We develop a global method to solve and estimate dynamic equilibrium models that treats prices as pseudo parameters and market clearing as moment conditions, and reduces estimation time from days to minutes. Our approach leverages AI algorithms, software, and hardware, and has three building blocks. First, we extend the state space to include equilibrium prices and model parameters, which allows us to clear markets and estimate parameters by solving the model once. Second, we approximate the mapping between parameters and moments by training neural networks on model-simulated data, which act as closed-form expressions for moment conditions. Third, we use this mapping to estimate parameters by minimizing the distance between the model and data moments, and to find equilibrium prices by targeting a market-clearing imbalance of zero. We also use this mapping to assess identification globally, verifying if the estimation objective function has a unique minimum for each parameter. We illustrate our method by estimating a dynamic general equilibrium model of leverage and investment with three state variables, three controls, endogenous default, costly equity issuance, and non-convex adjustment costs. After four days, the traditional approach does not reach the loss we achieve in under 20 minutes. We build an AI agent that applies our method to new models from natural language prompts.

In many areas of economics, relatively simple models developed for insight are used as quantitative guides. We study the accuracy of such simple quantitative guidance in an area where it has been widely adopted — lifecycle portfolio choice among stocks, bonds, and liquid accounts — by developing a machine-learning algorithm to solve for optimal portfolio choice in a calibrated lifecycle model that includes many features of reality modeled only separately in previous work. Both for optimizing households and for households that under-save, the average fully optimal portfolio at each age conforms well to current simple age-dependent prescriptive rules until shortly before retirement, validating existing analyses. We further show that the consumption-equivalent losses from conditioning portfolio shares on age alone are substantial, around 2 to 3 percent of consumption. Fully optimal equity shares have substantial heterogeneity, particularly by wealth level, state of the business cycle, and dividend-price ratio, implying substantial gains to further customization in these dimensions.

Media: Wall Street Journal, Morningstar

Mortgage borrowers are "locked in": forgoing moves to hold on to low rates. We show that missing downsizers stay in larger homes, raising net housing demand. We design a spatial housing ladder model with long-term mortgages, generating a distribution of locked-in rates and causal mobility effects consistent with the data. A temporary rate hike causes lock-in, increasing house prices by 4.4% and rents by 1.4% relative to a counterfactual without lock-in, offsetting a third of the house-price decline caused by higher rates. A starter-home seller subsidy modestly increases mobility at a high cost per marginal move, suggesting demand-based policies are poorly targeted responses to lock-in.

​

Media: NPR Planet Money (The Indicator), NPR Planet Money, NPR Morning Edition, Vox​

One in seven Americans carry medical debt, with $69 billion reported on consumer credit reports. In April 2023, the three major credit bureaus stopped reporting medical debt collections below $500. We study the effects of this information deletion on consumer credit scores, credit access, repayment behavior, and payday borrowing. Regression discontinuity estimates comparing individuals just above and below the $500 threshold show that the deletion reduced the reported number of medical debt collections by 61 percent. Despite expectations that removing negative credit information would improve credit outcomes for affected individuals, we find no evidence of benefits over the subsequent two years, ruling out even small effects. To interpret these findings, we build credit scoring models and show that medical debts, regardless of size, add minimal incremental information for default prediction beyond standard credit report variables, implying that they contribute negligibly to credit risk assessment. Our results suggest that eliminating medical debt collections entirely from credit reports would be unlikely to affect credit outcomes.