We examine how mutual funds form stock market expectations and the implications of these beliefs for asset returns, using a novel text-based measure extracted from Chinese fund reports. Funds extrapolate from recent stock market and fund returns when forming expectations, with more recent returns receiving greater weight. This recency tendency is weaker among more experienced managers. At the aggregate level, consensus expectations positively predict short-term future market returns, both in and out of sample. At the fund level, expectations are positively related to subsequent fund performance in the time series. In the cross-section, however, superior performance arises only when funds accurately forecast market direction and adjust their portfolios accordingly. This effect is stronger for optimistic forecasts and among funds with greater exposure to liquid stocks. Our findings highlight the conditional nature of belief-driven performance, shaped jointly by forecasting skill and the ability to implement views in the presence of execution frictions such as short-selling and liquidity constraints.

We study the influence of investor attention on mutual fund investors' fund selection and fund managers' portfolio choice. Using the Google Search Volume Index to measure investor attention on individual stocks, we find fund investors tend to direct more capital to mutual funds holding more high-attention stocks; fund managers tend to perform window-dressing trading to increase the portfolio holdings of high-attention stocks displayed to investors. Our results suggest that funds, particularly those with strong incentives, strategically trade on stock attention to attract investor flows. This strategic trading behaviour is also associated with fund underperformance and leads to larger non-fundamental volatility of holding stocks.

We investigate the economic impact of artificial intelligence (AI) adoption in the mutual fund industry by introducing a novel measure of AI adoption based on the presence of AI skilled personnel at fund management firms. We provide robust evidence that AI adoption enhances fund performance, primarily by improving risk management, increasing attentive capacity, and enabling faster information processing. Furthermore, we find that mutual funds with higher levels of AI adoption experience greater investor net flows and exhibit lower flow-performance sensitivity. While AI adoption benefits individual funds, we find no evidence of aggregate performance improvements at the industry level.

Using a unique dataset from China spanning 2005 to 2023, we investigate how superstitious beliefs influence mutual fund managers’ risk-taking behavior and how this influence evolves over their careers. We find a significant 6.82% reduction in risk-taking during managers’ zodiac years, traditionally considered unlucky in Chinese culture. This effect is particularly pronounced among less experienced managers, those without financial education backgrounds, and those with lower management skills. The impact also intensifies during periods of high market volatility. Our findings challenge the traditional dichotomy between retail and professional investors, showing that even professional fund managers can be influenced by irrational beliefs early in their careers. However, the diminishing effect of superstition with experience and expertise suggests a gradual transition towards more rational decision-making. Our results provide insights into the process by which financial professionals evolve from exhibiting behavior akin to retail investors to becoming the rational actors often assumed in financial theory.

This paper studies the impact of AI technology on the mutual fund industry. I develop a new measure of AI adoption based on hiring practices and find that this measure can predict fund performance. The funds with high AI ratio outperform non-AI funds, after I controlling for standard factors and fund characteristics. Further empirical evidence shows that funds with a high AI ratio tilt their portfolios toward high information intensity stocks, indicating that mutual funds benefit from AI technology adoption by improving their information capacity. Consistent with this channel, I find that the outperformance of these mutual funds mainly comes from better stock picking skills. Finally, AI technology adoption has a negligible effect on fund manager turnover.

We examine the performance of actively managed Chinese mutual Funds over the period 2002-2020. Using the bootstrap-based false discovery technique, we find that 19.25% of Chinese actively managed mutual funds produce positive-alpha, which contrasts with existing studies documented by others in developed markets. Our findings survive a battery of robustness tests. Unlike in developed markets, equilibrium accounting may not hold in China as the Chinese stock market is dominated by retail investors instead of mutual funds, and thus the mutual funds in China can be more skilled at the expense of the retail investors. We find supportive evidence of the applicability of the bootstrap-based false discovery rate method by conducting simulations.

