The B.E. Journal of Economic Analysis & Policy

Transcription

1 The B.E. Journal of Economic Analysis & Policy Advances Volume 10, Issue Article 82 Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Lars Lefgren Frank L. McIntyre Michelle Miller Brigham Young University, lars [email protected] Brigham Young University, frank [email protected] Rutgers University, [email protected] Recommended Citation Lars Lefgren, Frank L. McIntyre, and Michelle Miller (2010) Chapter 7 or 13: Are Client or Lawyer Interests Paramount?, The B.E. Journal of Economic Analysis & Policy: Vol. 10: Iss. 1 (Advances), Article 82. Available at: Copyright c 2010 The Berkeley Electronic Press. All rights reserved.

2 Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Lars Lefgren, Frank L. McIntyre, and Michelle Miller Abstract Households often rely on professionals with specialized knowledge to make important financial decisions. In many cases, the professional s financial interests are at odds with those of the client. We explore this problem in the context of personal bankruptcy. OLS, fixed effects, and IV estimates all show that attorneys play a central role in determining whether households file under Chapter 7 or Chapter 13 of the bankruptcy code. We present evidence suggesting that some attorneys maximize profits by steering households into Chapter 13 bankruptcy even when the households objective financial benefits are low and the probability of case dismissal is high. An attorney-induced Chapter 13 filing increases household legal fees and reduces the probability of long-term debt relief. KEYWORDS: consumer bankruptcy, principal agent problems We thank Tina Lindsay and Jocelyn Rick for helpful discussions regarding institutional factors affecting personal bankruptcy. We thank Paul Weitzel, Amanda Boren, Laura Summers, Justin Hansen, Brad Hunter, Mirinda Martin, Henry Tappen, and Daniel Sullivan for excellent research assistance. We are grateful to David Sims, Joseph Price, Richard Butler, Brian Jacob, Mark Showalter, Brad Larsen, and seminar participants at Brigham Young University, the University of California, Berkeley, and the Wharton School of Business for their thoughtful comments.

3 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Introduction Individuals rely on informed professionals for many of their important financial decisions. A growing literature documents the extent to which professionals sacrifice their clients interests to increase profits. An early example of such work is Fuchs (1978) who presented evidence that the incidence and price of surgery increased with the supply of surgeons. More recently, Gruber and Owings (1996) provide evidence that doctors are more likely to perform expensive caesarian sections when there is relatively low demand for their services. Levitt and Syverson (2008) show that real estate agents price clients houses more cheaply than their own. Similarly, Harrington and Krynski (2002) discuss how funeral directors often induce households to bury their deceased loved ones in lieu of a less expensive cremation. Chevalier and Ellison (1997) demonstrate that mutual fund managers adjust the risk of their portfolio to maximize fund inflows instead of risk-adjusted returns. In addition, Bergstresser, Chalmers, and Tufano (2009) show that mutual fund brokers and financial advisors provide no discernible financial benefit to consumers. In this paper, we examine the principal-agent problem in the context of personal bankruptcy. When filing for bankruptcy, households can file under Chapter 7 or Chapter 13 of the Bankruptcy Code. 1 For most debtors, a Chapter 7 bankruptcy involves a quick discharge of most unsecured debts, while a Chapter 13 bankruptcy requires households to make payments for three to five years. As discussed in great detail below, the advantages and disadvantages of each chapter are complex--- indeed the optimal chapter for any debtor depends on very detailed aspects of his financial situation. Previous research, however, has suggested that the chapter under which households file is not purely a function of their financial situation. 2 1 During the time period in question, Chapter 7 and Chapter 13 relief was available to nearly every debtor. Prior to 2005, the bankruptcy trustee could ask the judge to dismiss a case because the debtor s income was so high that to permit the debtor to discharge his debts in Chapter 7 was a substantial abuse of the bankruptcy system. However, the attitudes of trustees and judges about abuse varied from district to district. Therefore, in 2005 the Bankruptcy Code was amended to include a means test which is intended to provide a more objective approach to the issue of a debtor s ability to pay. However, as discussed in more detail later in this paper, this income restriction rarely applies. It should be noted that relief under Chapter 11 was also available to individual debtors, but few debtors chose that option because of its significant cost. 2 For example, Lefgren and McIntyre (2009) report that the fraction of personal bankruptcies filed under Chapter 13 ranges from 0.03 in North Carolina to 0.62 in Georgia, suggesting legal culture plays an important role in chapter choice. Sullivan, Warren, and Westbrook (1998) emphasize that debt loads and repayment ability appear similar for Chapter 7 and Chapter 13 filers. Braucher (1993) presents qualitative evidence of the attorney s role in the chapter decision. Published by The Berkeley Electronic Press,

