Contractual mutual fund governance: the case of China

The supervisory board and the board of directors are the two components of the two-tier governance structure of Chinese corporations.

For listed companies in China, corporate governance does not vary too much from Western practices in that companies are required to have a board of directors and adequate periodic financial disclosures that are audited by reputable accounting firms. There are, however, two areas of corporate governance where China tends to be quite different. First, the percentage of state control is extraordinarily high due to the large number of state-owned enterprises (SOEs) in the economy. SOEs may be charged with other social goals than the pure shareholder value maximization objective. This unique feature certainly has implications for resolving majority and minority shareholder interests. The other difference is that executive option pay is not widely used in China, although it is gradually gaining popularity. For a comprehensive review of corporate governance in China, see, for example, Clark (2003) and Liu (2006).

See the official website of the Securities Association of China (http://​www.​sac.​net.​cn).

Such as fund size, fee structure, trading activities and management characteristics. (Dahlquist et al. 2000; Golec 1996; Prather et al. 2004).

CSMAR is the leading financial data provider in China and is included in the WRDS database for Chinese research.

We use all listed A share stocks in the construction of the four-factor premia. As of the end of the formation months (June and December of each year), sample firms are classified into the top 50 % (B) and bottom 50 % (S) portfolios by their respective market capitalization. We further classify sample stocks into the bottom 30 % (L), middle 40 % (M) and top 30 % (H) portfolios by their book-to-market ratios. We then assign sample stocks into six portfolios, namely B/H, B/M, B/L, S/H, S/M and S/L. For each week, we calculate SMB as the weekly return difference between the value-weighted average return on the three small portfolios (S/H, S/M and S/L) and the value-weighted average return on the three big portfolios (B/H, B/M and B/L). Similarly, HML is the weekly return difference between the value-weighted average return on the two high portfolios (B/H and S/H) and the value-weighted average return on the two low portfolios (B/L and S/L). According to Carhart (1997), a momentum factor is formed by ranking all stocks on their past 11-month return lagged one month. The top and bottom 30 % stocks are assigned to the outperformed and underperformed portfolios and PR1YR is the weekly return difference between equal weighted returns on the two portfolios.

For quite a number of fund management companies in our sample, there are shareholders whose equity stakes tie for the top spot.

We have also considered the Herfindahl index of the top five shareholders of the fund management company as an ownership concentration measure. However, this measurement is highly correlated with TOP1 with a correlation coefficient above 0.9. Thus, we do not include this measurement in our analysis.

We have also considered the effect of various expenses on fund performance. As fund expenses are found to be highly correlated to fund fees, we exclude fund expenses from our empirical analysis. Empirical tests made using fund expenses rather than fund fees generate similar results and are available upon request.

See Kleinbaum et al. (1988, p. 210).

If we ignore a short rise, the stock indices have been declining since the middle of 2001.