Sullivan, Warren, & Westbrook, , “Folklore and Facts: A Preliminary Report for the Consumer Bankruptcy Project,” 60
Am Bankr. L.J.
11 U.S.C. § 522(a), (d) (1982 & Supp. IV 1986).
In 1981, the year that we have studied, Chapter 13 debtors were required to pay at least as much as their creditors would have received in Chapter 7 liquidation. 11 U.S.C. § 1325(a)(4) (1982). In most bankruptcy courts they were also required to pay enough to show their “good faith” in proposing a payment plan; e.g., Flygare v. Boulton, 709 F.2d 1344 (10th Cir. 1983). Since 1984, debtors in Chapter 13 must also show that their payments will include all their “disposable income” over and above “necessary” living expenses. 11 U.S.C. § 1325(b) (Supp. IV 1986). One study shows that these provisions have had little effect. American Bankruptcy Institute, Perception and Reality: American Bankruptcy Institute Survey on Selected Provisions of the 1984 Amendments to the Bankruptcy Code 73–75 (1987) (“ABI Survey”).
11 U.S.C. § 1328 (1982 & Supp. IV 1986). This is, of course, a very simplified version of the differences between Chapter 7 and Chapter 13. Secured debt adds other complications, so that debtors in Chapter 7 or Chapter 13 are put to the choice of maintaining payments on loans secured by a pledge of collateral (such as a home or car) or risking foreclosure on the property. For those wanting more detail on the legal aspects of consumer choices in Chapter 7 and Chapter 13, see, e.g., E. Warren, & J. Westbrook, The Law of Debtors and Crediton 176–79, 335–69 (Boston, 1986).
A fuller description of our data base will be available in Sullivan, Warren, & Westbrook, As We Forgive Our Debtors (forthcoming 1989) (“SWW, Forgive”). For a summary description, see Sullivan, Warren, & Westbrook, , 60
Am. Bankr. L.J.
293, 336 (1986) (cited in note 1). All tables in this article are based on that data set. The statistical analyses were performed using SPSS-X on an IBM 3081.
House Judiciary Comm. Rep., H.R. Doc. No. 1698, 57th Cong., 1st Sess. 406–7 (1902) (reprinted in Eastmann, Bankruptcy 412 (1903)).
32 Stat. 7970 (1903). Stricter provisions on discharge were adopted, and a series of ocher debtor-creditor compromises were put into place. Eastmann, Bankruptcy 407 (1903); Collier on Bankruptcy xi (7th ed. 1909).
47 Stat. 1467 (Railroad Reorganization) (1932), 48 Stat. 912 (Corporate Reorganization) (1933).
Act of June 22, 1938, ch. 575, § 1, 52 Stat. 893–905 (codified at 11 U.S.C. §§ 3601–3676).
In July 1970 Congress established the Commission on the Bankruptcy Laws of the United States for the purpose of analyzing, evaluating, and redrafting the Bankruptcy Act. Act of July 24, 1970, Pub. L. No. 91–354, 84 Stat. 468.
11 U.S.C. § 101 (1982).
Bankruptcy Amendments and Federal Judgeship Act of 1984, Pub. L. No. 98–353, § 306, 98 Stat. 333. The 1984 amendments also redressed the concerns of labor unions about the treatment of collective bargaining agreements with companies in bankruptcy, which partially explains the political coalition of labor and the credit industry to agree on restricting debtor protections. In 1986 Congress again amended the bankruptcy laws to create a special bankruptcy option (Chapter 12) for family farmers in financial distress. 11 U.S.C. §§ 1201–1229 (Supp. 1988); 100 Stat. 3105–3118 (1986).
81 Cong. Rec. H. 8,647 (daily ed. Aug. 10, 1937) (statements of Rep. Chandler).
As with all intellectual movements, the origins are complex. The premise that debtor behavior followed the predictable cause-and-effect relationship of the economic model surfaced most prominently in legal circles in a seminal exemption-reform article by the eminent bankruptcy scholar Frank Kennedy long before the economists formalized the model. Kennedy, , “Limitations of Exemptions in Bankruptcy,” 45
Iowa L Rev.
