We globally solve the canonical HANK model extended with transitory idiosyncratic productivity shocks and compute generalized impulse response functions for fiscal transfer shocks. Without aggregate risk, our global solution delivers the same impulse responses as a non-linear Sequence Space Jacobian solution. Negative transfer shocks are amplified relative to a first-order solution, while positive shocks are dampened. With aggregate risk, the stronger precautionary saving motive implies a higher bond-to-income ratio on average, leaving households better insured and dampening the response to negative shocks. We also present evidence that approximate aggregation might hold despite large fluctuations in the share of hand-to-mouth households.

An increase in permanent labor income inequality (PLI) increases aggregate savings because high-PLI households save more than poorer ones. For the asset market to clear, prices must adjust. This paper studies how this change in prices feeds back on the distribution of wealth. If higher households savings leads to more capital accumulation, the interest rate will decrease and wages will increase, “trickling-down” towards poorer households. If it only increases the price of financial assets, interest rates and wages will remain constant, and capital gains will increase the income of richer households. In a heterogeneous agent model with a non-homothetic taste for wealth and imperfect competition, we show that the level of markups dampens the trickle-down effect and increases the valuation effect. Our model calibrated to the U.S. economy suggests that the general-equilibrium effects of rising PLI inequality increased average wealth of the top 0.1% by 16% between 2020 and 1970, against 3% for the average household in the economy.

We introduce a novel deep learning algorithm for solving large life-cycle models with both continuous and discrete choices. This allows us to simultaneously account for both labor supply choices with human capital accumulation, portfolio choices with a risky and a risk-free asset, and housing and mortgage choices. We work directly on the Bellman equation and approximate both value and policy functions with neural networks, and use a simulated training sample instead of tensor product grids. This substantially alleviates the curse of dimensionality. We demonstrate this in a consumption-saving model with multiple durable goods subject to non-convex adjustment costs where our deep learning algorithm clearly outperforms a standard value function iteration. We confirm that we can accurately solve a large life-cycle model in 12 hours on a single GPU. We solve the model simultaneously across all periods instead of with backward induction. This simplifies transfer learning, where a new solution for a new set of parameters in a calibration or estimation can easily be achieved. An accompanying easy-to-use software package implements the method.

We study how rising permanent labor-income inequality shapes monetary and fiscal transmission in a Heterogeneous-Agent New-Keynesian model with a non-homothetic taste for wealth. Higher inequality increases the share of hand-to-mouth households, raising aggregate MPCs and lowering the effective intertemporal elasticity of substitution. As a result, the direct effect of monetary policy weakens, while the indirect effect strengthen, leading to a larger output response. A higher share of capital gains generated by monetary policy are paid to wealthy households, making monetary policy more regressive. For fiscal policy, higher MPCs raise impact and cumulative multipliers and make deficit-financed expansions more likely to be self-financing. Finally, the fiscal response to a monetary policy shock becomes more important to determine the output response to a change in interest rate.

Monetary tightening can generate inefficient firms exits, by exacerbating firm illiquidity constraints. Contrary to the common belief that only small firms are affected, new evidence shows that large firms are vulnerable too, leading to large aggregate effects. This paper investigates why monetary tightening causes large firms to exit and explores the macroeconomic implications. In this paper, we study why monetary tightening can lead to large firms exit, and explore its macroeconomic implications. To do so, we develop a model of endogenous firm exit with financial frictions and partial irreversibility. Financial frictions imply that only productive firms are able to take on debt, making them highly exposed to interest rate changes, as partial irreversibility makes deleveraging costly. The quality of firm selection can therefore endogenously worsen during monetary tightening episodes, highlighting a potential need for support for large firms in distress.