangement. Diversified firms are classified based on having above- or below-median variation in investment opportunities. Also estimated but not reported are a constant term, country dummies for four of the five countries, and industry dummies for 11 of 12 industries, as defined in Campbell (1996). Heteroskedasticity-consistent t-statistics are in brackets, and significance levels are: * = 10%, ** = 5%, and *** = 1%.
Notes: Reproduced from Mitton (2001). The table reports coefficients of regressions of stock returns over the periods indicated on corporate governance variables for East Asian sample firms. Stock returns are measured in local currency terms adjusted for local price changes. Firms with missing data on ownership, total assets, or debt ratios are excluded from regressions that include these variables. “ADR” means the firm had an American depository receipt listed in the U.S. Big Six auditor means the firm’s auditor is one of the Big Six accounting firms. Blockholder percentages are classified according to the holder being associated with mangement. Diversified firms are classified based on having above- or below-median variation in investment opportunities. Also estimated but not reported are a constant term, country dummies for four of the five countries, and industry dummies for 11 of 12 industries, as defined in Campbell (1996). Heteroskedasticity-consistent t-statistics are in brackets, and significance levels are: * = 10%, ** = 5%, and *** = 1%.

Lemmon and Lins (2001) confirm Mitton’s (2001) finding that firm-level corporate governance was important during the crisis. Using a sample of firms from eight East Asian countries, they show that firm values (as measured by changes in Tobin’s Q and stock price) declined significantly more for firms that had a divergence between the cash flow rights and voting rights of the largest owner. They conclude that value declined more for firms in which the incentive for expropriation of minority shareholders was greater.

Obviously a great deal more work needs to be done before we understand exactly when and how firm-level corporate governance matters for performance. Lins and Servaes (1999) also find a discount for diversified firms in seven emerging markets. Claessens et al. (1999) find a diversification discount for East Asian firms and worse performance for conglomerates during the East Asian crisis. Following the approach of Mitton (2001), Nalbantoglu and Savasoglu (2000) present evidence that Turkish firms with weaker corporate governance suffered a larger fall in stock price during that country’s 1998 crisis.

Institutions, Growth, and Crises

Weaker corporate governance at the firm level is correlated with higher corporate indebtedness. Weaker corporate governance is associated with worse performance at the firm level during the Asian crisis, with potentially both a direct effect and an indirect effect through higher debt. The evidence also suggests that weaker investor protection institutions contributed to worse performance at the country level.

If corporate governance institutions in Asia are so weak, why have most of the economies begun to recover? Are their institutions really too weak to prevent the resumption of sustained growth? If institutions are weak, how did many Asian countries achieve high growth rates over the past 40 years? To answer these questions we have to distinguish between two kinds of insitutions.

The first type of institution offers effective protection against expropriation for entrepreneurs. Acemoglu, Johnson, and Robinson (2001) find that institutions protecting entrepreneurs in the past are an important determinant of institutions protecting entrepreneurs today. Colonies suitable for European emigration developed good institutions, while those with high mortality for Europeans developed more exploitative institutions. More generally, they show that the mortality rates faced by early European settlers are a valid instrument for current institutions, because mortality affected European settlements, settlements determined initial colonial institutions, and these institutions have had persistent effects. Using two-stage least squares estimation, Acemoglu, Johnson, and Robinson (2001) find that variation in institutions that protect entrepreneurs against expropriation account for three-quarters of the income per capita differences across countries (measuring from the first quartile to the third quartile of the cross-country income distribution).

The second type of institution protects investors against expropriation, perhaps by the government, but most importantly by the entrepreneur. It is this type of institution that is represented by parameter k in the model presented above.

The empirical results presented here suggest it is possible to grow with weak investor protection, as long as entrepreneurs feel protected. However, weak investor protection makes an economy vulnerable to collapse. This vulnerability arises from two causes. First, weak investor protection contributes to higher debt levels, as we have documented in this paper. Second, weak investor protection can directly undermine investor confidence, particularly when an economy has just been hit by a negative shock.