4 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 Personal bankruptcy is one of the most important mechanisms through which Americans are insured against adverse financial, health, and personal shocks. Nearly five out of every thousand persons in the United States filed for personal bankruptcy in 2009, more than quadruple the rate for And given the current economic crisis, the bankruptcy filing rate is likely to increase. According to the Administrative Office of the U.S. Courts, bankruptcy transferred over $206 billion dollars in more than state unemployment insurance programs and Temporary Assistance for Needy Families (TANF) combined (Lefgren and McIntyre, 2009). We find a strong correlation between the debtor s chapter decision and the chapter decisions of his attorney s other clients. Specifically, our household-level data from California, Texas, and Utah, shows that an attorney s fraction of other bankruptcies filed under Chapter 13 explains percent of the variation in chapter selection. This correlation could be due to households matching to attorneys based on the household s preferred bankruptcy chapter. However, we argue that this correlation is more likely caused by attorneys steering clients into the attorneys preferred bankruptcy option. In particular, the correlation persists even after we control for the household s financial situation and neighborhood fixed effects. Additionally, we find the same results when we employ an instrumental variables strategy in which we instrument attorney fraction 13 with a measure of the average attorney filing behavior in a debtor s neighborhood. This strategy takes advantage of the fact that a particular household is more likely to file under Chapter 13 if attorneys close to them specialize in this chapter. To address concerns that our instrumental variables specifications are biased by variation in neighborhood level attitudes towards default, we show that our results are robust to the inclusion of Census block fixed effects or Census tract*year fixed effects. Attorney specialization appears to be consistent with firm profit maximization. We show that large (high volume) firms and firms with a client mix better suited for Chapter 13 bankruptcies tend to specialize in Chapter 13. This supports a model in which firms that expect to file a large number of Chapter 13 bankruptcies engage in fixed investments to reduce the marginal costs of filing such cases. Thus, firms encourage Chapter 13 when the cost is low enough to rationalize (from a profit-maximization perspective) pursuing the higher courtregulated attorney fees available under this chapter. Firms specializing in Chapter 13 collect more for these bankruptcies, although less payment is required upfront. Yet these firms do not offer benefits in terms of lower dismissal rates or more manageable payment plans. Indeed, our IV estimates, which focus on households that chose attorneys based on geographic convenience, suggest that those who file under Chapter 13 with 2

