is the personal bankruptcy system bankrupt? - core

Business Review Q1 2002 31www.phil.frb.org

OBY LORETTA J. MESTER

Is the Personal Bankruptcy SystemBankrupt?

ver the past few years, several attemptshave been made to reform the U.S.bankruptcy system, to help stem perceivedabuses of the system. In this article,

Loretta Mester outlines the components of reformproposals. She then looks at the empirical researchon bankruptcy to evaluate the rationale forreforming the system and the effectiveness ofproposed changes.

Neither a borrower nor a lender be;For loan oft loses both itself and friend,And borrowing dulls the edge of husbandry.

PoloniusAct 1, Scene 3Hamlet by William Shakespeare

Over the past several years,Congress has attempted to passlegislation to resolve perceived problemsin the personal bankruptcy system inthe U.S. Although it has not proposedanything as drastic as Poloniusrecommended, Congress has proposedseveral significant changes to thecurrent system. In the latest try, separate

Loretta Mester issenior vice presidentand director ofresearch at thePhiladelphia Fed.

bills were passed in the House (HR 333)and in the Senate (S 420) in March2001, but Congress adjourned beforereconciliation of the bills could becompleted. Legislation is again beingconsidered this year, but regardless ofthe outcome, the debate about theU.S. personal bankruptcy system isunlikely to be resolved anytime soon.Indeed, a number of studies provideevidence on the rationale for changingthe current system, on whether reformis necessary, and on whether theproposed revisions will have theintended effect.

After reviewing the currentpersonal bankruptcy system and theproposed changes, we�ll discuss someof the findings of these recent studies.These studies do shed some light on thedebate and cast some doubt that theproposed changes will yield benefitsas significant as intended. They alsosuggest further research is necessaryto resolve all the issues.

CURRENT PERSONALBANKRUPTCY SYSTEMIN THE U.S.

As Joseph Pomykala discussesin his interesting article, the word�bankruptcy� has two roots. �Bancarotta� is Latin for broken board. Inmedieval Italy, �creditors wouldbreak the workbenches of defaultingmerchants over the merchants� heads.��Banqueroute� is French for debtorson the lam (route), as bankruptcywas considered an act of debtorfraud. As Pomykala points out, beforethe mid-18th century, bankruptcy wasconsidered a crime, and in England,certain bankrupt debtors were subjectto capital punishment. The U.S.modified English law to be less harsh.For example, the PennsylvaniaBankruptcy Act of 1785 allowed forthose convicted of bankruptcy to beflogged while nailed by the ear to apillory, after which the ear would becut off. (Of course, how much morelenient this was is clearly debatable.)

Bankruptcy protection hasbeen part of U.S. federal law since 1898.Indeed, Article I, Section 8 of the U.S.Constitution authorizes Congress toenact �uniform Laws on the subjectof Bankruptcies� (see the article byLeonidas Mecham). The structure ofthe current bankruptcy system wasestablished in the Bankruptcy ReformAct of 1978. The idea is to allow a �freshstart� (within limits) for honest peoplewho, often through unfortunatecircumstances beyond their control,have gotten into trouble with debt andto allow for creditors to be repaid inan orderly fashion with the debtor�savailable assets.

CORE Metadata, citation and similar papers at core.ac.uk

Provided by Research Papers in Economics

https://core.ac.uk/display/6648817?utm_source=pdf&utm_medium=banner&utm_campaign=pdf-decoration-v1

32 Q1 2002 Business Review www.phil.frb.org

The bankruptcy provisionsallow for a sharing of risk betweenborrowers and creditors, offering someinsurance to borrowers if they findthemselves unable to repay their debts.The insurance gives consumers whoseincome may be low today but isexpected to rise in the future theconfidence to borrow now to pay forconsumption. This raises consumers�economic well-being. If things go asplanned, they will repay their debts.If some adverse event, like a job loss,prevents them from repaying, theycan file for bankruptcy and protecttheir future income from creditors.Bankruptcy procedures better enableconsumers to smooth their consumptionover time, thereby increasing economicefficiency. The bankruptcy system alsoprovides a joint debt-collection systemfor a debtor�s creditors (see the CBOstudy for a nice review of personalbankruptcy fundamentals). Abankruptcy system that is too harshwould lower economic welfare bydiscouraging borrowing and risk sharing.However, a system that is too lenientand allows debtors to escape fromtheir commitments too easily can hurteconomic efficiency by causing creditorsto restrict the supply of credit and raiseits cost to creditworthy borrowers.

The federal bankruptcy courtsadminister the bankruptcy system. Allbankruptcy cases are filed in thesefederal courts. There are 94 bankruptcydistricts in the U.S. and its territories,each with a court. Pennsylvania isbroken into three districts: eastern,central, and western, while New Jerseyand Delaware are each separate andsingle districts. The courts in all threestates (along with the District of theVirgin Islands) are part of the thirdfederal circuit. A bankruptcy case isoften overseen by a court-appointedtrustee.

Under current law, individualsconsidering bankruptcy can fileunder Chapter 7 or Chapter 13 of

the bankruptcy act.1 Chapter 7,sometimes called straight bankruptcy,is liquidation � a filer hands over hisassets (with some exemptions) to thetrustee, who then sells the assets anduses the proceeds to repay the debtor�screditors. The remaining debts (withsome exceptions) are then discharged� that is, wiped clean � and thedebtor retains control of his or herfuture income.2 In many cases thereare few assets available to repaycreditors, and most unsecured debt,such as credit card debt, is not repaidin bankruptcy (see the CBO study).In other cases, a debtor may want tokeep control of an asset, like a car,that is pledged as collateral againsta loan. In this case, the debtor can�reaffirm� the debt � the debtor andcreditor agree that the debtor will paythe creditor all or part of the debt,even though the debtor has filed forbankruptcy, and the creditor will notrepossess the property. A person can filefor bankruptcy under Chapter 7 everysix years.

Chapter 13 involves adjustmentof debts of an individual with regularincome. Under this chapter, some debtsare reduced, but then debtors andcreditors devise a plan by which thedebtors repay their remaining obligationsout of their future incomes. Repaymentis made in installments over a three-year period (which the court can

extend to five years). The debtor mustrepay creditors at least as much as theywould have received under Chapter 7,and claims entitled to priority must bepaid in full. The debtor cannot takeon any new debt without the trustee�sapproval, and the debtor�s securedand unsecured debt must be less thancertain specified limits to allow himor her to file under Chapter 13. Inexchange, debtors retain more of theirassets than they would under Chapter7. At the end of the repaymentperiod, any remaining unpaid debt isdischarged. Generally, someone canfile for bankruptcy under Chapter 13as often as he or she wants (exceptthat, in some cases, the filing cannotbe within 180 days of dismissal of aprevious case).

