错误估计房屋价值：对家庭财务的影响

Misestimating house values:consequences for household finance Stefano Corradin, José L. Fillat,Carles Vergara-Alert Abstract This study examines the effect of systematic household misestimation of home prices on fi-nancial decisions, including stockholdings, consumption, and asset allocation. Using exogenousvariation in house values, mortgage debt, and homeowner misestimation identified throughdifferences in local housing market characteristics, we find that a $60,000 increase in houseovervaluation (approximately one standard deviation) results in a 1.1 to 1.9 percent decreasein risky stockholdings, a 1.5 to 4.3 percent increase in consumption, and a 1.3 to 2.5 percentincrease in the share of risk-free assets over liquid wealth.The results highlight the need tobetter understand how housing wealth and beliefs about house values affect portfolio choice,spending, and overall household finance. JEL Classification: G11, D11, D91, R21, C61. Keywords: Household finance, Portfolio choice, Housing, Misestimation. Executive Summary For most households that own their primary residence, housing is their most important asset. Itserves as a means of saving and investing, while also providing shelter.The current value of ahouse is a key factor in determining financial decisions for households, including those relating toinvestment portfolios, consumption and savings. Using data on self-reported house values from households that had recently moved and wereparticipating in the U.S. Panel Study of Income Dynamics (PSID), and comparing this informationwith the CoreLogic Home Price Index at zip code level (used as a proxy for objective house marketvalue), we document that approximately half of the households in the PSID sample systematicallymiscalculated the value of their home between 1984 and 2021.While the average misestimationof house value across all owner-occupied households is close to zero (an average undervaluation of7,600 dollars in our sample), the standard deviation is large (59,800 dollars). This paper offers new insights into the economic impact of misestimating house values onhousehold finances.Through a combination of theoretical modelling and empirical analysis, weshow that misestimating house values can lead to substantial changes in portfolio allocation andconsumption. We document a robust negative relationship between house value misestimation andthe share of risky assets in household portfolios.Specifically, an increase in house overvaluationof around 60,000 dollars results in an average decrease of between 1.1 and 1.9 per cent in theallocation to risky stock holdings. This is consistent with the predictions of our theoretical model.Furthermore, our findings suggest that households that overvalue their home are less likely toparticipate in the stock market.We estimate that a one-standard-deviation increase in houseovervaluation results, on average, in a reduction of between 1.5 and 3.59 percentage points in stockmarket participation. These findings highlight a significant difference from standard portfolio choice models, whichusually assume an accurate perception of house value and, consequently, household wealth. 1Introduction Housing represents the most important asset for most households.Therefore, house values playa key role in household decisionmaking, notably about stockholdings and consumption. However,households’ estimates of their own house’s value do not often align with market prices. Althoughthe average house value misestimation across all homeowning households is close to zero ($7,600 ofundervaluation on average in our sample), its standard deviation is large ($59,800). We find that 5percent of homeowners undervalue their house by at least $87,500, while 5 percent overvalue theirhouse by at least $53,000. House value misestimation has been documented for more than half a century.1However,literature on portfolio choice that incorporates housing typically assumes that the value of a housereported by a household aligns with its market value (see, for example, Flavin and Yamashita(2002), Campbell and Cocco (2003), Cocco (2005), Yao and Zhang (2005), Fischer and Stamos(2013), Corradin, Fillat, and Vergara-Alert (2014), Chetty, S´andor, and Szeidl (2017), Vestman(2019) and Chen, Michaux, and Roussanov (2020)). In this paper, we study how house value misestimation affects households’ portfolio and con-sumption decisions.We first develop a simple theoretical framework to show the implications ofincorporating misestimation in the analysis of household choices in the presence of housing.Weshow that overvaluation tends to reduce the demand for risky assets because it increases a house-hold’s exposure to risk and illiquidity. Using household-level data, we then estimate the effects ofmisestimation on portfolio and consumption decisions, finding that more optimistic homebuyershave less exposure to stock holdings. We measure house value misestimation as the difference between the owner’s subj