With Vincent Sterk. Journal of Monetary Economics, 2021.
This paper studies quantitative easing in a heterogeneous-agent New Keynesian model with liquid and partially liquid wealth. Central-bank asset purchases transform long-term assets into reserves and bank deposits, expanding the liquidity available to households. Because households consume more out of liquid wealth than out of less-liquid wealth, QE can stimulate aggregate demand, output, and inflation even when conventional representative-agent models imply balance-sheet neutrality.
The direct effect of QE depends on the difference between marginal propensities to consume out of liquid and illiquid wealth. By increasing household deposits, QE raises spending precisely because the MPC out of liquid wealth is substantially higher.
In the estimated model, Federal Reserve asset purchases generated large-scale deposit creation and prevented a considerably deeper fall in output. The effects were strongest during QE1 and smaller in subsequent rounds as policy became anticipated and prices had more time to adjust.
QE initially reduces consumption and income inequality, but later increases inequality through growing differences in wealth accumulation. For a comparable aggregate expansion, these distributional effects are substantially larger than those generated by a conventional interest-rate cut.
This paper brings quantitative easing into the HANK framework by showing how asset purchases affect the economy through household portfolio liquidity rather than only through long-term interest rates. It connects the portfolio-balance channel of QE with the heterogeneous-agent literature on MPCs, incomplete markets, monetary-policy transmission, and inequality.