As part of the filing undereither chapter, the debtor submits a listof his assets, income, liabilities, creditors,and debts. After a debtor files undereither chapter, an automatic stay stopscreditors from collecting on unpaiddebts. This prohibits creditors fromfiling lawsuits against the debtor forrepayment, trying to garner the debtor�swages, or making telephone callsdemanding repayment (see Under-standing the Federal Courts).

WHY THE CALL TO REFORM?Bankruptcy protection is

designed to help people get out fromunder the burden of excessive debt andto get a fresh start. Some people, though,point to abuses of the system: Debtorswho actually have the ability to repaysometimes escape their obligations.Knowing this escape route exists maygive borrowers an incentive to take onmore debt, to the extent that lenders arewilling to supply them credit. Of course,lenders should respond to a lenientbankruptcy law that permits manydebts to be discharged by restrictingcredit or raising the cost of credit so thatonly the better credit risks could borrow.A bankruptcy system that is too

1 They may also file under Chapter 11, butthis is rare. Chapter 12, which is similar toChapter 13, applies only to family farmersin financial distress. See Report 106-49 onS. 625, U.S. Senate, and Mecham for fullerdescriptions of the U.S. personal bankruptcysystem.

2 According to Mecham, 18 categories ofdebt cannot be discharged under Chapter 11;there are fewer restrictions under Chapter 13.Certain types of tax claims, debts for spousalor child support or alimony, and debts forwillful and malicious injuries to person orproperty are examples of nondischargeabledebts.


lenient would lead to economicinefficiencies whereby the supply ofcredit was too restricted and its costtoo high.

Over the years, changes havebeen made to the system to try to limitthe scope for abuse. For example, in1984, judges were permitted to dismissChapter 7 cases if they thought grantingrelief would constitute a �substantialabuse� of the bankruptcy code. Butthe term �substantial abuse� was notdefined, and creditors and trustees werenot allowed to present evidence to thejudge in a particular case on whetherrelief should be viewed as substantialabuse (see Report 106-49, Senate).

Filings Have Increased inRecent Years. One factor pointed outin the most recent round of legislativeconsideration of the bankruptcy systemwas the substantial increase in thenumber of filings that began in the1980s. Total bankruptcy filings,including personal and business filings,hit a record 1.43 million in the yearended June 1998, and they have beenvery high since then, with 1.39 millionfilings in the year ended June 2001(Figure 1).3 Indeed, the 400,000 filingsin the second quarter of 2001 was themost ever for a three-month period.Over 97 percent of these filings arepersonal as opposed to business, and 70percent of personal filings per year aretypically Chapter 7 filings (Figure 2).4

FIGURE 1

Annual Number of Bankruptcy Filings(12-month periods ending in June)

FIGURE 2

Number of Personal Bankruptcy Filings,Total and by Chapter(12-month periods ending in June)

Note: Shaded areas represent economic recessions.Source: Bankruptcy data from Administrative Office of the U.S. Courts.

3 The number of personal filings beforeand after 1979 are not directly comparablebecause the Bankruptcy Reform Act of 1978allowed for spouses to file a joint petition forbankruptcy protection. See the CBO studyfor further discussion.

4 Note that some of the personal filingscould actually represent business failuresbecause some small businesses are fundedby the personal credit lines of their owners.See Appendix A of the CBO study forfurther discussion of the data on personalbankruptcy filings.

1379 1352

1240

1349

985 969

864

951

398375382394

1998 1999 2000 2001

Annual number of filings in thousands

1600

1400

1200

1000

800

600

400

200

0

Total Personal Filings

Chapter 7 Personal Filings

Chapter 13 Personal Filings

Source: Administrative Office of the U.S. Courts.


Figure 3 shows the number of personalbankruptcies per thousand households(the bankruptcy rate) in the threestates in the Third Federal ReserveDistrict (Delaware, New Jersey, andPennsylvania) and in the U.S.5 Asyou can see, these numbers rangedbetween nine and 13 bankruptciesper 1000 households in 2001. In otherwords, in 2001, 0.9 to 1.3 percent ofhouseholds filed for bankruptcy in thenation and in our three states. (Notethat there is much wider variation inthe personal bankruptcy rate acrossthe other states in the U.S. In 2001,Tennessee, which had 24.5 filings perthousand households, had the highestrate; Iowa, which had 4.1 filings perthousand households, had the lowestrate. See the Table.)

The number of personalbankruptcy filings began acceleratingin the 1980s, with an especiallylarge increase between 1995-98.6

For example, from 1961 to 1980, thepersonal bankruptcy rate rose, onaverage, about 3 percent per year.Since 1980, the average increase inthe personal bankruptcy rate has beenalmost 8 percent per year, with anespecially sharp increase of 14 percentper year between 1995 and 1998 (Figure4). We can look at the increase anotherway: In 1980, there was one personalbankruptcy filing for every 336 house-holds in the U.S.; in 2001, there wasone personal bankruptcy filing for every78 households.

Some of the rise in bank-ruptcies in the 1980s can be attributed tobad economic times: There was a shortrecession from January 1980 to July

FIGURE 3

Personal Bankruptcy Rate, Total and by Chapter(12-month periods ending in June)

FIGURE 4

Personal Bankruptcy Rate andGrowth in Personal Bankruptcy Rate(12-month periods ending in June)

Note: Sum of Chapter 7 and Chapter 13 does not equal total because there are a fewpersonal business filings under Chapter 11. Personal bankruptcy rate is the number ofpersonal bankruptcy filings per thousand households.Sources: Administrative Office of the U.S. Courts and U.S. Census.

Note: Shaded areas represent economic recessions.Personal bankruptcy rate is number of personal filings per year per 1000 households.Sources: Bankruptcy data from Administrative Office of the U.S. Courts; householddata from the Bureau of Census, www.census.gov.

5 Delaware has about 300,000 households,New Jersey about 3.1 million households,Pennsylvania about 4.8 million households,and the U.S. about 106 million households.

6 In contrast, business filings, whichincreased in the early 1980s, fell back inthe latter half of the 1980s and in the 1990s.


1980 and a long one from July 1981 toNovember 1982 (recessions are shownby shaded bars in the figures). But therapid rise in the bankruptcy rate in the1990s is more difficult to understand,since this was a period of very goodeconomic conditions � economicgrowth averaged 3.2 percent per yearin the 1990s, and the unemployment ratehad fallen to 4 percent by the end of thedecade. Even this is not unprecedented:The rate of bankruptcy filings roserapidly in the mid-1980s in the midstof an economic expansion, and it hasrisen in other periods of economicexpansion as well. (Note that the riseis not necessarily a bad thing, as itaccompanied an increase in creditavailability to households.)