445, 447–48 (1960). Evidently Professor Kennedy later concluded that the model was not so clear in predicting debtor behavior. He did not use it extensively in his later work, and as head of the Bankruptcy Commission, in his testimony advocating a single, minimum federal exemption standard, he minimized the expected impact of economic incentives on debtors' actual behavior. Bankruptcy Act Revision: Hearings on H.R. 31 and H.R. 32 Before the Subcomm. on Civil and Constitutional Rights of the Comm. on the Judiciary, 94th Cong., 1st & 2d Sess. 170 (1975–76). Other commentators carried the idea. See, e.g., Note, “Bankruptcy Exemptions: A Full Circle Back to the Act of 1800
Cornell L. Rev.
663, 673–74 (1968) (discussing the use of low exemptions to encourage greater use of Chapter 13, ultimacely rejecting the proposal on grounds other than trouble with the economic analysis); Comment, “Bankruptcy Exemptions: Critique and Suggestions,” 68
1459, 1509 (1959) (arguing for lower property exemptions to force more debtors into Chapter 13).
Meckling, , “Financial Markets, Default, and Bankruptcy: The Role of the State,” 41
Law & Contemp. Probls., 13, 27 (1977). Dean Meckling explained that the purpose of his paper was to use principles widely accepted by economists as “an analytical framework within which to clarify issues, identify more precisely the role which bankruptcy plays, and assess the social consequences of alternative legal structures for bankruptcy.”Id. at 13.
Other economists focused on other aspects of consumer bankruptcy laws, including the expectations about consumers' borrowing behavior and the response of credit markets; e.g., Weston, , “Some Economic Fundamentals for an Analysis of Bankruptcy,” 41
Law & Contemp. Probs.
47 (1977). Although Weston notes the absence of empirical evidence, he also adopts without question a simplistic cause-and-effect description of consumer debtor behavior and consumer credit markets.
Professor Shuchman, among others, forcefully pointed out the lack of empirical verification, but there was little discernible effect on the increasing reliance on the economic model as the basis for policy discussion. See, e.g., Shuchman, , “Theory and Reality in Bankruptcy: The Spherical Chicken,” 41
Law & Contemp. Probs.
66 (1977) (a spirited attack on Meckling's economic analysis challenging the unproven assertions of the model); Moore, , “Foreword” (to a symposium on the economics of bankruptcy reform), 41
Law & Contemp. Probs.
1, 4 (1977) (discussing a conference of economists and law professors, including Dean Meckling, at which questions were raised about empirical assertions). Although Meckling acknowledges that an earlier version of his Law and Contemporary Problems paper (cited in note 15) was presented at the earlier conference, he made no effort in his revised paper to respond to the questions raised about the unproven empirical assumptions. Professor Vern Countryman is another scholar who has long lamented the absence of useful data in the bankruptcy area and who has been wary of economic models with no data to back them up. See Countryman, , “For a New Exemption Policy in Bankruptcy,” 14
Rutgers L Rev.
678, 678 (1960) (discussing the lack of empirical data).
Hearings before the Subcomm. on Civil and Constitutional Rights of the House Comm. on the Judiciary, 94th Cong., 2d Sess., pt. 3, at 360 (1976) (statement of Alvin 0. Wiese, Jr., Chairman of the Subcommittee on Bankruptcy of the Law Forum of the National Consumer Finance Association, and Rovert B. Norris and Vernon L. Evans, General Counsel and Associate General Counsel, National Consumer Finance Association).
For a summary of the general economic arguments, see Moore, , 41
Law & Contemp. Probs.
1, 4–5 (1977).
Vukowich, , “Reforming the Bankruptcy Reform Act of 1978: An Alternative Approach,” 71
1129, 1153 (1983).
11 U.S.C. §§ 1322(b)(1), (2), 1325(a)(5), 1328 (1982 & Supp. IV 1986).
See Kaplan, , “Chapter 13 of The Bankruptcy Reform Act of 1978: An Attractive Alternative,” 28
DePaul L. Rev.
1045, 1058 (1979); Report of the Commission of the Bankruptcy Law of the United States, H.R. Doc. No. 137, 93d Cong., 1st sess., pt. 1, at 174 (1975). The ubiquity of debtors in bankruptcy who could pay was among the empirical assumptions in the debate, an assumption we find largely contradicted by data we discuss in separate analyses in Sullivan et al., Forgive (cited in note 5).
