Workshop on Bankruptcy

May 17–19, 2001
St. Louis, Missouri

Back to Workshop Program

Amanda E. Dawsey and Lawrence M. Ausubel
University of Maryland
January 2001

Preliminary

This copy contains only the Introduction of a preliminary draft.
Full text is available at the following URL:

http://www.bsos.umd.edu/econ/bankruptcy/informal-bankruptcy-jan01.pdf

or on Lawrence Ausubel’s web site:

www.ausubel.com

Abstract

In the economic literature on bankruptcy, the standard methodology is to model the individual’s bankruptcy decision as a binary choice between “bankruptcy” and “no bankruptcy.” We define an additional choice-non-repayment without seeking the formal protection of the bankruptcy system-as informal bankruptcy. Using data from a large credit card issuer, we find evidence that while both lenient exemption laws and garnishment laws increase bankruptcies in the standard model, loose garnishment discourages default in our expanded model, while at the same time more pronouncedly shifting individuals from informal to formal bankruptcy. This result suggests that previous research may substantially understate the degree to which garnishment laws drive defaulting individuals to choose bankruptcy. Moreover, lenient exemption laws increase both formal and informal bankruptcy. We also find that borrowers living in majority black neighborhoods are more likely to choose informal bankruptcy, and less likely to choose bankruptcy, than other borrowers.

We also test whether creditors’ strategic interactions increase bankruptcies. We develop a two-period model of the “creditor’s dilemma,” a popular hypothesis in the legal literature. We examine a testable implication of this model, namely, that increasing the number of creditors (while holding the amount and availability of credit constant) should increase the probability that a defaulting borrower enters bankruptcy rather than a workout, and either decrease or not affect the probability of informal bankruptcy. We find that the number of creditors indeed has a positive and statistically significant effect on the probability of bankruptcy, and, depending on the specification, a significant negative or insignificant effect on informal bankruptcy.

Send comments to:

Amanda E. Dawsey or Lawrence M. Ausubel
Department of Economics
University of Maryland
College Park, MD 20742-7211

dawsey@econ.umd.edu
(301) 405-3474

ausubel@econ.umd.edu
(301) 405-3495

I. Introduction

Personal bankruptcy provides a legal procedure giving individuals protection from their creditors. By filing for bankruptcy, a borrower halts collection activities and often obtains a discharge of debts, shielding some property in the process. However, the formal legal procedure of bankruptcy is not the only available alternative to repayment. Instead, a debtor can simply stop making payments-without seeking any legal protection-and place the burden on her creditors to collect. Especially if the debtor has few attachable assets or is difficult to find, the informal strategy may be extremely effective. We will refer to non-repayment without the benefit of the formal bankruptcy process as “informal bankruptcy.”

While informal bankruptcy has received minimal attention in the economics literature, it is rather prevalent in the real world. According to the 1996 Visa Bankruptcy Survey, some 65.2% of credit card loans were charged off¹ for reasons other than bankruptcy, and this percentage has been relatively constant in recent years.² In the data set of credit card accounts examined in the current paper, 50.7% of all written-off accounts (and 45.8% of credit losses in dollars) were lost without the debtor filing for bankruptcy.³

At first blush, the choice of informal bankruptcy by an individual may seem illogical. After all, the bankruptcy system has been severely criticized for its leniency: “Just as the old welfare system encouraged people not to get jobs and encouraged people not to even think about pulling their own weight, our lax bankruptcy system doesn't even ask people to consider paying what they owe, particularly when they have the ability to pay...Why pay your bills when you can walk away with no questions asked? Why honor your obligations when you can take the easy way out through bankruptcy?”⁴

However, a closer examination reveals that the costs of formal bankruptcy can be quite substantial. While both informal and formal bankruptcy can hurt an individual’s credit record significantly, bankruptcy is generally the more damaging choice.⁵ A poor credit record can interfere not only with an individual’s ability to borrow, it can impose other inconveniences, such as trouble writing checks, renting an apartment, or finding employment.⁶ Bankruptcy stays on a borrower’s record for 10 years, while a debt written off for other reasons is expunged after 7 years.

Other costs include the assets a borrower must forfeit in a bankruptcy proceeding, which, depending on her state, may be considerable. While a few states allow bankrupts to keep substantial property following a Chapter 7 bankruptcy, 24 states set homestead exemption limits of $15,000 or less. Delaware, for example, has no homestead exemption and allows borrowers choosing bankruptcy to keep only $5,000 worth of personal property. Borrowers who file under Chapter 13 submit a repayment plan rather than non-exempt assets, but they are required to pay creditors at least as much as they would receive under Chapter 7. Finally, the legal and court costs involved in filing a bankruptcy, though small relative to other proceedings, may be prohibitive to highly leveraged individuals.