Households did increase theirborrowing in the 1990s, and the rise inhousehold debt-service burdens � thatis, required payments on mortgagesand other consumer debt as a percentageof disposable income � in that decademay explain part of the rise in thebankruptcy rate (Figure 5, see nextpage). Yet the debt-service burden wasat comparable levels in the mid-1980s,and the bankruptcy rate was muchlower.7 So factors other than debt-service burdens appear to play somerole in the decision to file.

The most recent increase infilings in the first half of 2001 mightbe the result of pending legislationto change the bankruptcy system, aspeople contemplating bankruptcy mayhave accelerated their filings to get inunder the old rules. Still, the fact thatthe bankruptcy rate rose in the late1990s during good economic times andremains high is taken by many as anindication that the system is beingabused by people who actually have

TABLE

Personal Bankruptcy Rate by State inYear Ended June 2001(Number of Personal Bankruptcies per 1000)

PersonalTotal No. of Bankruptcy Rate

Nonbusiness Households (Nonbus. FilingsState Filings (2000 Census) per 1000 Hhs)

Tennessee 54,730 2,232,905 24.51Utah 16,915 701,281 24.12Georgia 63,800 3,006,369 21.22Nevada 15,833 751,165 21.08Alabama 36,116 1,737,080 20.79Mississippi 20,561 1,046,434 19.65Arkansas 19,466 1,042,696 18.67Indiana 42,537 2,336,306 18.21Maryland 32,956 1,980,859 16.64Idaho 7,578 469,645 16.14Oklahoma 20,892 1,342,283 15.56Washington 34,087 2,271,398 15.01Louisiana 24,730 1,656,053 14.93Kentucky 23,609 1,590,647 14.84Oregon 19,667 1,333,723 14.75Illinois 66,817 4,591,779 14.55Virginia 38,361 2,699,173 14.21Ohio 61,906 4,445,773 13.92West Virginia 9,630 736,481 13.08New Jersey 39,575 3,064,645 12.91Missouri 27,989 2,194,594 12.75California 143,174 11,502,870 12.45Florida 78,702 6,337,929 12.42Wyoming 2,343 193,608 12.10Kansas 12,554 1,037,891 12.10Hawaii 4,712 403,240 11.69Arizona 22,036 1,901,327 11.59Rhode Island 4,690 408,424 11.48New Mexico 7,420 677,971 10.94Michigan 41,251 3,785,661 10.90Colorado 16,921 1,658,238 10.20Pennsylvania 47,708 4,777,003 9.99Montana 3,578 358,667 9.98Washington, DC 2,410 248,338 9.70North Carolina 30,215 3,132,013 9.65Nebraska 6,382 666,184 9.58Wisconsin 19,624 2,084,544 9.41Delaware 2,730 298,736 9.14New York 63,642 7,056,860 9.02Texas 66,290 7,393,354 8.97Connecticut 11,144 1,301,670 8.56South Carolina 12,868 1,533,854 8.39Maine 4,198 518,200 8.10Minnesota 15,216 1,895,127 8.03North Dakota 2,021 257,152 7.86South Dakota 2,228 290,245 7.68New Hampshire 3,545 474,606 7.47Massachusetts 16,676 2,443,580 6.82Vermont 1,580 240,634 6.57Alaska 1,361 221,600 6.14Iowa 9,662 2,336,306 4.14

7 There have been periods � for example,between 1988 and 1991 � when debt-serviceburden and filings moved in oppositedirections.


the wherewithal to repay their debts.This view has led many observers tobelieve that the bankruptcy systemneeds to be revamped and has ledto proposed legislation to changethe system. Whether changes to thebankruptcy system will have much ofan effect on the rate of filings dependsboth on what changes will be enactedand whether the bankruptcy systemitself has encouraged filings.

BANKRUPTCY REFORMLEGISLATION

Although there have beenmany attempts to pass legislation overthe past several years, bankruptcyreform legislation has yet to be signedinto law. In March 2001, the Houseand Senate passed their own versions ofbankruptcy reform legislation (HR 333and S 420); similar bills were passed theprevious year by the 106th Congress,and hearings were held in 1997 by the105th Congress. Last year, Congressadjourned before reconciliation of thebills could be completed, but bankruptcyreform legislation is again beingconsidered this year. While theversions passed by the House and theSenate in 2001 differed in some ways,those differences have narrowed aslegislation has worked its way throughseveral sessions of Congress. In lastyear�s bills, there was general agreementon basic aspects of reform. Here I�llreview eight proposed changes tothe bankruptcy system. The first fiveof these reforms favor creditors bylimiting the benefits to debtors fromdeclaring bankruptcy. The last threereforms might be considered debtorprotections.

(1) Chapter 7 MeansTesting. The bankruptcy system wouldbe changed to what proponents of thebills call a �needs-based� system. If adebtor has sufficient income to repay alarge part of his or her debts, he or shecould not pursue Chapter 7 liquidationbut only a Chapter 13 repayment plan.

A means test would be applied todetermine which debtors would beforced into Chapter 13 and how muchdebt would have to be repaid over afive-year period. Debate has centeredon whether means testing is necessary.Creditors favor such testing, sayingthat some debtors have abused thecurrent bankruptcy system. Consumergroups say only 3 percent to 5 percentof Chapter 7 filers would have torepay some of their debt under theproposed means tests, and they arguethat reform is not necessary, since therate of filings dropped in 1999 afterthe record rate in 1998. As discussedbelow, arguments for and againstmeans testing are in dispute.

The 2001 bills barred Chapter7 filing if, after living expenses andthe cost of other necessities, the debtorcould afford to repay at least $100 permonth over a five-year period. Thebills generally used the standards the

IRS uses to figure out living expensesfor people owing back taxes, but thebill specified the use of actual costs ofother necessities (for example, childcare, union dues, and so forth). Someincome, such as Social Security andwar crimes compensation, would beexcluded from the calculations. Thoseearning less than the median incomein the applicable state would qualifyfor Chapter 7 regardless of their abilityto repay.

Under current law, if a debtorfiles under Chapter 13, the repaymentperiod is three years unless the court, forcause, extends it to a maximum of fiveyears. This provision would remain thesame for debtors whose family income isless than the median family income inthe applicable state. However, forfamilies with higher income (whowould be forced to file under Chapter13 by the means test), the repaymentplan would be extended to five years.