Chapter 13 filings jumped from about 15% of all consumer filings in 1978 to nearly 30% by the year following our study. Calculation from the Judicial Conference of the United States, Director of the Administrative Office of the United States Courts, Ann. Rep. 402, table F3B (1982). Whether the stated goals were met, however, is a very different question. There is a tendency to confuse the filing of a Chapter 13 with the goal of paying back creditors. Our data show that increased Chapter 13 filings by no means guarantee that result.
Several witness appeared before Congress during the hearings to discuss restricting the bankruptcy protections, including the representatives of the American Bankers Associations, the American Retail Federation, the Credit Union National Association, the Shopping Center Merchants Association, the National Consumer Finance Association, and the National Retail Merchants Association. Hearings Before the Subcomm. on Courts, Comm. on the Judiciary, U.S. Senate, 97th Cong., 1st Sess., April 3 & 6, 1983.
See Credit Research Center, Krannert School of Management, Purdue University, Consumer Bankruptcy Study, vol. 1, Consumers' Right to Bankruptcy: Origins and Effects (Purdue University Monograph No. 23) (1982) (“Purdue Study”).
The consumer credit industry “simultaneously commissioned a study of bankrupt debtors, hired a prestigious Madison Avenue advertising firm to publicize the results of the study, and lobbied for federal legislation to remedy the problems that it was certain the study would reveal” (citations omitted). Warren, , “Reducing Bankruptcy Protection for Consumers: A Response,” 72
1133, 1133–34 n.4 (1984). Testimony before Congress hammered home the economic incentive analysis. See, e.g., Hearings before the Subcomm. on Counts of the Senate Comm. on the Judiciary, 97th Cong., 1st Sess. 63–64 (1981) (“1981 Hearings”) (statement of Claude Rice, Alvin Wiese, Jr., and Jonathan M. Landers); id. at 181 (statement of Eldon Hoekstra); id. at 21–23 (statement of Andrew F. Brimmer).
E.g., Kolbenschlag, “Caveat Creditor,”Forbes, July 7, 1980, at 35–36; Sheik, Frons, Hewitt, & Abramson, “A Rush to Personal Bankruptcy,”Newsweek, Aug. 11, 1980, at 41–42; “Why More People Are Going Bankrupt,”U.S. News & World Rep., Nov. 5, 1979, at 76, 79; “Last Chapter? The Federal Bankruptcy Law, Claim Creditors, Should be Rewritten,”Barron's, Nov. 17, 1980, at 4, col. 1; Flanagan, “Debtors File for Bankruptcy at a Record Rate, Spurred by New Code, Recession, Tight Credit,”Wall St. J., June 30, 1980, at 34, col. 1; Rork, “Debtors Escaping Though They Can Pay,”USA Today, Dec. 17, 1982, at 10A, col. 1.
E.g., Vukowich, 71 Geo. L.J. at 1153 (cited in note 21); Boyes, & Faith, , “Some Effects of the Bankruptcy Reform Act of 1978,” 29
J. Law & Econ.
139 (1986) (statistical study purporting to show that bankruptcy filings rose when the 1978 Bankruptcy Code went into effect); Peterson, & Akoi, , “Bankruptcy Filings Before and After Implementation of the Bankruptcy Reform Law,” 35
J. Econ. & Business
95 (1984) (a macroeconomic study of bankruptcy filing rates and various exemption laws which the authors conclude shows that bankruptcies rise as a consequence of liberalization of federal bankruptcy laws); Shepard, , “Personal Failures and the Bankruptcy Reform Act of 1978,”27
