Research

Using Norwegian administrative data, we show that households at every level of liquid wealth reduce their risky share after a contractionary monetary policy shock, consistent with reaching for yield. The response is convex in wealth and flattens at the top. Yet portfolio flows move differently: the wealthiest households buy equities while less affluent households sell. We develop a model-free accounting identity that decomposes net purchases of risky assets into portfolio rebalancing, net saving, and capital gains. The decomposition shows that the flow reversal arises because capital losses increase strongly with wealth while the reduction in risky shares flattens at the top. Net saving plays only a small role. Direct estimates of portfolio flows confirm the same reversal over subsequent years. We develop a theory of reaching for yield based on certainty-equivalent portfolio adjustment, in which monetary policy changes households' marginal utility but not their sensitivity to risk. Higher interest rates make households safely richer and reduce their risky share. With countercyclical idiosyncratic disaster risk, the model also generates the convex wealth gradient: the lower sensitivity of marginal utility to resources among wealthy households dampens their reduction in risky shares. Together, these forces explain why monetary tightening reduces risky exposure throughout the wealth distribution while reallocating equity ownership toward the top.

The inter-temporal transmission of fiscal and monetary policy depends on how a shock reshapes the saving distribution across households. I develop an analytical framework built from the household Euler equation that traces saving from optimization to policy multipliers. A first-order perturbation shows that marginal propensities to consume satisfy a recursion, and yields a three-step method for the decomposition of consumption response to any shock. Aggregating, a heterogeneous-agent economy departs from a representative agent that matches its aggregate propensities to consume and save. The reason is simple: the households who accumulate the saving are not those who would spend it. In a two-period setting, closed-form multipliers reveal a structural asymmetry. For deficit-financed fiscal policy, heterogeneity unambiguously attenuates the multiplier: saving accumulates where it cannot be spent, and taxes fall where they cannot be smoothed. For monetary policy, the same forces cut both ways, so the sign is calibration-dependent.

Recent evidence shows that monetary tightening triggers a contrarian reallocation of equity: mid-wealth households sell while the wealthiest buy. I build a heterogeneous-agent New Keynesian model with endogenous portfolio choice to study the macroeconomic implications of equity flows. Cyclical tail risk accounts for a third of the risky-share response, the flat wealth gradient, and the direction of trading. On aggregate, monetary policy shocks permanently reallocate equity ownership: a contractionary monetary policy shock of 1 percentage point persistently lowers the top-5-percent equity holdings by 2.2 percent, entirely through accumulated flows, which cyclical risk amplifies by 30 percent.

Using data from the Survey on Household Income and Wealth (SHIW), we identify Marginal Propensity to Consume (MPC) and the patience of Italian households, and study the effect of patience level on the cross-sectional correlation between MPC and liquidity level.

I show analytically that uninsurable idiosyncratic income risk raises households' effective risk aversion and can reduce their demand for risky assets. Using Norwegian administrative data, I estimate how idiosyncratic income risk affects equity holdings across the wealth distribution, and I study how macroeconomic policies that insure income risk can indirectly shift household risk-taking.