FIGURE 5

Personal Bankruptcy Rate andDebt-Service Burden

Note: Shaded areas represent economic recessions.Source: Bankruptcy data from Administrative Office of the U.S. Courts; debt-serviceburden data from the Federal Reserve Board. Debt-service burden is household requiredpayments on mortgage and consumer debt as a percentage of disposable personal income.Quarterly personal bankruptcy rate is quarterly filings per thousand households. Note, sum of quarterly filings equal annual filings. Correlation between debt-service burdenand quarterly personal bankruptcy rate was 0.84 from 1980 to 1987, �0.71 from 1988 to1991, and 0.88 from 1992 to 2000.


(2) NondischargeableDebts. Bankruptcy courts currentlypresume that if a debtor bought morethan $1000 in luxury goods or servicesor took $1000 in cash advances inan open-ended credit plan within 60days before filing for bankruptcy, thesedebts were fraudulently incurred,and so they are not dischargeable.Any debt incurred to pay an existingnondischargeable debt is non-dischargeable if it was incurred withthe intent of not repaying. Non-dischargeable debts include certaintaxes, family support obligations, anddebts arising from fraud. The proposedlegislation would make more debtsnondischargeable in bankruptcy.

The 2001 bills extend thedefinition of fraudulently incurred(and, therefore, nondischargeable)debt. In the House bill, the threshold forpresumption of fraudulent purchases ofluxury goods was lowered to $250 within90 days of filing for bankruptcy, and thethreshold for cash advances was loweredto $750 within 70 days of filing. TheSenate bill agreed with the House,except that the threshold for luxurygoods was lowered to only $750.

(3) Homestead Exemption.Many states allow a debtor to keeppossession of his or her residence (up tosome limit) when filing for bankruptcy;the residence would not be available topay off creditors. Five states (Florida,Iowa, Kansas, South Dakota, andTexas) put no limit on the exemptionfor a primary residence; other statesare quite restrictive (for example,New Jersey allows no homesteadexemption).8 One of the majordifferences between the Houseand Senate bills passed last year wastheir treatment of the homesteadexemption. The Senate version would

put a federal cap of $125,000 on theexemption for a primary residence.This cap would apply to all states.The House bill maintained states�ability to opt out of the federal limitand reestablish an unlimited or otherexemption by passing legislation.9

Both bills lengthened the time fromsix months to two years that a debtormust live in a state before being able toclaim that state�s exemption.

(4) Lien Strip-Down.Currently, debts are secured only tothe value of the collateral, with anyremainder treated as unsecured debt.Thus, a debtor could pay the amountof the collateral�s current market valueand keep the collateral � this is a strip-down. For example, a debtor couldpurchase a car, file for bankruptcy, andkeep the car by paying off the value ofthe car, even though this is less than theamount he contracted to pay in theloan. Debtors who bought furniture,which has little resale value, are able tokeep the furniture by repaying the lowresale price rather than paying off themuch larger debt. Consumer groupsargue that changing the rules to allowcreditors to repossess the collateralwould force reaffirmations, whereindebtors agree to pay certain debts andnot discharge them in bankruptcy. The2001 legislation sought to limit strip-

downs. The House bill barred strip-downs for a motor vehicle acquired bythe debtor within five years prior to filingbankruptcy and for other personalproperty bought within one year prior tofiling. The Senate bill differed from theHouse bill in setting the period for motorvehicles at three years prior to filing.

(5) Repeat Filings. Currently,debtors can file for Chapter 13 bank-ruptcy at any time (except, in somecases, within 180 days of a priordismissal). Debtors can file for Chapter7 bankruptcy six years after a dischargein bankruptcy. The legislation wouldhave limited repeat filings. Both billswould have barred a Chapter 7 filingwithin eight years of prior discharge.The Senate bill would have barred aChapter 13 discharge within three yearsof a prior discharge under a Chapter 7,11, or 12 filing, and within two years ofa prior discharge under Chapter 13. TheHouse bill was harsher, disallowing aChapter 13 discharge within five yearsof any prior discharge.

(6) Reaffirmation of Debts.Consumer advocates say creditors arepressuring debtors into reaffirming debts� in other words, pressuring them intosaying that they will pay the debt andnot discharge it in bankruptcy. Suchreaffirmations are supposed to be filedwith and approved by the bankruptcycourt. But Sears was recently heldliable in a class action based onreaffirmations of debt that were notfiled with bankruptcy courts. The caseinvolved Sears� pressuring debtors intoreaffirming their debts with Searsrather then having them discharged inbankruptcy. Sears admitted to criminalfraud and paid a fine of $60 million.

The bills passed in 2001sought to stem abusive reaffirmationagreements by mandating that certainspecific disclosures be made in writingto the debtor, explaining the termsof the underlying credit agreementand the reaffirmation. The billsasked the attorney general to

8 These data are as of January 1, 2000. Seefootnote 13, page 490 of Report 107-3, Houseof Representatives.

9 Prior to 1978, there was no federalexemption; the states controlled exemptions.The Bankruptcy Reform Act of 1978 setfederal exemptions for certain assets,including personal goods, tools of trade,autos, and homesteads. But individual stateswere allowed to opt out and set their ownlimits, and by 1983, all of them had. TheBankruptcy Reform Act of 1994 raised thefederal exemption level and tied it to theConsumer Price Index starting in 1998. Asof January 1, 2000, the federal exemption was$16,150 per debtor; 35 states had set theirown exemption levels and did not allow thefederal exemption; and the remaining statesallowed debtors to choose between their stateexemption or the federal exemption when filingfor bankruptcy. See footnote 13, page 490 ofReport 107-3 Part 1, House of Representativesand the study by Jon Nelson.


10 Adverse selection in the credit-card markethas also been documented in my paper withPaul Calem. The fact that credit-card ratesare very sticky and don�t tend to come downwhen other rates do is partly attributable tothis adverse selection problem.

enforce prohibitions against abusivereaffirmations.

(7) Consumer CreditDisclosures. Some argue that creditorshave led consumers into bankruptcy bymisleading them about the true cost ofcredit. The legislation sought to makecredit card companies disclose moreabout the consequences to the borrowerof making only the minimum monthlypayment.

The bills would have amendedthe Truth-in-Lending Act to requiredisclosures on credit-card bills. Creditcard issuers would have had to providegeneric examples of the consequencesof making only minimum payments onbills in terms of how long it would taketo pay off the debt. Issuers also wouldhave had to give a toll-free numberwhere holders could get specificinformation about repayment scenariosfor their own accounts. Enhanceddisclosures for home-equity loans,introductory loan rates, and latepayment deadlines and penaltieswould have been required. Both billswould have barred the termination ofa credit card just because a customerhadn�t incurred a finance charge.

