My thesis aims at understanding funds' stock selection abilities and investors' fund selection behaviors with a focus on sector funds, and contributes to the literature in two main ways. First, I perform comprehensive analyses of sector fund performance and provide new results. I examine the distribution of sector funds' equity holdings across sectors and show that sector funds make concentrated investments in their own sectors but at the same time make investments across diverse sectors. These results shed light on the construction of proper sector benchmark returns. In addition to commonly used four-factor regression analyses, I employ both a matching portfolio approach and a trade-based approach. The aggregate results indicate that among subgroups of sector funds, technology funds have the ability to hold and trade stocks that perform better than other stocks with similar characteristics or in the same sector. I find that neither sector funds nor nonsector funds improve performance by timely changing the characteristics or sectors of the stocks in existing positions in their portfolios. Second, I find some smart money effects. A few papers report the existence of smart money which realizes higher returns by trading mutual funds. These studies examine whether money inflows into mutual funds earn higher returns than money outflows, or whether aggregate new money flows receive positive returns. However, because performance metrics in the previous studies do not control for momentum components in returns, smart money effects could mirror momentum effects. I take a new approach using matching portfolio-adjusted returns and find that positive returns on new money in sector funds are mostly driven by momentum returns. In addition, I compare the aggregate returns on new money with the returns on old money under the same fund managements and find that for sector funds, new money earns significantly higher returns than old money. The results are consistent with the previous finding in the literature that money tends to remain in poorly performing funds while this analysis quantifies the return difference between new money and old money.
