Labor-Based Asset Pricing [electronic resource].

9781088308479

Ann Arbor : ProQuest Dissertations & Theses, 2019.

1 online resource (159 p.)

Access is available to the Yale community.

Access restricted by licensing agreement.

My dissertation has three chapters. In the first chapter, I develop a labor-based asset pricing model empirically and theoretically that links expectations of returns and cash flows to firms' labor search decisions. Using a dataset that covers the near-universe of U.S. online job vacancy postings, I show that vacancy rates negatively predict stock returns and positively predict cash flows in the cross-section of firms and industries. Moreover, these results systematically depend on firms' labor-market conditions. The predictive power of vacancy postings is strengthened for firms with unfavorable labor-market conditions, including industries with low vacancy-filling rates, firms with low relative labor market concentration, and firms with high skill requirements. In aggregate time-series, I construct a new measure of employment value incorporating information about vacancies, hires, and the costs of vacancy postings. The measure strongly predicts aggregate stock and bond market returns, even in the presence of other known predictors such as the capital investment rate and various price factors such as the dividend-price ratio, earning-price ratio, consumption-wealth ratio, and relative bill rate.I build a dynamic production-side asset pricing model that combines heterogeneous firm production decisions with varying individual labor-market conditions. I show that the results of the model quantitatively match the empirical findings. When the expected return is low or the expected cash flow is high, the present value of additional labor is high, and therefore a firm responds by posting more vacancies. Moreover, firms are constrained by the labor-market conditions they face. When it is relatively hard to fill vacancies or when the labor market is unfavorable for firms, variations in vacancy posting rates are more informative about the underlying expectations of returns and cash flows.In the second chapter, we construct a time-varying network of labor competitors for all U.S. public companies. We show the importance of this network for transmitting labor and industry shocks. There are three main findings. First, the overlap between firms' labor competitors and product market rivals is less than 20 percent. Second, firm returns strongly respond to both the contemporaneous and lagged labor market shocks proxied by returns of the labor competitors. A long-short strategy exploiting the lagged response generates an average annualized excess return of 9.36 percent. Third, shocks to an industry can affect firms outside the industry through the labor network.The last chapter of my dissertation documents the existence of long-run risk in consumption growth. We take the novel approach of using news coverage to capture investor concerns about economic growth prospects. We provide evidence that consumption growth is highly predictable over long horizons: our measure explains up to 24 percent of cumulative future consumption growth at the 6-year horizon and beyond. Furthermore, we show a strong connection between this predictability and asset prices. Innovations to our measure can account for variations of 51 standard portfolios in the cross-section, and this 1-factor model outperforms many benchmark multi-factor models.

Dissertations & Theses @ Yale University.

Books / Online / Dissertations & Theses

English

January 17, 2020

Thesis (Ph.D.)--Yale University, 2019.

Yale University. Economics.