In testimony before theHouse in March 1999, Federal ReserveGovernor Edward Gramlich said theBoard of Governors wondered whetherrepeated disclosures to consumers mightcreate �information overload.� He saidthe Board does not generally favorlaws that restrict a creditor�s discretionto determine which accounts ortransactions it deems as economicallyviable. The Board�s view is that ifcreditors are terminating accounts ofcustomers who use their credit cardsonly for transactions purposes, they aredoing so because they consider theseaccounts unprofitable.

(8) Credit Counseling.Provisions are included to addressthe issue of unsophisticated borrowers.These provisions are intended to ensurethat the debtor made a good-faith effort

to negotiate a repayment plan withhis creditors on his own. However, therequirement might also delay filingsuntil debtors are unable to come up with a repayment plan in Chapter 13.Both bills required the debtor to haveundergone credit counseling within 180days before filing under Chapter 7 orChapter 13.

As suggested by this overview,while the House and Senate bills passedin 2001 differed in some of their details,there was general agreement on majoritems of reform. The real question iswhether these reforms are necessaryand, if so, whether they will be effective.

EVIDENCE RELEVANT TOBANKRUPTCY REFORM

There have been severalstudies of personal bankruptcy. Theimplications of the empirical work todate is mixed as to whether the proposedbankruptcy reform is needed andwhether it will have the intended effect.Partly at issue is the extent to whichborrowers respond to the incentivesprovided by the bankruptcy law �borrowing more than they otherwisewould and declaring bankruptcy eventhough they could eventually repay theirdebts � or, alternatively, whether theydeclare bankruptcy when they face anunexpected hardship that makes itimpossible for them to repay their debts.

There are four main issues.First, I�ll discuss empirical evidenceon the source of the recent rise inbankruptcies and the extent to whichmarket forces place limitations on thenumber of bankruptcies. If these forcesare effective, reforms are less needed.Similarly, I�ll discuss empirical

evidence on the relationship betweensocial forces (stigma) and the numberof bankruptcy filings. A lessening inthe effectiveness of these social forceswould help support arguments forreform. Next I�ll discuss evidenceon the extent to which the currentbankruptcy system is being abused.High levels of abuse favor thereformers. Finally, I�ll discuss evidenceconcerning the efficacy of proposedreforms. Even if one believes thecurrent system needs reform, it is notclear that the proposals will yield thedesired effect.

(1) Market Forces. In a studydone in 2000, Lawrence Ausubel arguedthat market forces have tempered someof the recent acceleration in personalbankruptcy filings and that legislationis unnecessary. As the bankruptcyrate increased, lenders responded bytightening their credit standards, thusleading to the decrease in the numberof bankruptcies between 1998 and 1999.In Ausubel�s view, if there ever was a�bankruptcy crisis,� it is self-correcting.

In congressional testimony in1998, Ausubel argued against the means-test approach because, in his view, theimmediate cause of the record numberof bankruptcies is the high level ofhousehold debt, which he attributes inpart to aggressive lending tactics. In a1999 study, Ausubel found that inrandomized trials on preapproved credit-card solicitations conducted by a majorU.S. issuer of credit cards, offers thatincluded higher interest rates and feestended to attract riskier borrowers withhigher delinquency, charge-off, andbankruptcy rates than offers with betterterms. In other words, issuers face aso-called adverse selection problem.10

The implications ofthe empirical work[on bankruptcy] todate is mixed.


Ausubel places some of the blame forhigher bankruptcies on the creditorswho have issued the debt. He favorsthe approach of earlier proposedlegislation that would restrict theclaims of lenders who caused a debtor�sratio of unsecured debt to income toexceed 40 percent. He also favors atime priority in bankruptcy: unsecuredlenders would be repaid in the orderin which they lent, with the earliestlender repaid first and the latestlender repaid last, giving the laterlenders more incentive to monitorthe borrower�s credit position.

Joanna Stavins reviewedstudies that have shown riskier borrowershave gotten better access to credit-card loans over time, noting that therise in credit-card borrowing in themid-1990s has coincided with theincrease in bankruptcy filings. Usingdata from the Terms of Credit CardPlans, a survey of about 200 of thelargest bank credit-card issuersconducted twice a year by the FederalReserve Board, Stavins providedempirical evidence that credit-cardissuers that offer higher rates andfees in order to compensate for higherrisk do tend to experience higherdelinquency rates (measured by thefraction of outstanding credit-cardloans 60 days or more overdue), afinding similar to Ausubel�s. But unlikeAusubel, Stavins found that theseissuers did not seem to have highercharge-off rates (the fraction ofoutstanding credit card loans thatare written off). Indeed, her empiricalresults showed that banks that chargedhigher rates and fees earned higher netincome from credit cards than banksthat charged lower fees. This impliesthat at least over the period covered(1990-99), when the economy was ingood shape, it was profitable for issuersto extend credit to riskier borrowers.Whether that would continue to betrue in an economic downturn remainsto be seen.

The relationship betweendebt levels and bankruptcies is morecomplicated than the studies byAusubel and Stavins might suggest.There is no doubt a strong correlationbetween debt burdens (measured bythe debt-to-income ratio, debt-to-assets ratio, or debt-service burden)and number of bankruptcies, since ahigher debt burden means a negativeshock can have a more severe effecton a household. However, we alsoknow that in the aggregate, debt anddebt-to-assets seem to increase afterhouseholds see their incomes rise andexpect their future incomes to rise.Debt seems to facilitate growth ratherthan inhibit it. Debt levels havebeen expanding not only because ofaggressive lending but also because of

the expanding economy and becausetechnological advances have allowedcreditors to offer loans to moreborrowers at a lower cost (see my1997 article on credit scoring).

(2) Stigma. Another factorthat may have contributed to theincrease in the bankruptcy rate isthe decreased social stigma associatedwith declaring bankruptcy. Certainly,bankruptcy continues to have anegative connotation. It can harm aperson�s reputation, and it can make itmore difficult to gain access to creditin the future. Federal law allowscredit bureaus to continue to reporta bankruptcy filing in the person�scredit report for up to 10 years afterthe filing. A study by David Mustoshowed that this does restrict theperson�s access to credit. His workusing credit-file data from 1994-97showed that when the bankruptcy flag

was removed from the report, the morecreditworthy past filers initiated newcredit relationships, especially high-limit credit cards, at a much faster ratethan normal � evidence that the flagwas a constraint on their gettingcredit.11 Nevertheless, some of thenegative effects from filing may havedeclined over time. Filing forbankruptcy may be more acceptedthese days, since it has become morecommon. There are other costsassociated with filing that may havefallen as the number of filings hasrisen. For example, it is easier to findinformation on how to file (the formsand the information are readilyavailable on the Internet); morepeople have experienced bankruptcy,so there are more people who can give

advice; and there are more bankruptcylawyers competing for business.

