Essays on macroeconomic linkages to equity risk premiums

Abstract

This dissertation mainly improves our understanding of the linkage between asset prices and the business cycle movements. In this dissertation, I first offer two extensions to enhance the performances of a consumption-based asset pricing model of loss aversion preference. Second, I propose a new habit sensitivity function in a production-based asset pricing model to explain the equity premium and size premium. Besides, I identify that long-run productivity risk is one of the potential driving forces behind the investment-based q-factor asset pricing model. The whole dissertation consists of three essays.

In the first essay, I incorporate the recursive utility into Pagel’s reference-dependent preference and study their aggregate implications in a consumption-based asset pricing model. In the case of recursive utility, the proposed model reproduces crucial asset pricing moments and time-varying risk premiums with a simple IID process for consumption growth. Second, the proposed model consistently predict that agent prefers a late resolution of uncertainty in both time-separable and recursive utility. My additional finding is that intertemporal substitution elasticity is more sensitive to asset prices given the recursive preference. Finally, the introduction of sluggish-updating can improve the model performances.

In the second essay, I study the interaction between lumpy investment and asset prices in both time-series and cross-section. To this end, I work with a variant of habit sensitivity function introduced in Campbell and Cochrane. The model produces 100% equity volatility of data by generating volatile marginal utility under the assumption of non-convex adjustment costs. Second, the model reproduces nearly 100% equity premiums of data because it assigns large weights on precautionary savings and constrained firms, respectively. Furthermore, the model can rationalise considerable size premiums as small firms absorb more productivity risks. Finally, the model matches key macroeconomic moments and the cross-sectional investment rate.

In the third essay, I incorporate the productivity risks into an investment-based q-factor asset pricing model. The productivity risks factors broadly summarise the cross-sectional portfolio return, in which the time-varying volatility plays an important role. A parsimonious q-factor model driven by productivity risks explains about 90% variation of return of 25 Size/BM portfolios and around 75% variation of return of 160 portfolios, which is comparable to the Fama-French multifactor models, the Carhart four-factor model, and the Hou et al. q⁵ model. As such, productivity growth risks can be one of the potential driving forces of investment-based factor models.

Date of Award	2 Jul 2021
Original language	English
Awarding Institution	University of St Andrews
Supervisor	Matthew Paul Knowles (Supervisor), Oliver de Groot (Supervisor), Ioannis Psaradellis (Supervisor) & Roderick McCrorie (Supervisor)