The implication is that basic institutions in Asian countries are consistent with economic growth. There is nothing fundamentally wrong with the protection of entrepreneurs in these countries. However, if investor protection continues to be relatively weak in an economy, this creates the potential for repeated severe crises and economic collapse.

Figure 3 provides represents alternative views of Asian growth in a stylized but hopefully useful fashion. Macroeconomic growth rates are on the horizontal axis and probabilities are on the vertical axis. The two graphs on the left-hand side summarize our interpretation of the conventional wisdom. Growth prospects before the crisis were considered good, with relatively small variance. The standard view is now that average growth will be lower and the variance on this forecast is high. In contrast, our view (on the right-hand side of Figure 3) is much more positive, in the sense that we place most of the probability distribution around rapid rates of growth. However, our reading of the evidence from the Asian crisis suggests that this distribution has two humps, so that there is a “fat” left tail to the distribution. It is this fat tail that could be eliminated by strengthening the institutions that protect investors.

Figure 3.Alternative Models of Crisis

Conclusion

There is growing evidence that institutions matter for macroeconomic outcomes in two ways. The fundamental need is for entrepreneurs to feel effectively protected against expropriation by the state. The evidence suggests that many Asian countries have done this relatively well on this dimension and strong “entrepreneur protection” institutions have helped produce impressive growth rates over the past 40 years.

Once entrepreneurs feel secure and are willing to reinvest their profits, a second related set of institutions also matter - the protection of outside investors against expropriation by entrepreneurs. It is possible to attract outside investors even when shareholder and creditor rights are weak, but when these rights are weaker debt will usually be higher. Higher debt levels make countries and companies vulnerable to collapse. Weak investor protection may also undermine confidence at critical moments, such as when the economy has just been hit by a negative shock. Taken together, these two effects of weak investor protection can turn a small shock into a big crisis.

A broader reevaluation of the macroeconomic implications of institutions is underway. For example, Blanchard (1999) argues that institutions in Western Europe were appropriate and well functioning, but could not handle the shocks they received in the 1970s and 1980s. In his view a functional set of institutions became dysfunctional due to a particular set of shocks. Our interpretation of the Asian crisis is similar. More generally, Blanchard (2000) suggests that macroeconomic behavior across countries may depend on the institutions that are in place. The Asian crisis suggests that the precise relationship between institutions and macroeconomic outcomes is of first order importance for economic policy.

Comments on Papers 8 and 9

Hasung Jang

These are both interesting papers. The paper by Professor Kim deals with the labor market, which experienced the most turbulent changes during the crisis. At the beginning of the crisis, many claimed that high wages and high financial costs were the cause of the weak competitiveness of Korean companies. However, little serious research has been done on the labor market since the crisis. So I was very happy to read the paper, which does a good job of documenting the many structural shifts that occurred during the crisis.

The paper looks at the labor market disaggregated by industry and by the size of companies. However, a comparison of chaebol and non-chaebol companies may reveal more interesting results. For listed companies from 1996-99, for example, the labor force in chaebol companies decreased by about 30 percent, while non-chaebol companies reduced their labor force by only about 5 percent. On the other hand, as Kim notes, wages increased during that period by over 20 percent among chaebol companies while in non-chaebol companies, wages increased only about 8.5 percent. Furthermore, even though the absolute wage has increased, the ratio of labor costs to total expenses has monotonically declined since 1990. This raises questions about whether labor market rigidity or wage rigidity has contributed to lower profitability, or has weakened competitiveness at the microeconomic level.

The paper makes very strong claims in the conclusion that so-called social programs have reduced job seeking on the part of the unemployed, but presents no evidence in support of this claim. The author highlights the basic livelihood protection program, and argues that the high allowance is an obstacle for job creation. Such strong claims should be supported by evidence. The basic livelihood protection program simply replaced the low income family protection program, and I do not believe there has been any increase in the number of the beneficiaries.