In an interesting study, DavidGross and Nicholas Souleles assembleda panel of over 25,000 individualcredit-card accounts, chosen to berepresentative of all open accounts inJune 1995. They studied the behaviorof these accounts for the next 24months or until they first defaulted

Another factor that may have contributed tothe increase in the bankruptcy rate is thedecreased social stigma associated withdeclaring bankruptcy.

11 Similarly, Stavins presented some data fromthe Survey of Consumer Finances for 1998indicating that among those who have ever filedfor bankruptcy (8.51 percent of respondents),the average level of credit-card debt is largestfor those who filed nine or more years ago. Butshe also found that the average credit-card debtfor someone who filed one or two years ago washigher than for those who filed three to nineyears ago. Stavins posited that this might bebecause once someone files under Chapter 7,he or she cannot file again for six years, soissuers might feel relatively safe lending in theinitial period after a filing.


or were closed in good standing, inan attempt to see whether the recentincrease in bankruptcies is betterexplained by supply effects or demandeffects. That is, did lenders increasethe supply of credit to less credit-worthy borrowers, who account forthe increase in bankruptcy filings?Or, even after researchers control forcreditworthiness, have people becomemore willing to default over time?Has their demand for bankruptcyincreased? According to Gross andSouleles� estimates, riskier borrowers� for example, those with lower creditscores, larger credit card balances, andsmaller monthly payments � aremuch more likely to default. Defaultrates were also higher for people livingin states where unemployment washigher, house prices were lower, andfewer residents had health insurance.12

The authors documented thatthere was an increase in credit to riskierborrowers. But increases in credit limitsand other changes in risk-compositionexplain only a small part of thesignificant increase in default ratesbetween 1995 and 1997. They foundthat all accounts, even those with thesame risk characteristics, age, and othereconomic fundamentals, became morelikely to default over the sample period.And this increase in the probability ofbankruptcy � about 0.06 percentagepoint per month between the start ofthe sample period in June 1995 to itsend in June 1997 � is comparable tothat which would occur if the creditscore of every account in the samplewere reduced by one standard deviation,which would be a very large increasein the overall riskiness of the sample.While not conclusive, this evidence

is consistent with the stigmahypothesis, that is, that a decline inthe cost of declaring bankruptcy �either the social, legal, or information-gathering costs � is largely responsiblefor the increased level of filings.13

(3) Abuse. A basic premiseof bankruptcy reform legislation is thatthe current system is being abused bypeople who declare bankruptcy whenthey can still afford to repay theirdebts. Here the evidence is verymixed, and the results depend onthe various assumptions made aboutthe type of means test that would beenacted and the way different types ofdebt would be handled. A 1997 studyby John Barron and Michael Statenpublished by the Credit ResearchCenter at Georgetown Universityestimated that if all secured debtwas reaffirmed, about 32 percentof Chapter 7 debtors in their samplecould repay about 31 percent of theirnonhousing, nonpriority debts. Thiswould be an average payment per filingof $3570 over five years. The studywas based on a sample of 3798 familieswho filed for bankruptcy in 13 majorU.S. cities during the spring andsummer of 1996. It was not a

nationally representative sample,and a review by the GeneralAccounting Office disputed someof the study�s findings on a numberof methodological grounds. For one

thing, the study used the informationdebtors provided at the time of filingabout their income, expenses, anddebts without verification and assumedthat the filer�s ratio of income toexpenses remained constant overthe five-year repayment period.

Visa and MasterCard havefunded several studies, including threeby Ernst &Young (by Tom Neubig andco-authors). The latest of thesestudies, published in March 1999 andbased on a nationally representativesample of 1997 filings, estimated that10 percent of Chapter 7 debtors wouldbe affected by a means test for abilityto pay either 25 percent or more ofunsecured debts or $5000 over fiveyears. This would yield $3 billion indebt recovery over five years. Butthese results assumed that the debtorsremained in payment plans for thefull five years and that their incomesrose as fast as their expenses and debt(Report 106-49, Senate, p. 88).

Marianne Culhane andMichaela White got significantlydifferent results from the Ernst & Youngstudies. Their results were based on adifferent sample of bankruptcy filings(from 1995) and different assumptionsabout, among other things, how long

Increases in creditlimits and other changesin risk-compositionexplain only a smallpart of the significantincrease in defaultrates between 1995and 1997.

12 Gross and Souleles also documenteda seasoning effect: They found that theprobability of delinquency rises from thetime the account is opened until it is abouttwo years old, then the probability falls.

13 There is a debate in the legal literatureabout whether an economic modelingapproach or a sociological approach isthe appropriate methodology for studyingbankruptcy decisions, especially when stigmais the factor being investigated (see MichelleWhite�s 1997 article for an interestingdiscussion). The economic modeling approach,favored, for example, by White (1997), assumesthat consumers act to maximize their welfareand, therefore, may act strategically when itcomes to filing for bankruptcy, as we�ll discussbelow. The sociological approach, favored, forexample, by Teresa Sullivan, Elizabeth Warren,and Jay Westbrook (1989) and Rafael Efrat(1998), assumes that people file for bankruptcywhen their financial problems become severeenough that they can no longer handle theirdebt; it is not something they anticipate orplan for, and they do not act strategicallyregarding filing for bankruptcy. My trainingputs me in the economic modeling camp;hence, the studies I review here largelyfollow that approach.


debtors remain in their payment plansand debtors� automobile expenses.They estimated that 3.6 percent ofChapter 7 debtors could afford torepay some of their debts, with totalrecoveries of $450 million over fiveyears. Because of their different sample,even when Culhane and Whitechanged their assumptions to thoseused by Ernst &Young, they stillestimated significantly lower recoveries� $930 million � than Ernst & Young.

