Research

Abstract: People can be ``locked-in” or constrained in their ability to make appropriate financial changes, such as being unable to move homes, change jobs, sell stocks, re-balance portfolios, shift financial accounts, adjust insurance policies, transfer investment profits, or inherit wealth. These frictions—whether institutional, legislative, personal, or market-driven—are often overlooked. Residential real estate exemplifies this challenge with its physical immobility, high transaction costs, and concentrated wealth. In the United States, nearly all 50 million active mortgages have fixed rates, and most have interest rates far below prevailing market rates, creating a disincentive to sell. This paper finds that for every percentage point that market mortgage rates exceed the origination interest rate, the probability of sale is decreased by 18.1%. This mortgage rate lock-in led to a 57% reduction in home sales with fixed-rate mortgages in 2023Q4 and prevented 1.33 million sales between 2022Q2 and 2023Q4. The supply reduction increased home prices by 5.7%, outweighing the direct impact of elevated rates, which decreased prices by 3.3%. These findings underscore how mortgage rate lock-in restricts mobility, results in people not living in homes they would prefer, inflates prices, and worsens affordability. Certain borrower groups with lower wealth accumulation are less able to strategically time their sales, worsening inequality.​

This paper studies how mortgage debt shapes the consumption response to cash transfers using an incomplete markets model with housing and long-term debt. Among homeowners, the model predicts those with mortgage debt have an average spending response over ten times larger than those without debt, and higher levels of leverage are associated with larger increases in spending. Responses in the model are found to be poorly correlated with income. By excluding many homeowners with debt, conditioning transfers on having low income reduces their efficacy in increasing aggregate spending. The opposite is predicted by a conventional heterogeneous agent model.

This paper studies the accumulation of external public debt in low-income countries (LICs) after receiving debt relief from multilateral lenders. While LICs who received debt relief from the International Monetary Fund in the early 2000s generally lowered their external debt in the initial years of relief, many experienced fast resurgence in debt after borrowing limits were lifted in 2006. Using a difference-in-differences model, we find countries that benefited from the relaxation are more likely to experience a significant increase in their debt-to-GDP ratio. We quantitatively evaluate the effects of debt limit relaxation using a model of sovereign default with two types of debt: subsidized loans from multilateral institutions and non-concessional loans from the private market. The model is calibrated using data from Mozambique prior to its default in 2016, and we find that an impatient government is necessary to match the data on debt accumulation and investment. In the calibrated model, lifting limits on non-concessional borrowing results in a 17.1 percent loss in households' consumption-equivalent welfare. Furthermore, the welfare benefits of relaxing borrowing limits with a counter-factually patient government are relatively small.

Abstract: I study how labor income taxation should vary with a household’s income history in a life-cycle model with differentiated skill types and general equilibrium wages. I consider a very general class of tax functions that can depend on the entire income history, which requires solving a dynamic maximization problem with up to 41 state variables. To maintain the feasibility of the optimal taxation problem with a large state space, I use a novel neural network approach to simultaneously solve the model and compute the labor income tax function that maximizes steady state welfare. I find that the optimal history-dependent tax function behaves very differently depending on a household’s income history. For most households, average tax rates increase over the life-cycle. However, for households with very high income histories, average tax rates decrease over the life-cycle. The welfare gains from history-dependent taxation are large, equivalent to about a 2 percent increase in lifetime consumption compared to a tax on current income. This suggests that history-dependent taxation may have large welfare benefits in reality, but the exact benefit will depend critically on the production structure of the economy and how complementary different types of labor are. Finally, I show that a parametric function that varies with just the average of previous income levels mimics the full history-dependent tax system by increasing taxes more slowly over the life cycle with higher levels of average past income.