The effect of personal bankruptcy exemptions on investment in home equity

Abstract

Homestead exemptions to personal bankruptcy allow households to retain their home equity up to a limit determined at the state level. Households that may experience bankruptcy thus have an incentive to bias their portfolios toward home equity. Using US household data for the period 1996–2006, we find that household demand for real estate is relatively high if the marginal investment in home equity is covered by the exemption. The home equity bias is more pronounced for younger and less healthy households that face more financial uncertainty and therefore have a higher ex ante probability of bankruptcy. These results suggest that homestead exemptions have an important bearing on the portfolio allocations of US households and the extent to which they insure against bad shocks.

Introduction

Individuals who file for personal bankruptcy according to Chapter 7 of the US bankruptcy code can generally retain some assets. Specifically, at the state level there tend to be exemptions for certain asset classes up to specific thresholds. The main exemption is the homestead exemption, which enables the filer to retain home equity in his primary residence (“homestead”) up to the exemption amount.¹ The homestead exemption ranges from $0 in Maryland to an unlimited amount in eight US states, including Florida and Texas, in 2006. Personal bankruptcy is quite common in the US, with about one million Chapter 7 filings in 2009, and homestead exemptions therefore frequently apply.

Limited liability, as implied by the ability to declare personal bankruptcy, provides households with some consumption insurance against possible negative income and wealth shocks. Zame (1993) shows that contingent debt repayment, implicit in the possibility of bankruptcy, can be welfare improving in a world of incomplete contracting, even if it implies a more limited ability to borrow and a reduced ability to smooth consumption over time.² Homestead exemptions allow households to emerge from bankruptcy with positive net worth. As shown by White (2011), the effect of exemptions is to further insure household consumption in the face of economic shocks and limit the ability to borrow due to higher interest rates.³

The homestead bankruptcy exemption potentially affects household portfolio choice, because a household needs to have home equity to benefit from the implicit consumption insurance. This paper provides an empirical and a theoretical investigation of the impact of the homestead exemption on investment in home equity as part of the household portfolio allocation and on home ownership.

The estimation uses household level data from the Survey of Income and Program Participation (SIPP) of the US Census Bureau. This data source provides information on wealth allocation and a host of personal and household characteristics for approximately 30,000 households. The homestead exemption is found to have an economically significant effect on a household’s investment in home equity. Specifically, the percentage change in home equity as a share of the percentage change in wealth is estimated to be 25% higher if the household’s home equity is below the exemption rather than above it.

The impact of the exemption level on home equity investment is estimated to be stronger for households with relatively young and unhealthy household heads, as these could face heightened income risk and a correspondingly higher probability of personal bankruptcy. A positive impact of the homestead exemption on home equity investment is confirmed by instrumental variables estimation where we use the historical exemption level in 1920 as an instrument for the more recent state-level exemption level. We also estimate a Heckman two-stage selection model, where in the first stage households decide on home ownership. This approach yields a consistent impact of the homestead exemption on home equity investment.

We find only a weak relation between the homestead exemption and home ownership. In probit regressions, in particular, we fail to find a significant impact of the homestead exemption on home ownership, except for households residing in states with unlimited homestead exemptions. This could reflect that households wishing to purchase a home on account of a high risk of personal bankruptcy are thwarted by a lack of mortgage financing necessary to complete the purchase.

The empirical work is motivated by a two-period model of the allocation of wealth between home equity and another asset category in the presence of a homestead exemption and major expense risk. This expense risk takes the form of an uninsurable medical expense, although it can easily be reinterpreted as any type of uninsurable income or other economic risk. The model implies that marginal household investment in home equity as related to wealth jumps down as home equity reaches the exemption level. This is corroborated in the empirical work.

Several empirical papers provide evidence that bankruptcy exemptions may insure consumption. Grant and Koeniger (2009), in particular, find evidence that the variability of household consumption growth at the US state level is negatively related to the homestead exemption in the state, consistent with an insurance role for bankruptcy exemptions. In addition, Mahoney (2012) shows that higher exemptions reduce households’ demand for health insurance, suggesting that the wealth insurance offered by exemptions reduces the need for explicit health insurance.⁴ Bankruptcy protection potentially also makes owning a business with unlimited liability less risky. Consistent with this, Fan and White (2003) find that the probability that a household owns a small business is higher in states with unlimited exemption than in other states.

Gropp et al. (1997) empirically investigate how exemptions affect aggregate credit to households. They argue that the protection offered by exemptions increases household demand for credit, while it reduces the supply of credit. They find that the net impact on credit is negative for less-well-off households, but it is positive for high-asset households.⁵ Fay et al. (2002) examine how bankruptcy exemptions affect the household bankruptcy decision, and they find that the financial gain that households can attain by filing for bankruptcy, as affected by the exemptions, is a main determinant of the bankruptcy decision.

There is also related a literature on the determinants of home ownership. Li (1977) relates home ownership to household characteristics such as the age of the household head, income and family size. King and Leape (1998) jointly consider the home ownership decision and the resulting household portfolio share in a general study of household portfolio allocation of US households and find that both home ownership and investment in owner-occupied housing respond positively to increases in wealth.

Recent work on housing and portfolio composition has recognized that housing is special because it is an asset as well as a durable consumption good, and because adjustments to housing wealth imply large transactions costs. Flavin and Yamashita (2002) consider the optimal household portfolio under the assumption that the household is constrained to live in the house that is owns and show that this implies that housing introduces considerable portfolio risk, especially for younger households with low net worth. Cocco (2005) provides empirical evidence that house price risk crowds out stockholdings, and that this crowding out is stronger for households with low net worth. Using data from the SIPP survey, Corradin et al. (2014) estimate a model of optimal housing wealth adjustment where house price movements are predictable and there are housing adjustment costs.

Homestead bankruptcy exemptions also set investments in home equity apart from other investments. To our knowledge, our paper is the first to investigate the empirical impact of homestead exemptions on investment in home equity and on home ownership. Homestead exemptions are found to provide bankruptcy protection especially to households that can be expected to need this, such as households that report poor health and low wealth. This protection, however, comes at a cost of biasing household portfolios toward real estate. Our findings inform the policy debate about the desirability of homestead exemptions, and contribute to the literature on the effect of personal bankruptcy on household portfolio choice.

The remainder of this paper is organized as follows. Section 2 discusses the role and evolution of exemptions in the US system of personal bankruptcy. Section 3 presents a simple two-period model of optimal investment in home equity in a world with bankruptcy exemption and major expense risk. Section 4 discusses the data, and Section 5 presents the empirical results. Section 6 concludes.