After World War II, capital markets in different countries were almost completely segmented from each other. Individual investors generally could not buy securities in other countries because they could not acquire the correct currency. Since then, dramatic changes have occurred. Currencies typically are convertible and most major obstacles to capital flows among developed countries have been removed. This evolution has allowed equity investors to invest in most countries in the world.

Unfortunately, throughout history, periods with limited barriers to international investment were followed by periods with stronger restrictions. Since the devaluation of the Thai baht last year, many countries have faced economic difficulties; many economists and policymakers have argued that the so-called "Asian flu" has spread too far and too decisively. Led by fears of contagion and encouraged by some prominent economists, some countries, for example Malaysia, have taken measures recently to restrict capital flows. Though fixed exchange rates can breed speculative attacks and contagion, we should not forget that countries benefit from open equity markets. There is little evidence that the presence of foreign equity investors will lead to irrational movements in equity markets.

In a recent paper,⁽¹⁾ I discuss the costs and benefits of opening equity markets to foreign investors. With closed equity markets, investors within a country have to bear all the risks of that country. Therefore, they charge a higher risk premium to bear risk, which translates into a higher cost of capital. The cost of capital determines which investments are worthwhile. An increase in the risk premium means that riskier investments with a greater expected return are replaced by safer investments with a lower expected return. Consequently, a country whose investors cannot diversify risks internationally has an economy that invests in safer projects with lower risk. This practice hampers economic growth. When the investors in a country become able to share risks with foreign investors, the cost of capital falls, and the economy invests in riskier projects that contribute more to growth because the cost of bearing risk is lower.

As countries have progressively liberalized, we have seen the emergence of a world capital market, which for a number of years included most developed countries. Eventually, it was joined by emerging markets. The theory predicts that, as the world capital market grows with the addition of new countries, the risk premium for bearing risk falls. As a country joins this world capital market, its cost of capital also falls. However, the evidence shows that the decrease in a country's cost of capital when it opens its capital market has been smaller than predicted. Recent papers by Geert Bekaert and Campbell R. Harvey⁽²⁾2 and P. B. Henry⁽³⁾ indicate that the fall in the cost of capital when an emerging market opens its capital market might be on the order of two hundred basis points when theory predicts a much larger drop.

The Home Bias and Portfolios of Foreign Investors

Part of the reason that the cost of capital does not fall as much as predicted is that when markets open up, foreign investors do not acquire as much equity as expected. In general, investors have a strong preference for their own country's equity. This phenomenon is called the home bias in equity holdings. Though the home bias has been studied for more than 20 years, it has recently become more of a puzzle, because some of the reasons for its existence seem to have disappeared. For instance, earlier research emphasized differences in transaction costs and taxes as an explanation for the home bias,⁽⁴⁾ but these differences have diminished over time. As Linda L. Tesar and Ingrid M. Werner point out,⁽⁵⁾ in some cases foreign investors trade more than domestic investors; thus higher transaction costs cannot explain why their holdings of foreign equity are so limited.

Because of the home bias, investors do not take as much advantage of international diversification as would be expected. Jun-Koo Kang and I⁽⁶⁾ examine the apparent home bias in holdings of equity by foreign investors in Japan. For that study, we had data on holdings of equity by foreign investors for each firm on the Tokyo Stock Exchange from 1975 to 1991. Strikingly, despite all the liberalization of capital flows, foreign holdings of Japanese securities were quite small over the entire sample period. In 1975, foreign investors held 4.64 percent of the market capitalization of the Tokyo Stock Exchange. In 1991, they held 5.59 percent. Their holdings reached a peak of 11.31 percent in 1984. Expressed as a fraction of the world equity market capitalization, the holdings of Japanese stocks by foreign investors were always small relative to the importance of the Japanese equity market. Even in 1991, the Japanese equity market was more than 30 percent of the capitalization of the Morgan Stanley Capital International World Index.

Why have investors been so slow to take advantage of the benefits of international diversification? Our paper on the behavior of foreign investors in Japan offers some clues. We find that foreign investors have an extremely strong preference for holding shares of large firms. On average during our sample period, foreign investors held 7.12 percent of a firm in the top quintile of Japanese firms in equity capitalization, but only 1.4 percent of a firm in the bottom quintile. Hyuk Choe, Bong-Chan Kho, and I⁽⁷⁾ demonstrate that this large firm bias is not restricted to Japan. We show a similar bias for foreign investors in Korea in 1997. Kang and I⁽⁸⁾ also show that, for a given firm size, foreign investors invest more in the more liquid stocks. This evidence is consistent with a model where foreign investors are better informed, or less at an informational disadvantage relative to domestic investors, for large firms. Choe, Kho, and I⁽⁹⁾ show that institutional investors account for almost all of the holdings of equity by foreign investors in Korea. Institutional investors generally exhibit a bias toward large stocks, and institutional investors who invest abroad are no different.

