Doctor of Philosophy
MacGee, James C.
My doctoral dissertation consists of three chapters on financial economics. In the first chapter, I examine whether a firm's share repurchases and issuances reveal information on its future 2-year stock returns. Firm-quarter observations with repurchases or issuances are each divided into twenty 5-percentile bins sorted by their magnitude. I regress 2-year future stock returns on the bins and find a non-linear relationship between the change in shares outstanding and returns. Firms making repurchases (issuances) of less than 9.3% (1.4%) of shares outstanding outperform the equal-weight portfolio by 6.4% (3.9%). These observations account for 90% of repurchases and 70% of issuances. Firms making larger repurchases (issuances) underperform by 1.9% (5.1%).
In the second chapter, I examine whether investors could use share repurchases and issuances to create outperforming portfolios from 2003-2019. First, I examine two exchange-traded funds which track repurchasing companies and find no evidence they outperformed. Second, I investigate whether repurchases and issuances of U.S. firms have predictive effects from 2003-2017, and I find large issuances predict lower returns. Finally, I construct a portfolio which short-sells the stocks of firms which make large share issuances and a portfolio which invests in all firms except the large share issuers. I backtest these strategies and find both outperformed the market.
In the third chapter, I show an economy with delegated investment management and assets with correlated tail risk will experience endogenous boom-bust cycles where longer booms lead to larger crashes. I examine a dynamic model populated by savers and investment managers, where savers delegate their wealth to investment managers to invest in a project. Risky projects produce returns that depend on the investment manager's ability. Tail-risk projects produce high average outputs after a good aggregate shock but produce no output after a bad shock. In equilibrium, savers fire managers who generate low returns. Low-ability managers to invest in tail-risk projects to reduce their chance of being fired. As such, the population of low-ability managers increases after good shocks and falls after bad shocks, which produces boom-bust cycles in output where longer booms are followed by larger crashes.
Summary for Lay Audience
My doctoral dissertation consists of three chapters on financial economics. My first chapter examines whether a firm's share repurchases and issuances predict the firm's future stock returns over the 1975-2017 period. Share repurchases occur when a firm buys back its shares from the stock market, and share issuances occur when a firm issues new shares on the stock market. I find that firms making repurchases (issuances) of less than 9.3% (1.4%) of its previous shares outstanding over 3 months have higher returns of 6.4% (3.9%) over the following 2 years, compared to firms which did not make repurchases or issuances. These observations account for 90% of repurchases and 75% of issuances. Firms making larger repurchases (issuances) underperform by 1.9% (5.1%).
In the second chapter, I examine whether investors can use share repurchase and issuance information to create portfolios which outperform the market. First, I examine two exchange-traded funds which track the returns of repurchasing companies and find no evidence they outperformed the market. Second, I investigate whether repurchases and issuances of U.S. firms have predictive effects in the 2003-2017 period, and find large issuances predict lower returns. Finally, I construct two portfolio which use the information from large issuances and show they outperformed the market from 2003-2019.
In the third chapter, I show that if assets with correlated tail risk exist, investment management can cause boom-bust cycles where longer booms lead to larger crashes. These tail-risk assets generate higher average returns than other assets most of the time, but there is a small chance that all tail-risk assets generate no returns and lose their value. Investment managers with less ability to generate high returns prefer to invest in assets with tail risk to improve their chances of attracting and retaining clients. When the tail risk event does not occur, the number of lower-ability investment managers and the amount invested in assets with tail risk increase . As such, the amount invested in assets with tail risk and the losses that would occur in a crash increase with a longer period of stability.
Li, Bohan, "Essays on Share Repurchases and Boom-Bust Cycles" (2020). Electronic Thesis and Dissertation Repository. 7389.