Similarly, we have seen a decrease in the participation rate, but I do not know if we can blame that on the public work programs. As the paper notes, more women lost jobs than did men. The author also points out that when we talk about labor market flexibility we always talk about layoff flexibility, not rehiring and re-entering flexibility; we all know that here in Korea, those who work for Daewoo will never get a job in Hyundai. For these reasons, the lower participation rates probably reflect issues related to gender and the class of workers that affected decisions about re-entering the market, rather than the impact of social programs.

Let me turn now to the Friedman, Johnson, and Mitton paper, which I liked very much. Until now, the corporate governance literature has focused on the quality of corporate governance, or the ownership structure, and how that relates to the performance of companies. This paper has taken the argument one step higher to relate corporate governance to macroeconomic issues. I was particularly happy to see the focus on country-and firm-level measures of corporate governance and h ow they affected macroeconomic outcomes such as the deterioration of exchange rates and the magnitude of economic crises.

One of the interesting results in the paper is that at the firm level, the concentration of ownership turns out to reduce the level of debt, as does non-management ownership. This looks controversial because recent research tends to find that highly concentrated ownership is associated with poor company performance. The explanation may be that owners have concentrated their ownership in the more profitable companies, whereas the less profitable companies are owned by the affiliated companies. It would have been useful to test for the effect of family control by including the discrepancies between cash flow rights and controlling rights.

Finally, it is disappointing that the variable representing disclosure, which is a very important measure for the quality of corporate governance, did not work out as expected, probably because the measure was rather simple. A more appropriate measure of disclosure would, in my opinion, work better than ownership and the other variables in explaining debt levels. There is a more fundamental problem with the data commonly used to analyze the relation between corporate governance and performance: companies with the most serious governance problems perform the worst and fail. This means that they are often not in our data sets. Thus, many papers that claim that corporate governance does not significantly affect company performance may have a serious bias in that the worst companies in terms of governance and performance are not included in the analysis.

The authors emphasize the differing importance of investor protection versus entrepreneur protection at different stages of economic development. The argument is that although entrepreneur protection was one of the main strategies for development policy, with the shift to a more developed market emphasis needs to be put on investor protection. Although I would have liked to see more evidence to support the argument, it is very interesting and important because corporate governance is usually neglected by economists who focus on development policy. The traditional view has been that growth and development needs to be led by the government, often with heavy government intervention. But as the market environment has changed, not only for developing countries but also for emerging market countries, policies to improve corporate governance can be a very effective development policy to induce private sector financial funds, rather than the international financial institutions, to provide financing. It will also be a better mechanism to monitor the implementation of those policies through improved transparency and more involvement of minority shareholders and other creditors. Enhancing the efficiency of development policies is thus another way that corporate governance can have macroeconomic implications.

Zia Qureshi

I will focus my comments on the Friedman, Johnson, and Mitton paper on corporate governance and debt, and attempt to relate some of its findings to the corporate governance reform agenda in Korea. I will also make some brief remarks about Professor Kim’s paper on the labor market.

The Friedman, Johnson, and Mitton paper is a valuable addition to a growing body of literature that shows that the institutions of corporate governance matter for macroeconomic outcomes. Some of the econometric results reported in the paper are not very strong, but taken as whole, and supported by corroborating evidence from other recent studies, including at the World Bank, the paper presents a convincing case. Weak corporate governance arrangements—low protection of outside investor rights, family/insider domination of corporate control and interlocking ownership, weak legal enforcement of contracts, lack of financial transparency—tend to produce a corporate financing structure that has a heavier dependence on debt, especially shorter-term debt. These factors also serve to undermine investor confidence, particularly at times of stress. Through both of these effects—higher corporate indebtedness and fragility of investor confidence—weak corporate governance arrangements can render an economy vulnerable to shocks and intensify their impact.

