| This thesis consists of two chapters. The first chapter develops a theoretical model of investment by mutual fund managers that generates the short-run persistence in fund performance and long-run reversal in performance observed in the data. The key elements of the model are that (1) trading has short-run price impacts, (2) better recent fund performance leads to greater new cash inflows, (3) firms incur trading costs selling shares so that there is a distinction between cash and invested funds, and (4) fund manager compensation rises with dollars under management. We show that these elements cause a fund manager to distort investment of new cash inflows toward stocks in which the fund holds larger positions. In turn, this leads to short-run persistence in fund performance, but eventually to a long-run reversal in performance. If cash inflows to the best performing funds are high enough, short-run poor performance will be more persistent than short-run strong performance. These results will lead mutual funds to under-perform the market. We show that these effects will tend to be larger in newer or smaller funds, growth funds and more narrowly focused (e.g. technology) funds. We also derive predictions for the timing of trades, and find supporting empirical evidence.; In the second chapter, we construct a model to test the impact of short sell restrictions on the price discovery process of Diamond exchange traded funds. Short sell restrictions prevent trading in individual stocks during down markets. Diamonds are traded continuously and hold a publicly known portfolio of DJIA stocks. We test the distributional implications of short sell restrictions by using trade level data to construct minute-by-minute trade level approximations of the Diamond portfolio and compare them to Diamonds during periods when the short sell restrictions bind. We find no evidence that short sell restrictions introduce persistent premiums or discounts. |