This study examines the impact of quantitative investment on stock price crash risk from the perspective of quantitative mutual funds holdings. The results show that quantitative mutual funds holdings can significantly reduce stock price crash risk, and this effect is more pronounced in subsamples characterized by executives with overseas backgrounds, higher internal governance efficiency, greater analyst attention, and higher profit volatility. Further research finds that quantitative mutual funds holdings can suppress the risk of stock price crash by smoothing the volatility of stock returns and optimizing the valuation of firms. This study sheds light on the effects of quantitative investment on stock price crash risk.

This paper finds that reported fund returns do not necessarily represent the returns of mutual fund investors, especially over long investment periods. We show that mutual fund’s reported returns are calculated using NAV and represent the mutual fund manager’s skill in extracting value from the capital market. However, the real returns earned by mutual fund investors depend not only on the mutual fund manager’s skill but also on the subscription and redemption activities. Using the inflow and outflow information reported in the mutual funds’ semi-annual reports in China, we are able to calculate mutual fund investors’ real returns. We further derive the adjusted gain coefficient (AGC) to capture the difference between the reported mutual fund returns and the mutual fund investors’ real returns. We find that the AGC is significantly lower than 1, which suggests that the real returns of mutual fund investors are significantly lower than reported mutual fund returns in China. The underperformance of mutual fund investors relative to the mutual fund managers they invest in is very persistent and is stronger in more recent years. A further investigation reveals that this underperformance is largely attributed to investors’ poor timing skills and additional fees incurred as a result of excessive subscription and redemption activities. We also identify skilled mutual fund investors using AGC and find that fund managers can benefit from investors’ timing skills. Skilled mutual fund investors flow in when the mutual fund managers have good investment opportunities and flow out when the mutual fund managers have extra cash. The synchronization of the mutual fund investors’ flow and mutual fund managers’ investment strategies can reduce the need for liquidity management and improve mutual fund performance. Using Chinese mutual funds data, we show that a 1% increase in AGC can increase fund riskadjusted return by 0.2% in the next six months.

As many ESG rating agencies have flourished to meet rising interests in ESG investing, we examine the information provider role of these rating agencies. We hypothesize that new ratings can add information useful to investors about rated firms besides any changes to the average level and dispersion in ratings. We exploited the empirical setting where the introduction of various ESG ratings in China is staggered over time and across firms. We show that an increase in the number of ratings by different agencies for a given firm will induce more mutual funds’ investments towards that firm. This is unexplained by rating inflation or rating shopping channels. We further show that such effect is more pronounced when incumbent and entrant agents provide complementary information. For different types of funds, we find different sensitivities to the arrival of new agents in accordance with their explicit requirements for ESG mandate. And interestingly ESG funds that track ESG indices are not responsive to new ratings as ESG indices are sticky in choosing the reference rating. We also provide evidence that the documented effects are not due to endogenous actions taken by incumbent agencies or the firms. Our paper provides interesting and causal evidence of the incremental information from additional ESG ratings which have important implications for the market competition and regulations of ESG rating agencies.

Common institutional ownership can enhance knowledge spillovers by increasing portfolio firms’ awareness about each other’s innovation. By investigating listed electronic hardware firms in China for 2000-2016, we find that when common ownership by mutual funds is higher between a firm pair, it is more likely that these two firms cite each other’s patents. To confirm causality, we show that even the exogenous increase in firms’ common ownership following their inclusion into the stock index still positively influences the citing likelihood. We also find that such citations are taken place in a timely manner. Additionally, this positive effect is robust when the effects of overlapping board members and common ownership by other types of institutional investors are controlled for. This effect is more pronounced among nonneighboring firms, when non-neighboring firms are close to their common owners, when common owners hold shares longer, and when firms’ executives have lower incentive to communicate (i.e., SOEs). Last, we find that common ownership by mutual funds also enhances knowledge spillovers through third-party patents. This paper deepens the understanding of knowledge spillovers among firms in developing countries.