Globalization and the Relationships Among Equity Markets

As barriers to international investment fall, equity markets everywhere are affected by changes in the risk premium on equity. Stock prices can fall because investors expect future economic activity to become less favorable, because of an increase in real interest rates, or because investors require a higher compensation for risk. With free markets, the fact that an increase in the risk premium that investors require on equity causes stock prices throughout the world to fall is a sign of efficiency rather than a reason for concern. In simple models, the risk premium on the world market portfolio depends on the volatility of the return on that portfolio. An unexpected increase in volatility increases the risk premium and causes stock prices to decrease. Everything else being equal, an increase in the volatility of a foreign market increases the volatility of the world market portfolio. Consequently, an increase in volatility abroad causes stock prices to drop in the United States.

K. C. Chan, G. Andrew Karolyi, and I⁽¹⁰⁾ examine this relationship between the volatility of a foreign market and the risk premium on U.S. stocks. We find that the risk premium on U.S. stocks, for example, depends on the volatility of the Japanese market. An increase in the volatility of Japanese stock returns leads to an increase in the risk premium on U.S. stocks and a fall in the price of these stocks. Opening an equity market to foreign investors increases the dependence of that market on events outside it. If this were not the case, there would be no benefit from opening the market. However, factors that affect the risk premium on stocks now affect the stocks in the newly opened market. As a result, the return on equity in that market is likely to be more correlated than before with the return on equity in the rest of the world. I have found some evidence of such an increase.⁽¹¹⁾ Further, when Karolyi and I⁽¹²⁾ examine the determinants of the correlation between U.S. stocks and Japanese stocks, we show that the U.S. market and the Japanese market move more closely together when volatility in either market is high.

Though the evidence just described relates to the rational links between markets that result from the removal of barriers to international investment, there has also been much concern about the existence of links that are attributable to panic and confusion of foreign investors. When I examine the evidence available before the most recent events, though, it does not seem that there is irrational contagion among equity markets.⁽¹³⁾

Do Countries Still Matter?

Globalization makes markets move together more than they did before. This has led some to believe that the focus of equity analysis should be on industries across frontiers rather than on countries. Some investment management firms are now reorganizing, so that instead of having analysts whose work stops at the border of a country, they have analysts who look at an industry across countries. This raises the question of whether globalization has gone so far that we can ignore countries. Both Karolyi and I⁽¹⁴⁾ and John M. Griffin and I⁽¹⁵⁾ find that we cannot. Industry effects are of surprisingly little importance in explaining the correlation of stock returns across countries, we find. Griffin and I⁽¹⁶⁾ also show that exchange rates explain very little of the difference in performance across industries. Even though there have been many analyses of how yen depreciation makes American car producers more competitive at the expense of Japanese car producers, in fact the impact of an unexpected change in the yen-dollar exchange rate on the equity of American car producers is economically trivial.

Are Foreign Equity Investors Dangerous?

Chan, Karolyi, and I⁽¹⁷⁾ demonstrate the existence of a completely rational relationship between the volatility of foreign markets and the equity premium on U.S. stocks. But recently, many observers and policymakers have been concerned about the ability of foreign investors to destabilize markets. Choe, Kho, and I⁽¹⁸⁾ investigate the role of foreign equity investors in Korea using a database that includes all trades made on the Korea Stock Exchange in 1997. This database makes it possible to identify trades made by foreign investors. We first explore the dynamics of foreign holdings of shares in Korea, because policymakers are concerned that foreign equity investors buy when markets are going up and sell when markets are going down. Such investment strategies are called positive feedback strategies. They can be destabilizing because investors can lead prices away from fundamentals. This is especially the case if different investors pursue similar strategies, so that they end up forming a herd. We find strong evidence that foreign investors engaged in positive feedback strategies in Korea. In particular, foreign investors were more likely to buy when Korea's market was up the previous day and to sell when it was down. We also find that foreign investors tended to act as a group, in that they were likely to buy together and sell together. Our evidence, therefore, supports those who are concerned that foreign investors engage in positive feedback trading and herding.

Next we ask whether the practices of foreign investors are destabilizing. We look in great detail at the last three months of 1997 and show that the practices of foreign investors that have caused concern diminished during that time, when Korea's economic crisis developed. In particular, we find that positive feedback trading dropped dramatically and perhaps disappeared during the last three months of 1997. Further, herding fell significantly during that period. Based on our evidence, it seems implausible that short-term destabilizing positive feedback trading on the part of foreign equity investors played a role in the Korean crisis. Perhaps the most telling evidence that positive feedback trading cannot have been important is that the fraction of the total capitalization of Korea's stock market held by foreign investors was higher at the end of 1997 than at the beginning.

Further investigating the impact of trading by foreign investors, we consider whether foreign investors might have started runs on stocks. In other words, we ask if stock prices start to increase following purchases by foreign investors and start to fall following sales by foreign investors. We first look at five-minute intervals: as one would expect, if foreign investors make a large purchase of a stock during a five-minute interval, the stock price increases. This is true both before and during the crisis period. Once the purchase is made, however, the market adjusts within five minutes. Therefore, there is no evidence that purchases by foreign investors have a destabilizing effect. The market also adjusts quickly to sales and the small negative impact of a large sale (on the order of 1 percent) gets reversed in part. When we examine daily returns, we again find that the actions of foreign investors do not lead to runs. Thus, no case can be made that foreign equity investors were disruptive.