5 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Chapter 13 specialists are often objectively poorly suited for this type of bankruptcy and have high dismissal rates. While it is plausible that some variation in attorney behavior may be driven by differing views regarding the advantages of each type of bankruptcy, attorney specialization is empirically consistent with debtor interests being subordinate to firm profits. Regardless of the cause of specialization, a typical lawyer-instigated decision to file under Chapter 13, as opposed to Chapter 7, leads to a substantial transfer of wealth from insolvent households to specialized attorneys with a reduced probability of long-term debt relief. Review of the Literature In addition to the empirical principal-agent literature cited in the introduction, our paper relates to a large existing literature on personal bankruptcy found in both the economics and legal disciplines. Most closely related to our analysis is a set of papers that explore factors driving a household s choice of bankruptcy chapters. Nelson (1999), Domowitz and Sartain (1999), Sullivan and Worden (1990), Li and Sarte (2002), and Sullivan et al. (1988) all find that financial incentives play a role in the household s choice of bankruptcy chapter. However, many economists note that legal culture and the choice of attorney also play important roles in the chapter decision. Specifically, Lefgren and McIntyre (2009) find that the propensity to file under Chapter 13 of the bankruptcy code varies greatly across localities. Because these differences are extremely persistent and exist across adjacent states with seemingly similar populations, they conclude that the differences in the proportion of Chapter 13 filings are likely due to legal culture. Sullivan et al. (1994) also argue that all of the variation across bankruptcy courts in chapter choice cannot be explained by state laws, by the behavior of particular individuals or by other non-legal factors. The authors develop a model in which the local legal culture is dominated by lawyers. According to this model, the lawyer, due to specialty, moral preference, or stereotype, influences the chapter choice. Braucher (1993), Neustadter (1986) and Sullivan et al. (1988) all provide important qualitative evidence that lawyers often steer households toward one particular bankruptcy alternative. Our paper is also closely related to the medical literature examining variation in treatment choice across providers and locations. Health economists have widely documented the variation in physician practice styles which cannot be explained by income, insurance, or patient preferences. Recent examples including Chandra and Staiger (2007), Epstein, Ketcham and Nicholson (2005), and Grytten and Sørensen (2003) show that choice of service provider plays an important role in the type of treatment patients receive. Chandra and Staiger Published by The Berkeley Electronic Press,

6 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 (2007) and Allgood and Bachmann (2006) highlight the health benefits that patients receive from physician specialization. The current study makes three contributions to the existing literature on chapter choice. First, we quantify the magnitude of the attorney s role in chapter choice. Second, we address concerns that the apparent role of lawyers is driven by the endogenous sorting of clients to attorneys. Third, we show that attorney Chapter 13 specialization is on average detrimental to clients and plausibly driven by profit maximization on the part of the attorney. Our paper also adds fresh evidence regarding the problems of principal-agent relationships when the principal (the insolvent household in our example) is unsophisticated. Institutional Background For insolvent households, personal bankruptcy is a primary instrument of debt relief. Federal authority for bankruptcy is found in Article 1, Section 8 of the United States Constitution. For this reason, many aspects of bankruptcy are uniform across states. When households file for bankruptcy, creditors must stop all collection measures; this means that creditors must cease all foreclosure proceedings and garnishments. In addition, creditors cannot send the debtor correspondence. 3 When they petition for bankruptcy, debtors must decide whether to file under Chapter 7 or Chapter 13 of the Bankruptcy Code. The complex differences between Chapter 7 and Chapter 13 are detailed in the subsection below. Under Chapter 7, often referred to as liquidation, households may only keep exempt assets. All other assets of value (if present) are liquidated by a trustee and distributed to creditors. 4 This process is fast and simple; Chapter 7 debtors are able to obtain a speedy discharge of most unsecured debts and remove some judicial liens. On the other hand, under Chapter 13, debtors keep all their property and instead agree to repay their debts using their future income. 5 Chapter 13 cases are comparatively complex and lengthy. From 1999 to 2001, 3 U.S.C (j) provides an exception for sending of correspondence for bills of the regular periodic payment on a home mortgage. 4 As detailed in the subsection below, in at least 95 percent of Chapter 7 cases, the debtor does not have any non-exempt assets to liquidate. 5 Before the Bankruptcy Code was amended in 2005, households could choose either a three or five-year repayment plan. If a debtor had a five-year repayment plan (as opposed to a three-year repayment plan), they typically paid less every month. Therefore, debtors typically repaid the same amount, regardless of the length of the plan. After the Bankruptcy Code was amended in 2005, household with income above the state median level were required to enter into a five-year repayment plan. Only debtors with income below the state median level could enter into a threeyear repayment plan. 4