One of the drawbacks ofmany of these types of studies is thatthey assume lender behavior wouldremain the same after bankruptcyreform. But if lenders lend even moreaggressively, the number (and cost)of bankruptcies might increase afterreform. The studies also assume thatborrower behavior would remain thesame, but models suggest that thisneed not be the case.14

For example, one potentialdrawback of the means test as proposedin the legislation is that shiftingdebtors with incomes above a certainthreshold from Chapter 7 to Chapter 13essentially imposes a high tax on futureearnings, since Chapter 13 requiresdebtors to use some portion of theirfuture earnings to repay their debt.Michelle White�s 1999 study pointedout that this sets up perverse incentives,reducing debtors� willingness to workand even giving them an incentive toquit their jobs to avoid the tax. Sheand Hung-Jen Wang have proposedcombining Chapters 7 and 13 sothat debtors filing for bankruptcywould have to use their assets andtheir future earnings, after certainexemptions, to repay their debts. Theirsimulations suggest that this systemwould not have deleterious effects ondebtors� incentives to work.

Wenli Li also developed ageneral equilibrium model of bank-ruptcy chapter choice and showed thatindividuals with fewer assets but higherincome were more likely to chooseChapter 7, while those with moreassets and lower income were morelikely to choose Chapter 13 and theywork less.15 The author�s theoreticalanalysis of proposed reforms indicatedthey will affect borrowers differently,depending on their level of wealth andincome. For example, a means test thatshifts filers into Chapter 13 would hurtthe ones with few assets and mediumincome levels. Anticipating this, thosefilers with a higher probability offiling for Chapter 13 will reduce theirborrowing but also not work as hard.This has important implications fortrying to assess the economic benefitsof bankruptcy reform or the amountof debt that borrowers would be ableto repay in Chapter 13.

Strategic behavior. A corollaryof the premise that people are abusingthe bankruptcy system by filing whenthey can repay is that people actstrategically when filing for bankruptcy.But the empirical evidence on just howstrategically people are behaving whenthey file is mixed. �Forum shopping� isone strategy. The exemption levels forpersonal bankruptcy vary widely acrossstates. For example, as discussed earlier,a single filer receives no exemption for

a home in New Jersey but an unlimitedexemption in Texas. So there is aneconomic incentive to move to astate with a higher exemption beforedeclaring bankruptcy.

In their provocative 1999study, Ronel Elul and NarayananSubramanian, using data from thePanel Study of Income Dynamics(PSID), estimated that about 3 percentof all moves to states with higherexemptions are driven by bankruptcyconsiderations. This percentage doublesfor moves made by households �atrisk� for bankruptcy (that is, withan estimated probability of filing forbankruptcy equal to the average filer).

Using data from the Universityof Michigan�s Panel Study on IncomeDynamics (PSID) for 1984-1995, ScottFay, Erik Hurst, and Michelle Whitefound evidence that borrowers respondto the incentives to file for bankruptcy.They measured these benefits as the

amount of debt that is dischargeableunder bankruptcy less any assets overthe exemption level, which wouldhave to be given up under bankruptcy.The authors found that for each $1000increase in benefits, the probability ofa household�s filing rises, on average,by 0.021 percentage point, which wouldimply a 7 percent increase in filings peryear.16 To see what this effect means,

A corollary of the premise that people areabusing the bankruptcy system by filing whenthey can repay is that people act strategicallywhen filing for bankruptcy.

14 In addition, these studies do not accountfor the administrative expenses of imposing ameans test.

15 In an empirical study, Ian Domowitz andRobert Sartain found that Chapter 13 filersmore often tend to be married and employedand have higher income and higher equity-to-debt ratios than Chapter 7 filers.

16 One drawback of the study is that only 254households included in the PSID had filedfor bankruptcy. The rate of filings in the PSIDwas only half of the national rate, suggestingthat PSID households underreported theirbankruptcy filings. See Fay, Hurst, and Whiteand the CBO study for further discussion ofthis point.


consider that there were about580,000 personal filings in the yearended June 1989, about the middlepoint of the period covered in thestudy. A 7 percent increase in filingswould have meant 40,000 more filingsin 1989. If the size of the effect werethe same in 2001, an increase of $1000in benefits would have added 94,000filings to the 1.35 million personalfilings in 2001.

Culhane and White alsostudied strategies and found thatsophisticated debtors could avoidbeing classified as having the ability torepay under a means test by taking onmore debt or increasing charitablecontributions. Note that if such strategicbehavior is the rule, then estimates ofcost savings under means testing thatdo not account for these reactions willbe overstated.

Other studies reviewed byMichelle White (1998b) did not findthat personal bankruptcy rates aresignificantly related to the incentivesto file.17 Indeed, the real conundrummight not be why the bankruptcy rateincreased so much in the 1990s, butwhy it didn�t increase more18 and whymore people haven�t moved to Texasand Florida where the homesteadexemption is unlimited. Two reasonssuggested by White (1998a) include

the fact that sometimes creditors donot take legal action against borrowerswho default, so borrowers have lessincentive to file for bankruptcyprotection, and that borrowers maywant to preserve the option to file inthe future, so they refrain from filingimmediately. Another possibility is thatthere is indeed stigma associated withfiling, which deters households fromdeclaring bankruptcy.

(4) Efficacy of RepaymentPlans. The reform proposals assumethat Chapter 13 repayment planswork. But a 1994 study by theAdministrative Office of the U.S.Courts found that 36 percent ofdebtors who voluntarily enteredChapter 13 repayment plans between1980 and 1988 completed their plans.The study did not indicate why 64percent of the plans failed. Only 14percent of all Chapter 13 cases wereconverted to Chapter 7. This findingon the efficacy of repayment plansis important and contrary to theassumption in many of the studies thatfound that bankruptcy reform wouldlead to significant cost savings, sincethe studies assume that debtors remainin the repayment plans the full fiveyears.

SUMMARYCongress continues to try

to pass legislation to reform thebankruptcy system. While the rate ofbankruptcy filings has risen over thepast several years, the reason for thatincrease is still debatable and thatmeans the rationale for reform isdebatable too. Some proponents ofreform argue that people with thewherewithal to repay their debts aretaking advantage of the system andthat significant cost savings would beforthcoming from reform. Othersargue that the real reason bankruptcieshave increased is that the level of debthas increased, perhaps because lendershave encouraged risky borrowers totake on excess levels of debt.

The empirical work onthe causes and incentives to file forbankruptcy and whether the proposedbankruptcy reform will have thedesired effect is decidedly mixed.While this means that proponents oneach side in the debate can find theammunition they need by choosing theright study, it also means that moreresearch is necessary in order to fullyunderstand the bankruptcyphenomenon. BR

17 White (1998b) and �Notes� review theliterature on whether debtors behavestrategically regarding bankruptcy decisions.See also the CBO study.

18 Fay, Hurst, and White found that about 18percent of households would have benefittedfrom filing for bankruptcy over the 1984-94period, and White (1998a) found that 15percent of households would have benefittedfrom filing in 1992. Yet, on average, fewer than1 percent of households filed over these periods.


