| This thesis employs multi-period overlapping generations (OLG) models with aggregate risk to study questions at the interface of macroeconomics, public finance, and finance.;The first chapter addresses the equity premium puzzle. The equity premium puzzle refers to the inability of standard models to reproduce the large difference, on average, between the returns to risky and safe assets observed in the data. First proposed three decades ago, this puzzle has remained a challenge to economics. A large literature has tried to resolve it using complex machinery, such as nonstandard preferences or risk structures. I show that to resolve the puzzle it suffices to impose increasing marginal costs of borrowing in an otherwise standard OLG model.;In the second chapter, co-authored with Laurence Kotlikoff, we quantify the size of generational risk in an 80-period OLG model for the first time. Generational risk is the extent to which aggregate shocks are spread across contemporaneous generations. Under prefect risk-sharing, the consumption of all generations changes by the same percentage when a shock hits. The deviation from that perfect risk-sharing world is a measure of generational risk. Contrary to standard assumptions in the literature, we find that generational risk is small and that government policy can easily exacerbate it. We also show that a bond market can mitigate risk-inducing policy.;In the third chapter, also co-authored with Kotlikoff, we consider a long-standing problem: how to value government obligations when markets are incomplete. Our approach consists of determining the current wealth equivalent of future government promises using the change in remaining expected lifetime utility converted into current consumption units. We find that discount rates for policies involving sure payments each period to the elderly aren't uniform over time or agents of different cohorts. They also depend on the size of the payments and attendant general equilibrium effects. For infinitesimal promises, the discount rates are remarkably close to the prevailing short-term interest rate.;A technical contribution of this thesis is solving models like the above for the first time, by employing and extending Judd, Maliar, and Maliar (2011). |