7 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Lefgren and McIntyre (2009) report that 70.5 percent of all personal bankruptcies were filed under Chapter 7. Institutional Procedure: The Difference between Chapter 7 and Chapter 13 A Chapter 7 case begins when the debtor s attorney files a petition with the bankruptcy court. 6 In addition to the petition, the debtor must detail his assets, liabilities, monthly income, and average monthly expenses. After the petition is filed, a Chapter 7 bankruptcy trustee holds a meeting of creditors, during which the trustee and creditors may ask the debtor questions regarding these documents. According to conversations with bankruptcy attorneys, creditors typically do not even attend these meetings. Following the meeting of the creditors, the trustee will gather and sell some of the debtor s assets. The Bankruptcy Code, allows the debtor to keep certain exempt property. Specifically, debtors may keep assets with a value below the personal and homestead exemption levels. These exemption levels vary dramatically across states. For example, Florida has an unlimited homestead exemption while Delaware has none. The large majority of Chapter 7 cases involve no non-exempt assets that are profitable for the trustee to sell. According to the 1997 National Bankruptcy Review Commission Report, less than 5 percent of Chapter 7 consumer bankruptcies were denominated as asset cases. After the debtor s assets are liquidated, the proceeds are used to repay creditors and the debtor is discharged (released from liability) of any remaining debts. In other words, all remaining debts are then forgiven. Because the typical Chapter 7 bankruptcy does not involve contested filings or the liquidation of assets, it is quick and simple for both the attorney and the client. Furthermore, only a small fraction of such bankruptcies are dismissed. In our sample of Chapter 7 filings, only 3 percent were listed as being dismissed. Not all households, however, can file for Chapter 7 bankruptcy. Households who already filed for Chapter 7 bankruptcy within the past six years are ineligible. 7 Additionally, United States trustees, who represent creditor interests, may attempt to force individuals to file under Chapter 13 if it is clear that the debtors possess the ability to pay a substantial fraction of their debts percent of Chapter 7 and 4.1 percent of Chapter 13 cases are filed without an attorney. We ignore such cases in our analysis. 7 The Bankruptcy Abuse Prevention and Consumer Protection Act of 2005 (BAPCPA) extended the time households must wait between consecutive Chapter 7 cases to eight years. 8 Indeed, while courts have long had the discretion to mandate that high-income households file under Chapter 13, the BAPCPA required that households earning income above the state median level file under a Chapter 13. Households are allowed to deduct certain expenses from their income, however. Most economists argue that this income restriction rarely applies. In a sample of 1,938 cases filed nationwide between November 1998 and August 1999, Flynn and Bermant (2000) only found two petitions which were filed under Chapter 7 but failed the means test. Published by The Berkeley Electronic Press,

8 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 Rarely are the income restriction or trustee objections an obstacle for households considering bankruptcy. Indeed, in our sample, only 47 cases were converted from a Chapter 7 to a Chapter 13. As an alternative to Chapter 7, households may file for bankruptcy under Chapter 13 of the Bankruptcy Code. As in a Chapter 7, a Chapter 13 case begins when the debtor s attorney files a petition with the bankruptcy court. Again, the debtor must detail their assets, debts, monthly income, and average monthly expenses. In addition to the bankruptcy petition, when a debtor files under Chapter 13 of the Bankruptcy Code, their attorney must create a Chapter 13 repayment plan. This plan specifies the amount debtors will pay to the trustee every month; debtors pay their projected monthly disposable income, generally calculated as the difference between their monthly income and monthly budgeted living expenses, into the Chapter 13 payment plan. 9 Thus, to create this plan, attorneys must collect and investigate their client s receipts and bank statements, and generate a comprehensive itemized budget for their clients to follow over a three- to five-year period. The paperwork, planning, and organization required to complete this task is substantial. Therefore, filing under Chapter 13 of the Bankruptcy Code involves substantially more time and paperwork than a Chapter 7. After the repayment plan is submitted, as in a Chapter 7, the bankruptcy trustee holds a meeting of creditors, during which the trustee and creditors may ask the debtor questions regarding these documents. After the meeting with the creditors, the Chapter 13 repayment plan must be confirmed by the court. This process can also be fairly difficult and time Similarly, Culhane and White (1999) estimated that only 3.6 percent of debtors would be barred from Chapter 7 when a means test was enacted. True to speculation, within the first year of its enactment, the Acting Director of the Executive Office of the U.S. Trustees testified that approximately 0.5 percent of Chapter 7 filers were being affected by the means test (Tabb and McClelland 2007). White (2007) also finds that only a relatively small minority of filers were affected by the means test. 9 Before the BAPCPA of 2005, households used their projected monthly income and actual monthly expenses to determine the monthly amount paid into a Chapter 13 plan. After 2005, households use the means test to determine the monthly amount paid into a Chapter 13 plan. Instead of using their projected monthly income, households use their average monthly income over the past six months. If the household s average income over the past six months was below the state median income level, the household deducts actual monthly expenses--- the difference between its average monthly income and its actual monthly expenses is the amount the household will pay into a Chapter 13 plan each month. If the household s average income over the past six months was above the state median income level, the household subtracts standard deductions for expenditures---the difference between its average monthly income and standard expenditure deductions is the amount the household will pay into a Chapter 13 plan each month. A recent Supreme Court decision, however, stated that if a household s circumstances changed over the past six months, and its average monthly income over the past six months differs from its projected monthly income, the household must use its projected monthly income to calculate the amount paid into the Chapter 13 plan. 6

