This paper investigates household leverage-dependent fiscal policy effects in a two-agent New Keynesian DSGE model with occasionally binding borrowing constraints. Our model successfully replicates empirical evidence showing that fiscal policy's effectiveness differs significantly across the household leverage cycle. Fiscal multipliers are persistently above unity when government spending rises at the peak of the household leverage cycle. In contrast, increases in government spending at the trough of the household leverage cycle imply fiscal multipliers below unity. We test the model's predictions on post-WWII U.S. data.

Replication Codes

In this paper, we use the property that certainty equivalence, as implied by a first-order approximation to the solution of stochastic discrete-time models, breaks in its equivalent continuous-time version. We derive a risk-sensitive first-order perturbation solution for a general class of rational expectations models. We show that risk matters economically in a real business cycle (RBC) model with habit formation and capital adjustment costs, and that neglecting risk leads to substantial pricing errors. A first-order perturbation provides a sensible approximation to the effects of risk in continuous-time models. It reduces pricing errors by around 90 percent relative to the certainty equivalent linear approximation. 

Additional Materials:  Online Appendix, Codes

This paper studies the preferential tax treatment of housing that can be observed in many industrialized countries. It provides a rationale for it by means of an optimal taxation approach taking into account an important feature of housing, namely its usage as collateral. In a borrower-lender framework where private loans are assumed to be non-enforceable and have to be collateralized by housing, optimal fiscal policy should disburden constrained borrowers by subsidizing their housing.

Working paper version can be downloaded here.

This paper studies the quantitative contribution of MBS purchases to the macroeconomic effects of the US Federal Reserve's first round of quantitative easing, QE1. We develop a macroeconomic model with costly financial intermediation, collateralized lending, and an explicit specification of distinct central bank asset purchase programs. We show that MBS purchases induce a fall in mortgage rates and increases in the price of collateral and in bank lending, which stimulates aggregate demand and real activity. We replicate the time paths of purchased asset and of associated price effects at the onset of the global financial crisis. Our calibrated model predicts that MBS purchases accounted for more than 85% of the (cumulated) output effects of QE1 and that an equally sized intervention with purchases of treasuries only would have led to about 30% smaller output effects.

This paper argues that the choice of frequency, or the length of a time period, in real business cycle (RBC) models is not only theoretically relevant, but also carries important economic implications. First, we demonstrate analytically that in the stylized RBC model with log utility and full depreciation marginal propensities to save and consume, and thereby, the annualized risk-free rate depend on the chosen frequency. We then explore a richer RBC model that incorporates capital adjustment costs and habit formation, known for its ability to generate significant risk effects. Within this framework, we demonstrate that the monthly calibration, compared to the annual calibration, results in a 50% higher (annualized) risk-free rate. Similar patterns emerge when examining the equity risk premium.

(Preliminary version, available upon request.)

This paper develops a continuous-time two-agent endowment economy with lenders and indebted households facing collateral constraints. The stylized framework in continuous time makes it possible to analytically derive new insights on the interconnection between household debt and different types of inequalities. We consider in this stylized framework inequalities in consumption of non-durables and durables (housing), and in housing, financial, and total wealth as well as in income and welfare.

(Preliminary version, available upon request.)