In the United States, market hours worked are approximately flat across the wealth distribution. Accounting for this phenomenon is a standing challenge for standard heterogeneous-agent macro models. In these models, wealthier households consume more, enjoy more leisure, and work less. We propose a theory that generates the cross-sectional wealth-hours relation as in the data. We quantify this theory in the context of a new general-equilibrium heterogeneous-agent incomplete-markets model with three key features: a quality choice in consumption, non-homothetic preferences, and a multi-sector production structure. We show that the model produces expenditure patterns that are consistent with the data, as well as realistic “quality Engel curves.” 

We use supervised machine learning to approximate the expectations typically contained in the optimality conditions of an economic model in the spirit of the parameterized expectations  algorithm (PEA) with stochastic simulation. When the set of state variables is generated by a stochastic simulation, it is likely to suffer from multicollinearity. We show that a neural network-based expectations algorithm can deal efficiently with multicollinearity by extending the optimal debt management problem studied by Faraglia et al. (2019) to four maturities. We find that the optimal policy prescribes an active role for the newly added medium-term maturities, enabling the planner to raise financial income without increasing its total borrowing in response to expenditure shocks. Through this mechanism, the government effectively subsidizes the private sector during recessions.

How does housing illiquidity affect household risk-aversion and saving behavior? This paper shows that when housing services provide utility, the risk over the relative consumption ratio of nondurables and houses determines household relative risk aversion and drives asset prices. I show in a calibrated heterogeneous-agents model thataccounting for the relative consumption risk (i) helps to explain challenging asset pric-ing facts, such as the countercyclical market price of risk and a stable risk-free rate,(ii) greatly amplifies the business cycle fluctuations, (iii) illuminates the new source ofbusiness cycle costs (iv) helps to understand the endogenous variation in uncertainty. 

Can governments use Treasury Inflation-Protected Securities (TIPS) to tame inflation? We propose a novel framework of optimal debt management with sticky prices and a government issuing nominal and real state-uncontingent bonds. Nominal debt can be monetized giving ex-ante flexibility, whereas real bonds are cheaper but constitute a commitment ex-post. Under Full Commitment, the government chooses a leveraged and volatile portfolio of nominal liabilities and real assets to use inflation to smooth taxes. With No Commitment, it reduces borrowing costs ex-ante using a stable real debt share strategically to prevent future governments from monetizing debt ex-post. Such policies rationalize the small and persistent real debt share in U.S. data, with higher TIPS shares effectively curbing inflation. Reducing future governments' temptation to monetize debt renders debt and inflation endogenously sticky.