What About Banks?

If the actions of foreign investors seem unlikely to explain dramatic falls in stock markets, we must look elsewhere to understand such episodes. The drop in the Japanese stock market in the early 1990s offers fertile ground for researchers. Kang and I⁽¹⁹⁾ try to understand the contribution of the difficulties in the banking sector to the collapse of equity prices. We show that those Japanese firms that relied more heavily on banks for their financing at the end of the 1980s experienced a sharper loss of equity value during the early 1990s.

We then explore possible explanations for this result. We find that the impact of bank reliance on equity performance cannot be explained by other firm characteristics. Rather, the explanation for the poor performance of firms that relied more on banks seems to be that these firms had to cut back their investment relative to firms that did not rely on banks as much. Everything else being equal, the firms that relied more on bank financing at the end of the 1980s invested less at the beginning of the 1990s. Shocks to banks' ability to finance investment thus have a significant economic impact on those firms that depend on them. We further confirm the importance of bank health on bank-dependent firms by examining the performance of Japanese firms on days when important announcements about the Basle Accord on capital requirements for banks were made. On days when Japanese bank stocks fell because of announcements about the Basle Accord, the firms that relied more on bank debt performed poorly, we find.

Our evidence does not imply that firms should not rely on banks. Rather, it shows that finance provided by banks is fundamentally different from finance provided by equity investors. If foreign investors want to reallocate their holdings away from a country, they have to worry about the price impact of their trades. If a bank has to reduce the exposure of its loan portfolio to firms in one country, it benefits from being the first to do so. This is because the borrower is likely to run out of liquid assets to pay back loans and the firm that had obtained bank finance no longer has it once the bank has called the loan. As a result, when too many firms in a country rely excessively on bank debt, it does not take much of an economic shock to force them to shrink investment and to put some of them into default.

Endnotes

^1. R. M. Stulz, "International Portfolio Flows and Security Markets," Dice Center Working Paper, Ohio State University, 1997, pp. 97-112.

^2. G. Bekaert and C. R. Harvey, "Foreign Speculators and Emerging Equity Markets," unpublished working paper, Duke University, 1998.

^3. P. B. Henry, "Stock Market Liberalization, Economic Reform, and Emerging Market Equity Prices," unpublished working paper, Graduate School of Business, Stanford University, 1998.

^4. R. M. Stulz, "On the Effects of Barriers to International Investment," Journal of Finance, 36, (1981), pp. 459-68.

^5. L. L. Tesar and I. M. Werner, "Home Bias and High Turnover," Journal of International Money and Finance, 14, (1995), pp. 467-93.

^6. J. Kang and R. M. Stulz, "Why Is There a Home Bias? An Analysis of Foreign Portfolio Equity Ownership in Japan," Journal of Financial Economics, 46, (1997), pp. 3-28.

^7. H. Choe, B. Kho, and R. M. Stulz, "Do Foreign Investors Destabilize Stock Markets? The Korean Experience in 1997," NBER Working Paper No. 6661, July 1998.

^8. J. Kang and R. M. Stulz, "Why Is There a Home Bias? An Analysis of Foreign Portfolio Equity Ownership in Japan."

^9. H. Choe, B. Kho, and R. M. Stulz, "Do Foreign Investors Destabilize Stock Markets? The Korean Experience in 1997."

^10. K. C. Chan, G. A. Karolyi, and R. M. Stulz, "Global Financial Markets and the Risk Premium on U.S. Equity," Journal of Financial Economics, 32, (1992), pp. 137-68.

^11. R. M. Stulz, "International Portfolio Flows and Security Markets."

^12. G. A. Karolyi and R. M. Stulz, "Why Do Markets Move Together? An Investigation of U.S.-Japan Stock Return Comovements," Journal of Finance, 51, (1996), pp. 951-86.

^13. R. M. Stulz, "International Portfolio Flows and Security Markets."

^14. A. Karolyi and R. M. Stulz, "Why Do Markets Move Together? An Investigation of U.S.-Japan Stock Return Comovements."

^15. J. M. Griffin and R. M. Stulz, "International Competition and Exchange Rate Shocks: A Cross-Country Industry Analysis of Stock Returns," NBER Working Paper No. 6243, October 1997.

^16. Ibid.

^17. K. C. Chan, A. Karolyi, and R. M. Stulz, "Global Financial Markets and the Risk Premium on U.S. Equity."

^18. H. Choe, B. Kho, and R. M. Stulz, "Do Foreign Investors Destabilize Stock Markets? The Korean Experience in 1997."

^19. J. Kang and R. M. Stulz, "Is Bank-Centered Corporate Governance Worth It? A Cross-Sectional Analysis of the Performance of Japanese Firms During the Asset Price Deflation," NBER Working Paper No. 6328, October 1997.