Feixue Xie

Ownership Structure and Firm Value in China's Privatized Firms: 1991-2001

This paper investigates the relation between ownership structure and firm value across a sample of 5,284 firm years of Chinas partially privatized former state–owned enterprises (SOE) from 1991–2001. We find that state and institutional shares are significantly negatively related to Tobins Q, and that significant convex relations exist between Q and state shares, as well as between Q and institutional shares. We also find that foreign ownership is significantly positively related to Tobins Q. We test for potential endogeneity of ownership, and find that Q and state/foreign ownership are not jointly determined. We also test for time-series, industry, and geo-economic location effects, and find our results to be robust.

Market Development and the Asset Growth Effect: International Evidence

A number of studies of U.S. stock returns document what is referred to as the investment or asset growth effect. Specifically, firms that increase investment or total assets subsequently earn lower risk-adjusted returns. This study finds substantial cross-country differences in the asset growth effect. In particular, the asset growth effect is stronger in countries with more developed financial markets, but it does not seem to be associated with corporate governance or the costs of trading. Overall, the evidence is consistent with a q-theory where financial market development captures either managers’ willingness or ability to align investment expenditures to the cost of capital, but it is inconsistent with the hypothesis that the asset growth effect is due to bad governance and overinvestment.

Accruals, Capital Investments, and Stock Returns

The evidence from this study shows that the “accruals anomaly” and the “capital investment anomaly” are distinct, even though capital investments and accruals may be related in a certain way. The results also indicate that, after adjustment for the Fama–French three risk factors, investors earn substantially higher returns by using a strategy that exploits both anomalies at the same time than by exploiting either anomaly alone. Using current accruals as the measure of accruals produced similar results to using total accruals, and the results are robust to various measures of return. The evidence suggests that managers in companies ranked highest in both accruals and capital investments may be overly optimistic about future demand for their products. This study examined whether the “accruals anomaly” and the “capital investment anomaly” capture the same underlying force. Both anomalies may be linked by way of managers’ overly optimistic expectations about future product demand that lead to a buildup of production capacity and inventory. We further examined whether one anomaly provides new information not provided by the other even if both anomalies are driven by the same expectation of future demand. If both signals are incrementally valuable, an investor can obtain additional abnormal returns by basing a strategy on combining the anomalies, so this possibility should be of great interest to practitioners and investors. Using data from the CRSP/Compustat Merged Database for the period 1972–2005, we found that the two anomalies are distinct from each other and that investors can earn substantially higher returns by using a strategy that exploits both anomalies at the same time. Specifically, a trading strategy of buying the shares of companies in the lowest quintile formed on the basis of total accruals and the lowest quintile formed on the basis of capital investments and simultaneously shorting the shares of companies in the highest total accruals quintile and the highest capital investment quintile would have generated a risk-adjusted return of 12 percent per year, on average, in the sample period. The return to such a strategy is substantially larger than the returns from exploiting either the investment anomaly–based strategy (7.66 percent) or the total accruals–based strategy (6.74 percent) alone. Using current accruals as an alternative measure of accruals produced similar results. Moreover, we showed that our results are robust to the measure of returns. An important implication of this study is that portfolio managers and investors should exploit the capital investment anomaly and the accruals anomaly at the same time to enhance their trading profitability.

Capital Investments and Stock Returns in Japan

The negative relation between capital investments and subsequent stock returns, found in the United States, is not observed in Japan, which is inconsistent with the risk-based explanation. More specifically, we find no significant relation between capital expenditures (CE) and subsequent stock returns for either the entire sample or for keiretsu firms. However, in the pre-1990 subperiod, there is a positive relation between increased CE and subsequent risk-adjusted returns among independent firms, especially for those firms that have high cash flows and/or low leverage. These results are consistent with existing evidence that independent firms are financially constrained in the pre-1990 period and that keiretsu main bank monitoring effectively controls the overinvestment problem.