Research

Abstract: We analyze the economic effects of two major family policies in the United States, the Child Tax Credit and the Child Care and Development Fund childcare subsidy, in an overlapping generations framework where altruistic parents invest in their child’s skill using their own time and purchased childcare time. The model incorporates differences in the design of these policies and endogenizes low rates of childcare subsidy receipt by including application costs and subsequent rationing. We compare the effects of a recent child tax credit expansion with a spending-equivalent expansion of the childcare subsidy implemented by reducing access frictions. Across steady states, the childcare subsidy expansion generates a larger increase in average adult skill, which leads to larger welfare gains behind the veil of ignorance compared to the tax credit expansion. However, the two policies yield similar average welfare gains for adults who know their own skill level, and the tax credit benefits a larger share of this group. 

Abstract: How should governments with a preference for redistribution design tax policies when facing limited borrowing? This paper studies optimal taxation in a small open economy with heterogeneous agents and endogenous debt constraints arising from the government's limited commitment to fiscal policies. The optimal labor tax decreases over time and is non-zero in the limit, and the optimal capital and domestic borrowing taxes are positive in the limit, deviating from the standard Ramsey tax results. The government’s redistributive motive directly affects optimal tax levels, whereas binding debt constraints influence optimal tax dynamics. In the numerical analysis, a stronger redistributive preference requires greater initial tax distortions and a higher external debt level in the long run.

Abstract: The recent decrease in U.S. money velocity raises debates about its unit root behavior. This paper revisited the random walk hypothesis (RWH) of the U.S. money velocity in 1960-2010 and two sub-periods 1960-85 and 1986-2010 by applying the Variance Ratio methodologies, including new nonparametric tests by Wright (2000) and Belaire-Franch and Contreras (2004). The results suggested that the velocity would likely increase, and the U.S. monetary policy will soon stimulate GDP and employment. Furthermore, past velocity is important to predict the future outcomes, and changes in financial structural could alter the empirical characteristics of the velocity series.

Abstract: This paper develops a theory of sovereign debt sustainability driven by the government’s redistributive motive. I study a heterogeneous-agent small open economy where redistribution relies on distortionary labor taxation and the government lacks commitment. Access to international credit markets lowers the cost of redistribution, while default into financial autarky raises it, generating an endogenous cost of default. Quantitatively, the model accounts for the buildup of Italy’s external debt and the positive cross-country correlation between pre-tax income inequality and external debt. Optimal austerity is more gradual when distributional concerns are present.

Abstract: Sudden stops are often accompanied by high income inequality and increases in consumption inequality during crises. We rationalize this fact in a sudden stop model of collateral constraints by incorporating income inequality and nonhomothetic preferences. Nontradable goods are more income-elastic than tradable goods. Borrowings of high income households have a stronger effect on future real exchange rates than low-income households. Excessive international debt accumulation by high income households increases the frequency and severity of sudden stop crises. On the other hand, low income households underborrow from a social perspective. Both income inequality and nonhomotheticity of preferences amplify the inefficiency of the pecuniary externality. Therefore, macroprudential policies are more welfare improving than in standard models.

Abstract: This paper develops a return variance decomposition model to examine how sovereign credit risks respond to different sources of information and noise. Using trading data on sovereign credit default swaps (CDS), we decompose sovereign CDS spreads into four components: noise, global-market information,  country-specific private information revealed through trading, and country-specific information disclosed through public announcements. We find that the variance in spreads is largely driven by country-specific information. We use the model to evaluate the role of noise in sovereign credit spread movements and examine how information respond to changes in global and domestic macroeconomic factors.