Ausubel, Lawrence. �Testimony Beforethe Subcommittee on Commercial andAdministrative Law of the Committee onthe Judiciary of the U.S. House of Represen-tative,� hearing on Consumer BankruptcyIssues, March 10, 1998.

Ausubel, Lawrence. �Adverse Selection inthe Credit Card Market,� Working Paper,University of Maryland, 1999.

Ausubel, Lawrence. �Personal BankruptciesBegin Sharp Decline: Millennium DataUpdate,� manuscript, University ofMaryland, January 18, 2000.

Barron, John M., and Michael E. Staten.�Personal Bankruptcy: A Report onPetitioners� Ability to Pay,� Credit ResearchCenter Monograph #33, GeorgetownUniversity, October 1997.

Calem, Paul, and Loretta J. Mester.�Consumer Behavior and the Stickinessof Credit Card Interest Rates,� AmericanEconomic Review, 85 (December 1995),pp. 1327-36.

Congressional Budget Office (CBO).�Personal Bankruptcy: A LiteratureReview,� September 2000.

Culhane, Marianne B., and Michaela M.White. �Taking the New ConsumerBankruptcy Model for a Test Drive: Means-Testing Real Chapter 7 Debtors,� AmericanBankruptcy Institute Law Review (1999),pp. 27-77.

Domowitz, Ian, and Robert L. Sartain.�Determinants of the Consumer BankruptcyDecision,� Journal of Finance 54 (February1999), pp. 403-20.

Efrat, Rafael. �The Moral Appeal ofPersonal Bankruptcy,� Whittier Law Review,20 (1998), pp. 141-67.

Elul, Ronel, and Narayanan Subramanian.�Forum-Shopping and Personal Bank-ruptcy,� manuscript, Wharton School,University of Pennsylvania, July 1999.

Fay, Scott, Erik Hurst, and Michelle J.White. �The Household BankruptcyDecision,� American Economic Review,forthcoming.

Gross, David B., and Nicholas S. Souleles.�An Empirical Analysis of PersonalBankruptcy and Delinquency,� Reviewof Financial Studies, 15, 2002, pp. 319-47.

Li, Wenli. �To Forgive or Not to Forgive:An Analysis of U.S. Consumer Bank-ruptcy Choices,� Federal Reserve Bankof Richmond Economic Quarterly, 87(Spring 2001), pp. 1-22.

Mecham, Leonidas Ralph. BankruptcyBasics. Administrative Office of the UnitedStates Courts, revised second edition,June 2000.

Mester, Loretta J. �What�s the Point ofCredit Scoring?� Federal Reserve Bank ofPhiladelphia Business Review (September/October 1997), pp. 3-16.

Musto, David K. �The Reacquisitionof Credit Following Chapter 7 PersonalBankruptcy,� Wharton Financial Institu-tions Center Working Paper 99-22,June 1999.

Nelson, Jon P. �Consumer Bankruptciesand the Bankruptcy Reform Act: A Time-Series Intervention Analysis, 1960-1997,�Journal of Financial Services Research(September 2000).

Neubig, Tom, Gautam Jaggi, and RobinLee. �Chapter 7 Bankruptcy Petitioners�Repayment Ability Under H.R. 833:The National Perspective,� Ernst &Young, March 1999,www.ey.com/global/vault.nsf/US/Chapter_7_Bankruptcy_Petitioners_Repayment_Ability_Under_H.R._833:_the_National_Perspective/$file/Report99.pdf.

Neubig, Tom, Fritz Scheuren, Gautam Jaggi,and Robin Lee. �Chapter 7 BankruptcyPetitioners� Ability to Repay: AdditionalEvidence from Bankruptcy Petition Files,�Ernst & Young, February 1998,www.ey.com/global/vault.nsf/US/Chapter_7_Bankruptcy_Petitioners_Ability_to_Repay:_Additional_Evidence_from_Bankruptcy_Petition_Files/$file/feb98_report.pdf.

REFERENCES

Continued on next page


REFERENCES

Neubig, Tom, Fritz Scheuren, GautamJaggi, and Robin Lee. �Chapter 7 Bank-ruptcy Petitioners� Ability to Repay:The National Perspective, 1997,�Ernst & Young, March 1998,www.ey.com/global/vault.nsf/US/Chapter_7_Bankruptcy_Petitioners_Ability_to_Repay:_the_ National_Perspective,_1997/$file/Report97Fin.pdf.

�Notes: A Reformed Model of ConsumerBankruptcy,� Harvard Law Review, 109(April 1996), pp. 1338-56.

Pomykala, Joseph. �Bankruptcy Reform,�Regulation (Fall 1997), pp. 41-78.

Report 106-49 of the Committee onthe Judiciary, Senate, 106th Congress,1st Session, To Accompany S. 625,Bankruptcy Reform Act of 1999, Togetherwith Additional and Minority Views,May 11, 1999.

Report 107-3, Part 1, of the Committeeon the Judiciary, House of Representatives107th Congress, 1st Session, To Accom-pany H.R. 333, Bankruptcy AbusePrevention and Consumer Protection Actof 2001, Together with Dissenting Views,February 26, 2001.

Stavins, Joanna. �Credit Card Borrowing,Delinquency, and Personal Bankruptcy,�New England Economic Review (July/August 2000), pp. 15-30.

Sullivan, Teresa A., Elizabeth Warren,and Jay L. Westbrook. As We Forgive OurDebtors: Bankruptcy and Consumer Creditin America. New York: Oxford UniversityPress, 1989.

Understanding the Federal Courts.The Federal Judiciary, 1999,www.uscourts.gov/UFC99.pdf.

Wang, Hung-Jen, and Michelle J. White.�An Optimal Personal BankruptcyProcedure and Proposed Reforms,�Journal of Legal Studies 29 (January2000), pp. 255-86.

White, Michelle J. �Economic VersusSociological Approaches to Legal Re-search: The Case of Bankruptcy,� Lawand Society Review, 25 (1997), pp. 685-709.

White, Michelle J. �Why Don�t MoreHouseholds File for Bankruptcy?� Journalof Law, Economics, and Organization 14(1998a), pp. 205-31.

White, Michelle J. �Why It Pays to Filefor Bankruptcy: A Critical Look atIncentives Under U.S. Bankruptcy Lawsand a Proposal for Change,� University ofChicago Law Review, 65 (Summer 1998b),pp. 685-732.

White, Michelle J. �Viewpoints: Too ManyIncentives to Go Bankrupt,� AmericanBanker, November 19, 1999.

Continued from previous page

is the personal bankruptcy system bankrupt? - core

Documents