This is indeed what appears to have happened in East Asia. I agree with the authors that weak corporate governance arrangements in these countries provide an important part of the explanation of their vulnerability to the shocks of the late 1990s and of the severity of the economic collapse that followed. Weaknesses in corporate governance also raise agency costs and the cost of capital, and lower corporate value.

These messages were not lost on the governments of these countries. In their policy response to the crisis, improvements in corporate governance were included as an important element of the reform programs they adopted with the support of the international financial institutions (IFIs). Ironically, one criticism of the IMF- and World Bank-supported reform programs that was made soon after the crisis was that they were too broad in scope and encompassed elements such as legal and governance reforms, rather than focusing more narrowly on the immediate task of macroeconomic stabilization. Recent research, of which this paper is a good example, demonstrates that reform of the institutions of corporate governance is germane to the macroeconomic policy agenda, not an unrelated element that can be deferred until much later. Incidentally, these findings are also relevant to the current debate on narrowing the scope of conditionality in IFI-supported programs.

In Korea, the country whose post-crisis reform program I am most familiar with by virtue of m y direct involvement in World Bank support for the Korea reform program, substantial progress has been made since the crisis to improve the legal and institutional framework for corporate governance. The corporate governance component of the Korean Government’s reform program, which the World Bank actively supported, focused on four main areas: (i) strengthening minority shareholder rights and removing restrictions on the voting rights of institutional investors; (ii) clarifying the role and enhancing the independence of corporate boards of directors; (iii) improving financial transparency by raising standards for disclosure, accounting, and auditing toward international best practice; and (iv) strengthening creditors’ rights by improving the bankruptcy system.

Much progress has been made in all of these areas. Indeed, among the East Asian crisis countries, progress on reform has been the quickest in Korea. Yet, a sizable reform agenda remains unfinished. While a further deepening of reform is needed in each of the areas I just mentioned, three elements will be particularly important going forward:

  • First, enforcement needs to be strengthened. More explicit investor rights and other governance rules are useful only if there is a credible threat of sanctions under the legal and court system. This includes, for example, removal of obstacles to the prosecution of class action law suits by shareholders; enforcement of penalties and sanctions for breach of rules on disclosure, accounting, and auditing; and making the observance of the Code of Best Practice for Corporate Governance that was issued last year mandatory for companies listed on the Korea Stock Exchange.

  • Second, for companies with a large controlling shareholder, typically a founding family, which account for a substantial proportion of listed companies in Korea, a priority is to strengthen legal rules relating to conflict of interest and their enforcement. Compared with companies that are widely held, voting rights of minority shareholders are less effective in protecting their interests in companies with a large controlling shareholder. More important are legal safeguards in dealing with transactions where the actions of managers/directors conflict with their fiduciary “duty of loyalty” to act in the best interests of the company and its shareholders. Examples of such safeguards are board or shareholder approval requirements for transactions with related parties, dissenters’appraisal remedies for M&As, and pre-emptive subscription rights for shareholders.

  • Third, market discipline needs to be strengthened to permit market forces to exert greater pressure on firms to adhere to sound corporate governance practices. This includes, for example, further progress in removing obstacles to M & As and strengthening the insolvency system in order to make takeovers or insolvency more credible threats to firms; stronger exercise by the creditor financial institutions of their fiduciary responsibilities in monitoring corporate performance; and strengthening of complementary competition policies and their enforcement to limit non-arm’s length transactions, insider-dealings, and interlocking-ownership that create opportunities for abuse of outside investor rights.

Complementing these improvements in corporate governance, Korea’s corporate culture needs to shift in two desirable directions: separation of management from ownership; and maximization of shareholder value rather than corporate size as the main driver of corporate effort. Moreover, Korea needs to continue to move away from the old system of public-private relations characterized by government intervention and provision of implicit insurance (and the associated moral hazard).

