Who Files for Personal Bankruptcy in the United States?

• Published date: 01 December 2019
• Author: Jonathan D. Fisher
• Date: 01 December 2019
• DOI: http://doi.org/10.1111/joca.12280

WINTER 2019 VOLUME 53, NUMBER 4 2003

JONATHAN D. FISHER

Who Files for Personal Bankruptcy in the United States?

Who les for bankruptcy in the United States is not well understood.

Previous research relied on small samples from surveys or a small

number of states from administrative records. Using over ten million

administrative bankruptcy recordslinked to the 2000 Decennial Census

and the 2001–2009 American Community Surveys, I document who

les for bankruptcy.Compared to the US population, bankruptcy lers

are middle income, more likely to be divorced, more likelyto be black,

more likely to be veterans, less likely to be immigrants, and more likely

to have a high school degree or some college. Filers are more likely to

be employed. The bankruptcy population is aging faster than the US

population as a whole. Lastly,using pseudo-panels I study what happens

in the years around bankruptcy. Individuals are likely to get divorced

in the years before bankruptcy and then remarry. Income falls before

bankruptcy and rises after bankruptcy.

Bankruptcy protection is part of the social insurance system. Bankruptcy is

social in that it is paid for by risk-adjusted interest rates on debt (Berkowitz

and White 2004; Gropp, Scholz, and White 1997). Bankruptcy is insurance

in that it provides debt forgiveness and asset protection in the face of

negative events (Sullivan, Warren, and Westbrook 2000). Most social

insurance programs in the United States target fairly narrow groups with

strict eligibility requirements, usually depending on income or earnings.

Personal bankruptcy protection, on the other hand, is available to “any

person who owes debts,”1and 75% of households have debt (Bricker et al.

2014). White (1998) estimated that at least 15% of all households in 1992

would have experienced an increase in net worth in bankruptcy.

Bankruptcy is an important part of social insurance system. In

2015, 795,000 consumer debtors led for personal bankruptcy, with

1. Fifty-fth Congress, Session II, Chapter 541, Chapter 1, Section 4.

Jonathan D. Fisher (jonathan.sher@stanford.edu) is a Research Scholar at Stanford Center on

Poverty and Inequality, Stanford University. I want to thank the editor, associate editor, and the three

anonymous reviewers.I also want to thank Tal Gross, Matt Notowidigdo, and Jialan Wangfor providing

the administrative bankruptcy records. Thanks to Nathaniel Johnson for research assistance. This

research was carried out at the U.S. Federal Statistical Research Data Center facility. The results and

conclusions of the paper are those of the author and do not indicate concurrence by the Census Bureau.

These results have been screened to avoid revealingcondential data.

The Journal of Consumer Affairs, Winter 2019: 2003–2026

DOI: 10.1111/joca.12280

Copyright 2019 by The American Council on Consumer Interests

2004 THE JOURNAL OF CONSUMER AFFAIRS

approximately $47 billion of debt forgiven (U.S. Courts 2015). Debt for-

giveness amounts to an average transfer of future income equal to $59,000

per household, not counting the interest that would have accrued had the

debt not been forgiven. In comparison, Medicaid covered approximately

72.8 million people and spent about $461 billion in 2013, for an average

government outlay of $6,300 per recipient (U.S. Department of Health

and Human Services 2014).

Bankruptcy provides not only a large transfer, but it has also become

a more important component of the social insurance system. In 1980,

there were 287,564 lings, and annual lings steadily increased until they

reached over 1.5 million in 2004. After the passage of the Bankruptcy

Abuse and Consumer Protection Act (BAPCPA) in 2005, bankruptcy

lings decreased. Several elements of BAPCPA may have contributed to

the decrease in the number of lings, namely the increase in the cost of

ling for bankruptcy in attorney fees and court fees (Gross, Notowidigdo,

and Wang 2014) and the means test meant to push some debtors to the less

generous chapter 13.

While research has investigated the causes and consequences of l-

ing for personal bankruptcy, the demographic prole of who les for

bankruptcy is less understood. Research that reports some demographics of

bankruptcy lers often comes to different conclusions. For example, using

data from 1989 to 1995 Fisher (2005) nds that mean income in 1991 dol-

lars is about $32,000, while Sullivan, Warren, and Westbrook (2000) use

lers from 1991 and nd mean income around $20,000. Fisher (2005) nds

that 44% of lers are terminal high school graduates while Sullivan, War-

ren, and Westbrook (2000) nd only 32% of lers with a terminal high

school degree. Bauchet and Evans (2019) examine the determinants of

bankruptcy during the Great Recession using the panel component of the

2007–2009 Survey of Consumer Finances. They nd some of the same

correlations I do here, but most coefcients are statistically insignicant in

their model, most likely because they only have 60 bankruptcy lers and a

total sample of 3,800. Emmons and Ricketts (2016) study the demography

of loan delinquency, but bankruptcy and delinquency comprise two differ-

ent groups. Households do not need to be delinquent on debt in order to

le for bankruptcy, and some households will be delinquent but never le

for bankruptcy.

My approach is closest to the approach taken in Lefgren and McIntyre

(2009). They use zip-code level ling rates and examine the correlation of

demographics and the ling rate. They do not know who in the zip code

actually led but instead nd that zip codes with more black individuals

have a higher bankruptcy rate. I improve on their results by knowing the