Borrowers choosing informal bankruptcy face the cost of resisting their creditors’ collection attempts, but these costs will depend on the actions lenders are willing and legally able to take. A creditor can garnish a borrower’s wages, but 75% is shielded from garnishment by federal law, %, and 16 states protect a higher percentage. Creditors may also seize property, but they cannot seize property that is exempt from a Chapter 7 bankruptcy proceeding. Before seizing non-exempt property, creditors must obtain a judgment and surpass several other legal hurdles. Finally, creditors can simply demand repayment, but they are limited by the Federal Fair Debt Collection Practices Act,⁷ which restricts credit collections agents from inconvenient or harassing contact with borrowers.

While it would seem that these costs would impact a borrower’s choice between informal and formal bankruptcy, we have no economic studies of informal bankruptcy to support this hypothesis. Economists are not alone in their neglect: informal bankruptcy has also received little attention from lenders and collectors. According to an article in American Demographics, “Demographic analysis has been common for years, if not decades, among companies that offer credit - using sophisticated formulas to figure out who to offer credit cards to. But even the credit card companies haven't expended much energy trying to understand why people stop paying their bills, and how best to persuade them to resume paying.” Where firms have begun to utilize precise statistical methods, the results are largely proprietary and inaccessible to the public.⁸

The study informal bankruptcy is likely to contribute significantly to our understanding of insolvency is several ways. First, informal bankruptcy can clarify the effects of federal and state laws governing the interactions between creditors and insolvent borrowers. The feasibility of collection, and therefore the likely costs of informal bankruptcy, clearly vary according to state laws governing garnishment and exemptions. These laws have been examined in the context of formal bankruptcy alone, but they should affect the borrower’s choice of informal bankruptcy. We test the impact of these statutes using a nested logit model, in which a borrower first chooses between repayment and delinquency, and if delinquent, chooses among informal and formal bankruptcy, and delayed repayment. We find that garnishment laws affect both the probability of repayment and the choice between informal and formal bankruptcy, effects which are conflated in the dichotomous bankruptcy model. Garnishment shifts individuals to repayment from non-repayment, while at the same time shifting individuals from informal bankruptcy into bankruptcy. This result suggests that the possibility of garnishment following a default deters borrowers from choosing not to repay their loans, but having defaulted, garnishment increases the likelihood they will choose bankruptcy. The large and significant impact of garnishment laws found in dichotomous models may actually underestimate the degree to which garnishment influences individuals to choose bankruptcy after a default. On the other hand, it appears strict⁹ bankruptcy exemptions do not significantly deter default, but do decrease the probability of bankruptcy following default.

Second, a model which includes informal bankruptcy can present a more complete picture of the level of insolvency in some groups. Race has been shown to be statistically and economically insignificant in standard models,¹⁰ and we also find that race does not have a significant effect in a dichotomous model. However, in a model that includes informal bankruptcy, we find that borrowers living in majority African American neighborhoods are less likely than other borrowers to choose repayment and less likely to choose formal bankruptcy, which results in the insignificant parameter estimate in our dichotomous model. Instead, these borrowers are far more likely to choose informal bankruptcy. This result suggests that analyses of formal bankruptcy alone may overlook the real problem of financial distress among racial minorities, who may be less likely to avail themselves of the protections of the formal legal system.

Finally, the inclusion of informal bankruptcy allows an examination of the validity of the “Creditor’s Dilemma” hypothesis, a prevalent theory in collections law.¹¹ The claim of the this literature is that self-interested creditors pursuing repayment often impose negative externalities on other creditors, which leads to inefficient outcomes. Inefficiency stems from externalities in the creditor’s collection decision: outside of bankruptcy, creditors’ payoffs depend crucially on their diligence relative to other creditors, so their own collections necessarily impose a negative externality on other lenders. An implication of the creditor’s dilemma is that lenders will seize a borrower’s assets and garnish her wages rather than attempting a workout, even if a workout would allow her to pay more of her total debt. This model of creditor behavior only applies when a borrower has defaulted on multiple loans but has not declared bankruptcy, which would forestall collections efforts, i.e. when the borrower is in informal bankruptcy. A clear implication is that increasing the number of creditors should increase the probability a borrower chooses bankruptcy rather than a workout, and either decrease or not affect the probability of informal bankruptcy. Our results support the creditor’s dilemma as an appropriate model of strategic interactions among creditors and borrowers.¹²

One reason informal bankruptcy has been neglected is a lack of appropriate data. Though some data on losses and charge-offs is available, because individuals can charge off any number of loans under bankruptcy, determining which are due to bankruptcy and which are not is difficult in aggregated data. We are able to overcome several difficulties faced by previous researchers by using an unusual data set collected by a large credit card issuer. The data includes a rich set of variables on the borrowers’ loan histories and behavior, as well as their zip code, from which we obtain additional demographic and geographic information. We can also control for the creditor’s own selection criteria, which should mitigate concerns over supply and selection effects.