Through costly corporate debt restructuring, Korea has made considerable progress in reducing high corporate debt-equity ratios that prevailed prior to the crisis toward levels that are more sustainable. However, unless there are fundamental improvements in the underlying corporate governance regime, this progress runs the risk of being only temporary. If corporate leverage begins to rise again, the systemic vulnerability to crisis would rise with it.

If firms are to reduce their dependence on debt financing, growth will slow if they are then to rely mainly on their own retained earnings, and resources will be less efficiently allocated. A longer-run strategy for sustainable growth must be based on developing equity markets. And if these are to develop, improved corporate governance, including strong protection for minority shareholders, is essential.

I have two additional comments on the Friedman, Johnson, and Mitton paper.

  • The paper explains well why weak corporate governance arrangements create incentives for firms to take on more debt than equity. However, what is less clear is why, on the supply side, the providers of debt finance continue to lend large amounts in the face of poor corporate governance practices and rising debt-equity ratios, and hence increasing corporate financial vulnerability. For instance, in Korea in the 1990s, lenders continued to lend heavily to the chaebol even in the face of declining corporate performance and mounting debt-equity ratios from levels that were already high. True, being the residual risk bearers, equity owners bear more risk as their claim is only to whatever is left after all prior claims, including those of lenders, have been paid. But lenders face risks too which rise as the financial vulnerability of the borrower rises with increasing indebtedness. I suspect that expectations of a public bailout in the event of loans turning sour provide part of the explanation for the behavior of lenders, both domestic and foreign.

  • Second, if this presumption is correct, shouldn’t such moral hazard play a more important role in the analysis of corporate indebtedness than it does in the paper? The paper looks briefly at the role of political allocation of lending, but the moral hazard issue I am referring to is broader than that. There is some support for the role of moral hazard in the data. Korea’s corporate governance indicators, though lower than those of the advanced economies, have been better than those of most of the other East Asian emerging market countries. Yet, Korea’s corporate sector indebtedness has been much higher. Perhaps the traditionally close public-private relations in Korea, and the “too big to fail” phenomenon from which Korea is now beginning to move away, provide part of the explanation.

I turn now to the paper by Professor Kim on Korea’s labor market. The paper provides an analysis of labor market changes in Korea since the crisis in impressive detail. I will make only two points:

First, the detailed microeconomic analysis of the labor market would have been more useful if it had been placed in a more macroeconomic and dynamic context. The paper finds that some of the changes in labor market policy, such as liberalization of layoffs and use of temporary labor, have contributed to a rise in wage inequality, and an increase in inferior and less secure jobs at the lower end of the skill range. In a macroeconomic context, these results do not imply that liberalizing layoffs and freeing the use of contract or temporary labor were the wrong policies to adopt. Of course, increased labor market flexibility was a correct policy response, and a necessary part of the reform package in order to allow the needed corporate and industrial restructuring to proceed. Under a proper assignment of policies, increased inequality should be dealt with through other policies, not maintenance of rigidities in the labor market. For example, K im refers to increased wage inequalities resulting from the rising wages of those with IT skills and declining wages of unskilled workers. The correct policy response to this, of course, is not to prevent such labor market adjustment by limiting labor market flexibility but, more positively, to expand education and broaden access to IT knowledge, as the Korean Government is n ow trying to do. It would be unwise to overburden labor market policy with a multiplicity of objectives.

Similarly, if the effects of the rise in unemployment and underemployment and the decline in job security are placed in a dynamic rather than static context, one could argue that these negative short-term effects of increased labor flexibility could be more than offset in the medium term by increased efficiency and growth that enhanced labor market flexibility would contribute to. In the interim, of course, it is important to mitigate the impact on the unemployed, especially the poor, by appropriately designed and targeted social safety net programs.

Second, and here I fully agree with Professor Kim, in designing the safety nets, such as defining the level of benefits and the actuarial structure of unemployment insurance, it is vitally important to ensure that these safety nets do not introduce new rigidities and distortions in the labor market that lead to a permanent rise in unemployment—a higher natural rate of unemployment. The unhappy European experience in this area provides important lessons for other countries.

