Do Homeowners Mark to Market? A Comparison of Self‐Reported and Estimated Market Home Values During the Housing Boom and Bust

Abstract

This article examines homeowners' self-reported values in the American Housing Survey and the Health and Retirement Study from the start of the recent housing price run-ups through recent price declines. We compare ZIP-Codelevel market-based estimates of housing prices to those derived from homeowners' self-reported values. We show that there are systematic differences which vary with market conditions and the amount of equity owners hold in their homes. When prices have fallen, homeowners systematically state that their homes are worth more than market estimates suggest, and homeowners with little or no equity in their homes state values above the market estimates to a greater degree. Over time, homeowners appear to adjust their assessments to be more in line with past market trends, but only slowly. Our results suggest that underwater borrowers are likely to understate their losses and either may not be aware that their mortgages are underwater or underestimate the degree to which they are.It has been widely reported that the value of homes in the United States fell by almost one-third between the middle of 2006 and the middle of 2009. Estimates suggest that the value of homeowners' equity in their homes fell by more than 50% over this period (Brown et al. 2010). Faced with such massive reductions in wealth, households may change their consumption, investment and labor supply decisions and they may be more likely to walk away from their mortgages, as they find themselves owing more than their house is worth. However, behavioral responses to wealth changes and negative home equity positions, including strategic default, are likely triggered by a homeowner's individual perception of changes in her or his property value rather than by 2 Chan et al.any externally estimated change in its market price. If homeowners overstate the value of their homes in declining markets, then we should observe less dramatic shifts in behavior than expected.This article examines homeowners' assessment of their home values and draws out implications for expected changes in behavior. We discuss possible mechanisms and behavioral biases that may drive differences between selfreports and market values, and our empirical analysis aims to demonstrate a reduced form effect. We build on the existing literature in several respects. First, we examine and compare homeowners' self-reported values in two national, longitudinal surveys: the American Housing Survey and the Health and Retirement Study. Second, we follow households from the start of the recent housing price run-ups through recent price declines and explore how the discrepancy between owners' reported values and market estimates varies across the market cycle and with levels of home equity. Third, in contrast to prior studies, we use restricted, geocoded versions of these datasets, which identify the neighborhood in which each household resides, allowing us to compare homeowners' self-reported values to the estimated market value of their homes, formed by inflating the reported origin...

Self‐Reported vs. Market Estimated House Values: Are Homeowners Misinformed or Are They Purposely Misreporting?

“…Even after including control variables and MSA fixed effects (columns (2) and (3)), the coefficient on the estimated negative equity dummy variable remains significant and large. These results are consistent with Chan, Dastrup and Gould ().…”

“…Using the geocoded Panel Study of Income Dynamics (PSID) from 1985 to 2013, this study first confirms the results in Chan, Dastrup and Gould (). We find that majority of homeowners classified as underwater according to the market estimated house value report their house value significantly higher than the market estimate.…”

“…Chan, Dastrup and Gould () suggest the level of home equity can also affect how households report their house value. Using the American Housing Survey and the Health and Retirement Study, the study tests how the level of home equity affects how homeowners report their house value.…”

“…Chan, Dastrup and Gould () only compares how the reported and the estimated house value differs between those with LTV≥80% and those with LTV<80%.…”

The Accuracy of Senior Households’ Estimates of Home Values: Application to the Reverse Mortgage Decision

“…The study closest in the literature to ours is Chan, Dastrup and Ellen (CDE, ) who also use data from senior households in the HRS. An important difference in the studies is the method of assigning a home's market value.…”