Firm Organization Matters: Evidence from Investment Responses of Centralized and Decentralized Mutual Funds to Information Shocks

Abstract

In the world of money management, professional intermediaries constitute a large and important group. How much does internal firm organization affect the performance and allocation of resources in these intermediaries? This question has received relatively little attention in academic literature. We examine this question in the context of mutual funds and provide direct evidence that internal organization structure -- in particular the decision making process about investments -- impacts their investment behavior and performance. Our empirical evidence is consistent with tradeoffs generated by theories of organizational economics. Decentralized mutual funds provide more discretion in investment decisions to their managers and as a result have higher performance than their centralized counterparts. However, since decision making is done individually by managers in decentralized structure, there is a diversification loss due to individual funds not accounting for the correlation of their own portfolio returns with the returns of other managers in the fund. These results obtain both in the panel setting and in a quasi-experiment that involves an exogenous shift in the precision of publicly available information on some stocks. The results of the paper offer several broad implications for issues related to delegated portfolio management. First, the results help understand what drives some funds to deliver better performance than others. This has potential to shed light on the long standing academic debate on whether or not mutual fund managers have skills. If a large population of funds have centralized investment decision making, it may explain why many studies find that managers in the population do not have superior skills. Second, by showing that organizational structure is an important determinant of fund strategies and performance, it connects the literature on firm boundaries with financial intermediation. Besides shedding light on the "black box" of mutual fund management, this insight has the potential to inform us on the shifts in the mutual fund industry over time. For instance, our research could inform us on how competition for talent from hedge fund industry (with more decentralized decision making) may have influenced mutual fund families to changing their decision making process to a decentralized one as well. Alternatively, our research can inform better on how the difference in investment and performance across different organizational forms may be expected to evolve depending on the labor market conditions.