Did the American Dream Kill the Social Safety Net? Evidence from Experiment and Survey Data

Abstract

This paper examines the claim that housing wealth reduces support for social insurance. Homeowners who rely on their housing wealth for retirement or view their home as their private rainy day fund may become less supportive of social insurance programs that benefit all strata of society. However, it is also known that the effects of material self-interest on policy preferences are cognitively narrow, and well-identified studies show no effect of wealth on attitudes toward income redistribution. This paper presents two studies that explore the effect of home values on support for social insurance. A survey experiment, in which information about historical home price appreciation (HPA) is randomly assigned to respondents, finds homeowners informed that historical HPA has been high are about 8 percentage points more likely to prefer existing Social Security arrangements to a privatized Social Security scheme, compared to those informed that HPA has been low. I find no statistically detectable effect on various measures of support for social insurance among renters. An analysis of ANES panel data finds that homeowners who experience HPA in the highest quartile are neither more nor less likely to support status quo spending on Social Security than those in the middle quartiles. However, those who experience HPA in the lowest quartile are about 12 percentage points less likely to be satisfied with the status quo than all other homeowners. The evidence suggests that housing wealth gains, rather than reducing support for Social Security, increase preference for the status quo. I organize these results by drawing on existing findings in political behavior and behavioral economics, in particular research on narrow framing, mental accounting, and prospect theory.

Keywords: safety net, housing, wealth, social insurance