J. Law & Econ.
419 (1984) (a statistical study correlating various “government transfer payments” and bankruptcy filings that the author concludes shows a rise in bankruptcy filings related to the adoption of the debtor protections in the 1978 Code). The largest study was the two-volume publication cited in note 26 (Purdue Study) commissioned by the consumer credit industry. This study purports to show that $1.1 billion in debt was discharged by consumer debtors who could have repaid their debt but instead responded to the incentives of the bankruptcy code to declare bankruptcy and “enrich themselves” as often as the law allows. For a detailed discussion of the data from the Purdue Study, see T. Sullivan, Warren, & Westbrook, “Limiting Access to Bankruptcy Discharge: An Analysis of the Creditors' Data,” 1983 Wis. L. Rev. 1091; C. Sullivan, Reply: “Limiting Access to Bankruptcy Discharge,” 1984 Wis. L. Rev. 1069; T. Sullivan, Warren, & Westbrook, Rejoinder: “Limiting Access to Bankruptcy Discharge,” 1984 Wis. L. Rev. 1087 (collectively “Purdue Study Analysis”). Woodward and Woodward summarize the academics' approach: The relationship between bankruptcy filings and provisions of the Reform Act is commonly expressed in terms of “incentive:” attractive provisions of the Act are said to provide debtors with “incentives” to choose bankruptcy. More specifically, if the provisions of the bankruptcy law are made more attractive to debtors, more of them will select that option as a solution to their financial problems. A corollary of this idea is commonly used to deal with rising bankruptcy rates: make the law less attractive to debtors and less of them will choose bankruptcy. [Footnotes omitted] Woodward, & Woodward, , “Exemptions as an Incentive to Voluntary Bankruptcy: An Empirical Study,” 88
309, 311 (1983).
By 1982, 34 states had opted out of the federal exemption scheme. Two more had opted out by 1984. Bankr. Service Lawyers Ed., Code Commentary & Analysis§§ 22:120–22:172 (1988).
Woodward & Woodward, 88 Com. L.J. at 311. See also Woodward, , “Exemptions, Opting Out, and Bankruptcy Reform,” 43
Ohio St. L.J.
335, 345 (1982).
Lacy, , “South Carolina's Statutory Exemptions and Consumer Bankruptcy,” 30
643, 687 (1979). South Carolina opted out two years later. S.C. Code Ann. § 15–41–425 (Law. Co-op. Supp. 1981).
11 U.S.C. § 101 (1982).
11 U.S.C. §§ 707(b), 1325(b) (Supp. IV 1986).
Vuchowich, 71 Geo. L.J. (cited in note 21).
See Boyes & Faith, 29 J. Law & Econ.; Peterson & Akoi, 35 J. Econ. & Bus.; Shepard, 27, J. Law & Econ. (all cited in note 29). See also White, , “Personal Bankruptcy Under the 1978 Bankruptcy Code: An Economic Analysis, 63
1, 18 (1987).
Woodward & Woodward, 88, Com. L.J. at 311 (cited in nore 29)
Id.; Vukowich, 71 Geo. L.J. at 1153 (cited in note 21).
These small differences appear once the sample has been sorted by homeowner status. For example, among the nonhomeowners, there is a significant different (p= 0.02) in the debt-income ratio between the debt-income ratios of Chapter 7 filers (mean value of 2.1) and Chapter 13 filers (mean value of 1.1). Similarly, among the homeowners there is also a difference between the debt-income ratios of Chapter 7 and Chapter 13 filers, although the difference just misses statistical significance (p= 0.051). For homeowners, the differences are also in the predicted direction, with higher ratios reported among the Chapter 7 filers. The point is not that there is no sorting effect, but rather that the sorting effect is so minor.
Purdue Study 88–91 (cited in note 26). For a detailed analysis of this data, see Purdue Study Analysis (cited in note 29).
In SWW, Forgive (cited in note 5), we show, among other things, that a very large proportion of Chapter 13 cases fail–at least two-thirds of those sampled–chiefly because most Chapter 13 debtors are staggering under debt-income ratios that would make it very difficult for them to pay any substantial part of their debts.