References

    AcemogluDaronSimonJohnson and JamesRobinson2001The Colonial Origins of Comparative Development: An Empirical Investigation,American Economic Review911369-1401.

    AlbaPedroStijnClaessens and SimeonDjankov1998Thailand’s Corporate Financing and Governance Structures,World Bank working paper.

    BebchukLucian AyreKraakmanReinierTriantisGeorge2000Stock Pyramids, Cross-Ownership, and Dual Class Equity,in MorckR. (Ed.) Concentrated Corporate OwnershipThe University of Chicago Press pp. 295-315.

    BeckThorstenRossLevine and NormanLoayza2000Finance and the Sources of Growth.Journal of Financial Economics58261-300.

    BlanchardOlivier1999“European Unemployment: The Role of Shocks and Institutions,”manuscript MITJanuary.

    BlanchardOlivier2000“What Do We Know about Macroeconomics that Fisher and Wicksell Did Not?”Quarterly Journal of Economics.

    BonginiPaolaGiovanniFerri and HongjooHahm1999Corporate Bankruptcy in Korea: Only the Strong Survive?World Bank working paper.

    BurkartMikeDenisGromb and FaustoPanunzi1998Why Higher Takeover Premia Protect Minority Shareholders,Journal of Political Economy106172-204.

    ChoiBuhmsoo1999“Lessons of Crisis Management from the Korean Experience,”paper presented at An International Forum on The Korean Financial CrisisHarvard UniversityNovember18-19.

    ClaessensStijnSimeonDjankov and LarryLang1998Corporate Growth, Financing and Risks in the Decade Before East Asia’s Financial Crisis,World Bank working paper.

    ClaessensStijnSimeonDjankovJoseph P.H.Fan and Larry H. P.Lang1999The Benefits and Costs of Internal Markets, Evidence from Asia’s Financial Crisis,World Bank working paper.

    CoffeeJohn C.Jr.1999The Future as History: The Prospects for Global Convergence in Corporate Governance and its Implications,Northwestern University Law Review93631-707.

    Demirguc-KuntAsli and VojislavMaksimovic1999Institutions, Financial Markets, and Firms Debt Maturity,Journal of Financial Economics54295-336.

    DyeR.1993Auditing Standards, Legal Liability, and Auditing Wealth,Journal of Political Economy101887-914.

    EasterbrookFrankH.1997International Corporate Differences: Markets or Law?Journal of Applied Corporate Finance Volume 9Number 4Winter.

    FriedmanEric and SimonJohnson2000“External Finance with Weak Investor Protection,”manuscript MITMarch.

    GlaeserEdwardSimonJohnson and AndreiShleifer2001Coase versus the Coasians,Quarterly Journal of Economics116853-900.

    HoshiTakeoAnilKashyap and DavidScharfstein1991Corporate Structure, Liquidity, and Investment: Evidence from Japanese Industrial Groups,Quarterly Journal of Economics10633-60.

    JensenMichaelC.1986Agency Costs of Free Cash Flow, Corporate Finance and Takeovers,American Economic Review76323-29.

    JensenMichaelC. and William H.Meckling1976Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure,Journal of Financial Economics3305-360.

    JohnsonSimonPeterBooneAlasdairBreach and EricFriedman2000Corporate Governance in the Asian Financial Crisis, 1997-98.Journal of Financial Economics58141-186.

    JohnsonSimonRafael LaPortaFlorencioLopez-de-Silanes and AndreiShleifer2000“Tunnelling,”American Economic Review (papers and proceedings) May.

    JohnsonSimon and ToddMitton2001“Who Gains from Capital Controls? Evidence from Malaysia,”Journal of Financial Economicsforthcoming.

    KhannaTarun and KrishnaPalepu2000Is Group Affiliation Profitable in Emerging Markets? An Analysis of Diversified Indian Business Groups,Journal of Finance55867-891.

