An understanding of the variation of stock price exposures in different countries helps to determine equilibrium compensation and evaluate portfolio performance and assists making decision on the allocation of international portfolio. However, there are two unanswered questions regarding international asset pricing. How do different developmental stages of countries affect the risk of individual stock price? Do the risks of stock prices in the different regions demonstrate variation? In this study, we first explain the relationship between international integration of local markets and individual stock price volatility. We then report the result of an empirical test of relative magnitude, measured by non-parametric Mann-Whitney statistics, of price exposures in the nations of different developmental stages and areas by using firm-level data. We found that the stocks in emerging markets are significantly riskier than in rich countries. On the other hand, in contrast to developed countries, the volatilities in most developing nations, on the whole, are steadily decreasing. We also found similar levels of risk in equity values in the countries of the same area.
One explanation to the cross-country variation of asset price risks is the degree of integration of a domestic market with international financial markets. Bekaerk and Harvey (1995) and Bae, Bailey, and Mao (2003) suggest the influence of liberalization of financial market to equilibrium price of domestic assets differs from nation to nation. Following the opening of financial market, the exchange of capital and information flows of local market with foreign markets may trigger escalation of volatility of security prices. On the other hand, as the degree of international financial integration and market efficiency increases, any investment should only be compensated by the amount of global systematic risk. One may also expect the shrinkage of the idiosyncratic risks of asset prices. However, the empirical results suggest the mixed effect of global integration of financial markets on asset risk. The empirical findings by Bekaert and Harvey (1997, 2003) and De Santis and ?mrohoro?lu (1997) support the argument that there is no straight finance theory about the change of volatility after market liberalization. Kim and Singal (2000) suggest that market openings of developing nations decrease the risk in the long-run by increasing market efficiency.
The variation of legal tradition, major financing sources, cultural background, and natural resources can also help to explain the disparity of stock risks across nations. These factors affect the protection of property right, awareness of uncertainty, and attitude toward wealth and its distribution, all of which are relevant to risk aversion. These are relevant to the willingness of investment as well as the financial structure of corporations. Furthermore, the developmental stage of economy is related to the maturity of a financial market and integration with the economies of the rest of the world. Although it is hard to quantify the qualitative elements in a pricing model, previous research has suggested that they have an impact on asset price volatility.
Previous studies indicate that the distribution of systematic risk can not be modeled. Collins, Ledolter and Rayburn (1987) demonstrate the randomness of beta of stock yield. They also find that the systematic risk does not show autocorrelation, and that it relates with firm size. On the other hand, Chatterjee and Lubatkin (1990), DeJong and Collins (1985), and Denis and Kadlec (1994) argue that the variation of beta of stock comes from the change of interest rate, of dividend policy of a company such as stock split and dividend payout, and of investment activities such as merger and acquisitions. Pettengill, Sundaram, and Mathur (1995) found that the systematic risk is adjusted by previous abnormal returns as measured by the Sharpe-Litner-Black asset pricing model and that there is a positive relationship between beta and return.
Empirical testing on the significance of cross-country stock risks helps us clarify the mixed theoretical effects of exposure in emerging markets. Since the assumption of a Gaussian distribution of financial parameters rarely holds, we employ distribution-free Mann-Whitney test to avoid problems with the departures from normality. In addition, to avoid the mixture of returns of stock returns brought by market-level data suggested by Carrieri, Errunza, and Sarkissian (2004), data of 4,916 stocks from 22 developed countries and 15 developing countries were collected. Our empirical findings indicate that the stock prices in emerging markets are comparatively riskier than those in developed countries, as measured by both conditional volatility and unconditional global beta. In addition, our results also support the geographical variation of stock risk. Specifically, the equity values in East Asia, Southeast Asia, South Europe, and Latin America are more volatile than the rest of the world. Although there are some exceptions in the country-level tests, relative size of stock price risks in most countries are similar to the ones of their developmental stage as well as area. In addition, the analysis of time-series of volatility suggests that the stocks of high price exposures tend to be less volatile, and the conditional volatilities of less risky stocks tended to be steadily enlarging. This finding can be viewed as evidence of the enhancement of integration of international financial markets.
Download
Variation of Stock Return Volatility: An International Comparison