In addition to the exemption structure, Congress enacted legal incentives to attract more debtors to Chapter 13. These incentives included a broader discharge in Chapter 13, something attractive only to debtors with problems under the Chapter 7 scheme. 11 U.S.C. §§ 525, 727, 1328. Our interviews with judges and attorneys suggested that the legal (as opposed to economic) incentives to use Chapter 13 (or to avoid Chapter 7) were rarely used. The data permit us to test the direct impact of some of the extra incentives. Some reasons for avoiding Chapter 7 or for entering Chapter 13 could not be tested because the files contained no relevant information. For example, we could not tell if the debtor or attorney feared that a debt might be claimed as a fraud by embezzlement and not be eligible for discharge in Chapter 7 but dischargeable in Chapter 13. Other reasons, however, we were able to examine. Virtually no debtors had filed bankruptcy within the previous six years. The broader discharge in Chapter 13 might be more attractive to persons with educational loans or with intentional tort debts, neither of which can be discharged in Chapter 7 but both of which can be discharged in Chapter 13. Contrary to expectations, petitioners in Chapter 7 and Chapter 13 were equally likely to report educational loans (t= −1.56). Furthermore, Chapter 7 debtors were significantly more likely to report tort debt (t= 3.33, p= .00) of any kind, including both intentional tort debts, which can only be discharged in Chapter 13, and ordinary tort debt, which can be discharged in either chapter. We separately coded data on intentional torts, when the files gave enough information from which we could draw this inference reliably. Seven debtors with pending intentional tort suits were in Chapter 7, and 27 of 32 cases reporting other or undetermined tort cases pending were in Chapter 7. The Chapter 13 forms collect much less information on pending lawsuits than do Chapter 7 forms. Nevertheless, these data suggest that neither educational loans nor torts lead people to file Chapter 13. The circumstances may vary, but many debts involving fraud are not dischargeable in Chapter 7. We could not detect any cases of fraud, nor did we find any creditor objections to a debtor's discharge based on an allegation of fraud. Having lost or destroyed financial records or having transferred property within the preceding year could be an indication of fraud. The data show that debtors who had lost or destroyed records were significantly more likely to be in Chapter 7 (t= 3.82, p= .00). Those who had transferred property were equally likely to be in Chapter 7 or Chapter 13 (t= −0.55). Again, these indicators did not have the expected association with Chapter 13. Chapter 13 can offer some protection to a nonspouse co-debtor, protection not available in Chapter 7. Here we found a split in the expected direction. Chapter 13 petitioners were significantly more likely than Chapter 7 petitioners to have a nonspouse cc-debtor (t= −3.09, p= .000). Relatively few cases were involved, however: 66 cases in Chapter 7 and 71 cases in Chapter 13. Finally, we examined the choice of chapter for debtors who had any one of these conditions, and found that Chapter 7 and Chapter 13 debtors were equally likely to have at least one such condition (t= −0.42). It appears that the additional incentives, to the extent they can be tested directly, have negligible influence on debtor behavior. They certainly have no demonstrable influence on sorting can-pay from can't-pay debtors.
See Boyes & Faith, 29 J. Law & Econ.; Peterson & Akoi, 35 J. Econ. & Bus.; Shepard, 27 J. Law & Econ. (all cited in note 29). Those without statistical studies have also concluded that incentives in the new laws caused the rise in bankruptcy filings. See, e.g., Woodward & Woodward, 88 Com. L.J. at 311 (cited in note 29).
See, e.g., Vukowich, 71 Geo. L.J. at 1153 (cited in note 21), who explains: The exemption policy should be modified by reducing bankruptcy exemptions. Reducing exemptions would have major benefits. First, the number of bankruptcy filings should decline because reducing exemptions would make bankruptcy more costly. Some debtors who now can file in chapter 7, obtaining a discharge and retaining all or most of their assets under the current exemption policy, would not file if they had to surrender some valued assets. These debtors instead will resolve their financial problems outside of bankruptcy, thereby increasing the level of debt repayment. Second, egregious cases, in which large amounts of debt are discharged while debtors retain significant wealth, would be eliminated. Third, and affecting many cases, debtors with nonexempt assets would have a stronger incentive to file in chapter 13, which permits debtors to retain all their property. Fourth, this approach would result in larger payments in chapter 13 to creditors with unsecured claims. Id.
See, e.g., 1981 Hearings 63–64 (cited in note 27) (statement of Claude Rice, Alvin Wiese, Jr., and Johnathan M. Landers); id. at 181 (statement of Eldon Hoekstra); id. at 21–23 (statement of Andrew F. Brimmer). See, e.g., Woodward & Woodward, 88 Com. L.J. at 311: “Perhaps more important, the argument that larger federal bankruptcy exemptions will induce the filing of voluntary bankruptcy petitions has probably influenced the opt-out decisions of the thirty-two states which have thus far opted-out.” See also Woodward and Woodward's text quoted in note 29, summarizing the approach of many academics, an approach with which Woodward and Woodward disagree.