    KimByungmo and InmooLee2000“Do Agency Problems Explain the Post Asian Financial Crisis Performance of Korean Companies? Chaebol vs. Non-Chaebol firms,”manuscriptKorea University.

    KimSe-Jik2000Bailout and Conglomeration,Revised version of IMF Working Paper WP/99/108August.

    KimSe-Jik and IvailoIzvorski2000“Aggregate Shock, Capital Market Opening, Optimal Bailout and Structural Reform,”manuscript IMF KIEP and Institute of International FinanceDecember.

    KimSe-Jik and Mark R.Stone1999Corporate Leverage, Bankruptcy, and Output Adjustment in Post-Crisis East Asia,IMF Working paper WP/99/143October.

    LaPortaRafael and FlorencioLopez-de-Silanes1998Capital Markets and Legal Institutions,” in ShadidBurki and GuillermoPerryeditorsBeyond the Washington Consensus: Institutions MatterWorld BankDecember.

    LaPortaRafaelFlorencioLopez-de-Silanes and AndreiShleifer1999Corporate Ownership Around the World,Journal of Finance54471-517.

    LaPortaRafaelFlorencioLopez-de-SilanesAndreiShleifer and RobertVishny1997aLegal Determinants of External Finance,Journal of Finance521131-1150.

    LaPortaRafaelFlorencioLopez-de-SilanesAndreiShleifer and RobertVishny1997b“Shareholders’ Rights: Appendix,”unpublished appendix to LLSV1998.

    LaPortaRafaelFlorencioLopez-de-SilanesAndreiShleifer and Robert W.Vishny1998Law and Finance,Journal of Political Economy1061113-55.

    LaPortaRafaelFlorencioLopez-de-SilanesAndreiShleifer and Robert W.Vishny1999aThe Quality of Government,Journal of Law Economics and Organization15222-279.

    LaPortaRafaelFlorencioLopez-de-SilanesAndreiShleifer and Robert W.Vishny1999b“investor Protection and Corporate Valuation,”manuscriptHarvard and University of ChicagoJuly.

    LaPortaRafaelFlorencioLopez-de-SilanesAndreiShleifer and Robert W.Vishny1999c“Government Ownership of Commercial Banks,”manuscriptHarvard and University of ChicagoNovember.

    LaPortaRafaelFlorencioLopez-de-SilanesAndreiShleifer and Robert W.Vishny2000a“Investor Protection and Corporate Governance,”Journal of Financial Economicsforthcoming.

    LaPortaRafaelFlorencioLopez-de-SilanesAndreiShleifer and Robert W.Vishny2000b“Agency Problems and Dividend Policies around the World,”Journal of Financeforthcoming.

    LeeJong-WhaYoung SooLee and Byung-SunLee1999The Determination of Corporate Debt in Korea,Harvard Center for International Development working paper.

    LemmonMichael and KarlLins2001“Ownership Structure, Corporate Governance, and Firm Value: Evidence from the East Asian Financial Crisis,”manuscriptUniversity of Utah.

    LinsKarl2000“Equity Ownership and Firm Value in Emerging Markets,”manuscriptUniversity of Utah.

    LinsKarl and ServaesHenri2000“Is Corporate Diversification Beneficial in Emerging Markets?”manuscriptUniversity of Utah.

    LinsKarlDeonStrickland and MarcZenner2000“Do Non-US Firms Issue Stock on US Equity Markets to Relax Capital Constraints?”manuscriptUniversity of Utah and Ohio State UniversityOctober.

    MichaelyRoni and W.Shaw1995Does the Choice of Auditor Convey Quality in an Initial Public Offering?Financial Management24:415-30.

    MittonTodd2001“A Cross-Firm Analysis of the Impact of Corporate Governance on the East Asian Financial Crisis,”Journal of Financial Economicsforthcoming.

