UC Berkeley

This dissertation studies how behavioral forces shape household financial decision-making. Across three chapters, we combine administrative credit-panel data, structural modeling, and a controlled laboratory experiment to show that departures from the frictionless, fully-informed benchmark of standard models are not peripheral: they govern how households respond to debt relief, how they form beliefs about asset returns, and how they invest in financial knowledge. Taken together, the chapters argue that the behavioral channel is central to understanding, and designing policy around, household finance. The first chapter examines the long-run consequences of temporary debt relief, using the federal student loan repayment pause from 2020 to 2023 as a natural experiment. Exploiting variation in loan eligibility through an instrumental variables strategy applied to the University of California Consumer Credit Panel (UC-CCP), we show that the pause initially reduced delinquency and raised consumption---but that the consumption increase persisted and grew gradually, consistent not with a neoclassical liquidity response but with the slow formation of new spending habits. When payments resumed, delinquency rose sharply across credit categories. A life-cycle model with adaptive reference-dependent consumption reproduces these dynamics and, in counterfactual simulations, shows that a shorter pause would have curtailed habit formation and reduced post-pause delinquency. The chapter establishes a recurring theme of the dissertation: temporary changes in household circumstances can leave durable behavioral imprints. The second chapter shifts from debt management to homeownership, asking how lived experiences of local housing markets shape the decision to buy a home. We develop a life-cycle model in which households form beliefs about both the mean and the volatility of future housing returns from their personal histories, and use the UC-CCP linked to zip code-level Zillow price data to construct individual-level measures of experienced returns and volatility since age 18. Consistent with the model, a one percentage point increase in experienced returns raises the probability of transitioning into homeownership by roughly seven percent, while a one percentage point increase in experienced volatility lowers it by roughly ten percent. Volatility effects are detectable only at the zip-code level and attenuate with age, underscoring that perceptions of risk are formed at a granular, local scale. The third chapter turns to the acquisition of financial knowledge itself. Combining administrative data on UC Berkeley students enrolled in a large personal-finance course with an incentivized online experiment, We document strong positive selection into financial literacy education on baseline ability, family income, and quantitative preparation. A controlled lab experiment identifies the mechanism: low-ability individuals systematically overestimate their starting knowledge and underestimate the benefits of education, producing demand that tracks perceived rather than actual returns. Varying the cognitive cost of the material through AI-based, video, and lecture formats raises aggregate demand but does not repair the selection pattern, suggesting that the binding constraint is belief miscalibration rather than access or difficulty. A common thread runs through the three chapters. Households do not move smoothly along the optimal paths that standard models prescribe: they form habits from temporary windfalls, extrapolate from the local markets they happen to have lived in, and misjudge the returns to knowledge they do not yet have. Each of these departures carries first-order implications for debt policy, housing policy, and the design of financial education. The chapters share a common empirical backbone---linked administrative credit-panel data and a careful mapping from behavioral models to causal estimates---and together they make the case that behavioral mechanisms deserve a central place in the analysis of household finance.