For example, compare the Alabama exemptions with those in California, as summarized in the Collier Bankruptcy Treatise. 7 Collier's 15th 5–18.5; 63–92.11 (1984).
1 Purdue Study 90 (cited in note 26).
E.g., Vukowich, 71 Geo. L.J. at 1152-53 (cited in note 21); Vukowich asserted that lowering exemptions would produce greater creditor repayments and that those with assets would be able to pay.
Total debt-income ratio and total nonmorrgage debt-income ratios do not vary by state (one-way analysis of variance, F= .231, p= .79; F= .63, p= .53.) When we performed a two-way analysis of variance using both state and chapter as independent variables and total debt-income ratio as the dependent variable, we again recorded no significant effects. When we compared total nonmorrgage debt income ratios in the same way, we still found no significant variations by state. The only significant effect was the same one reported earlier between capital Chapter 7 and capital Chapter 13. See note 40. The main effects were significant at p= .05 and the chapter effects at the same level.
The dichotomous dependent variable, choice of chapter, is not too skewed for ordinary least squares regression; the split is about 60–40 in our sample. We considered this an acceptable departure from hornoscedasticity. To check our assumption, the regression equations were rerun using logistic regression (results nor shown) with no resulting change in the substantive conclusions.
See Neustadter, , “When Lawyer and Client Meet: Observations of Interviewing and Counseling Behavior in the Consumer Bankruptcy Law Office,” 35
Buffalo L. Rev.
177, 239–40 (1986). This ground-breaking qualitative study of attorney impact on the consumer hankruptcy process is the sort of work that will be important in devising new quantitative studies.
The variable is dichotomized as married or nonmarrird; where there were no indications of a spouse, the variable was coded as “nonmarried.” This probably understates the number of married petitioners because some single filers may have been married but failed to mention a spouse.
A knowledgeable bankruptcy practitioner can immediately spot problems in the Administrative Office's classifications of the data. The category “business 7” includes everything from the demise of Braniff to the failure of a small Amway distributor. “Petitions filed” may or may not include cases that are reopened, a situation that some clerks reported to be increasing in frequency. Even the information that a case was filed in Chapter 7 may mislead if substantial numbers of cases are filed in Chapter 7 or Chapter 13 but convert to another chapter and are shown in subsequent years as “pending” or “terminated” in those chapters. The category “business 13” highlights these data problems. Because Chapter 13 filings are limited to real live human beings (not corporations), a “business 13” must mean, we suppose, that the Chapter 13 debtor was in business sometime before he filed. The “joint business 7,” which might at first blush suggest the failure of a small, incorporated business and the personal bankruptcy of its owner, must instead refer to the filing by a husband and wife, at least one of whom owned a business. The data are rife with ambiguities that lack detailed and consistently employed explanations.
Administrative Office of the Courts, Ann. Rep. 544–47, table F2A (1981); Ann. Rep., 393–95, table F2A (1982).
This result comes from a series of F-tests with probabilities that exceed our criterion level of p= 0.05.
Ideally, we would like to have included the district information within the correlation matrix and the subsequent regression equations. But data about districts are discontinuous. A person is in District 1 or 2 or 3, etc., but there is no hierarchical ranking of districts, unlike a person's debt, which may be $10,000 or $10,001, so one debt is always more than another. Furthermore, in four districts we had oversampled Chapter 13s, so that predicting choice of chapter using dummy variables for district residence would be erroneous. As a result, we are unable to offer a regression equation using district as a direct predictor of chapter choice. What we have done, however, is to test for the possible effects of groups of districts. The group of oversampled districts did not show a biasing effect, but the group of urban districts did have a significant predictive effect and is included in the regression equation.
These provisions required both the attorney and the court clerk to certify that each debtor had been advised of the Chapter 7/Chapter 13 alternatives. 11 U.S.C. § 342(b) (Supp. IV 1986). See also ABI Survey 73–75 (cited in note 3), which shows that little change resulted from the 1984 amendments.
We examined, for example, multicollinearity with the other predictor variables and found none. The highest correlation between specialist attorney and any other variable (except chapter) was with residence in the state of Texas (r=0.23). This correlation is far below the level at which problems of multicollinearity would be encountered.
See Neustadter, 35 Buffalo L. Rev. at 248–50 (cited in note 52).