    MorckRandallDavidStrangeland and BernardYeung2000“Inherited Wealth, Corporate Control and Economic Growth: the Canadian Disease?” in MorckR. (Ed.) Concentrated Corporate OwnershipThe University of Chicago Press.

    MyersStewartC.1977Determinants of Corporate Borrowing,Journal of Financial Economics5147-175.

    NalbantogluOsman and SerkanSavasoglu2000“Impact of Corporate Governance and Foreign Trading on Firm Returns During Crises: the Case of Turkey,”manuscriptHarvard UniversityApril.

    PivovarskyAlexander2001“Does Legal Origin Matter for Financial Development?”Chapter 2 in“Essays on Institutions and Finance,”Ph.D. ThesisHarvard UniversityMay.

    PivovarskyAlexander and YunyongThaicharoen2001“Institutions and the Severity of Currency Crises,”Chapter 3 in“Essays on Institutions and Finance,”Ph.D. Thesis (Alexander Pivovarsky) Harvard UniversityMay.

    RahmanM.Zubaidur1998The Role of Accounting in the East Asian Financial Crisis: Lessons Learned?Transnational Corporations71-51.

    RajanRaghuram and LuigiZingales1998aFinancial Dependence and Growth,American Economic Review88559-586.

    RajanRaghuramG. and LuigiZingales1998bWhich Capitalism? Lessons from the East Asian Crisis,Journal of Applied Corporate Finance1140-48.

    RajanRaghuram and LuigiZingales2001“The Great Reversals: the Politics of Financial Development in the 20th Century,”revised manuscriptUniversity of Chicago.

    ReedB.M.Trombley and D.Dhaliwal2000Demand for Audit Quality: the Case of Laventhol and Horwath’s Auditees,Journal of Accounting Auditing and Finance15: 2183-198.

    ReeseW. Jr. and MichaelWeisbach2001Protection of Minority Shareholder Interests, Cross-Listings in the United States, and Subsequent Equity Offerings,NBER working paper #8164.

    ScharfsteinDavidS. and Jeremy C.Stein2000The Dark Side of Internal Capital Markets: Divisional Rent-Seeking and Inefficient Investment,Journal of Finance552537-2564.

    ShleiferAndrei and RobertVishny1997A Survey of Corporate Governance,The Journal of Finance52737-783.

    SteinJeremy1997Internal Capital Markets and the Competition for Corporate Resources,Journal of Finance52111-133.

    StulzR.1999“Globalization of Equity Markets and the Cost of Capital,”ManuscriptOhio State University.

    TitmanSheridan and B.Trueman1986Information Quality and the Valuation of New Issues,Journal of Accounting and Economics8159-172.

    TitmanSheridan and R.Wessels1988The Determinants of Capital Structure Choice,Journal of Finance431-19.

    WolfenzonDaniel1998“A Theory of Pyramidal Ownership,”manuscriptHarvard University.

    WurglerJeffrey2000Financial Markets and the Allocation of Capital,Journal of Financial Economics58187-214.

Simon Johnson thanks the MIT Entrepreneurs hip Center for support. For helpful comments we thank Damn Acemoglu, Stewart Myers, and Andrei Shleifer.

We use the terms investor protection and corporate governance interchangeably in this paper. The more important distinction is between country-level institutions and investor protection/corporate governance institutions that are chosen by firms.

This section draws heavily on Friedman and Johnson (2000).

No tunneling occurs if α R is sufficiently high. Given that α is often high in emerging markets, a reasonable economic boom may make it optimal for the group not to tunnel anything.

We model the case where the firm has debt, but it could be equity with some debt-like characteristics (e.g., the firm is punished if it has below market “expectations” for earnings.) There just needs to be some incentive to smooth earnings.

We do not deal with the possibility that the debt is renegotiated. As long as both groups and investors lose something when the firm “goes bankrupt,” the intuition behind our results holds.

    Other Resources Citing This Publication