9 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? consuming. On average the Chapter 13 confirmation process takes four months to complete. 10 At the confirmation hearing, the Chapter 13 repayment plan can be dismissed, converted, or confirmed. Dismissal can occur if the repayment plan is not in the best interest of the creditors, if the plan was not proposed in good faith, if the plan is infeasible, or if the debtor does not cooperate in providing sufficient documentation. If the case is dismissed, the debtor does not receive a discharge, and thus, does not receive any long-term debt relief. If the case is converted to a Chapter 7, the debtor only receives long term debt relief if they are able to receive a Chapter 7 discharge. If the Chapter 13 repayment plan is confirmed, the Chapter 13 trustee will begin to distribute funds to creditors. 11 Only upon completion of the plan, which, as mentioned, takes three to five years, are the household s remaining debts discharged. 12 If the debtor is unable to complete their repayment plan, the case will either be converted or dismissed. If the case is ultimately dismissed the household is once again liable for any remaining debts; thus, they do not receive any long term debt relief. It should be noted that debtors can file under Chapter 13 as frequently as every six months. 13 Figure 1 contains a diagram outlining the legal procedures discussed above. The Advantages and Disadvantages of Chapter 13 Relative to Chapter 7, Chapter 13 is advantageous for some households but is detrimental to many debtors. One of the main benefits of a Chapter 13 is that debtors may keep all of their assets. However, in order to discharge its debts under Chapter 13, a household must complete its repayment plan. Thus, Chapter 13 is designed for households with a continuing ability to earn income and minimize future expenditures. As mentioned, if a debtor is unable to follow through on a long-term plan, his case will be dismissed, and he will be liable for all his original debts as well as additional court and lawyer fees. According to 10 By comparison, a Chapter 7 bankruptcy may never appear in court. Unless there is a problem, a Chapter 7 bankruptcy sails through the administrative process without appearing before a bankruptcy judge. 11 The funds are first disbursed to the debtor s lawyer, then to creditors with priority claims, and finally, any remaining funds are used to repay creditors with unsecured claims. 12 The Chapter 13 discharge is often referred to as the super-discharge; in addition to the debts discharged under Chapter 7, Chapter 13 debtors can discharge debts from property settlements following a divorce, debts for willful and malicious injury, child support payments, and debts incurred to pay non-dischargeable tax obligations. Before the BAPCPA, debtors filing under a Chapter 13 could also discharge debts from embezzlement, larceny and fraud. 13 The BAPCPA extended this time period; households must now wait two years between Chapter 13 cases. Published by The Berkeley Electronic Press,

10 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 Lefgren and McIntyre (2009), while 29.5 percent of bankruptcies are filed under Chapter 13 of the bankruptcy code, the majority of these bankruptcies (60 percent) are dismissed largely due to nonpayment on the debtor s part. 14 Another 12 percent of the Chapter 13 filings are ultimately converted to Chapter 7 for the same reason. Thus, Chapter 13 bankruptcies only represent 10 percent of bankruptcy discharges. For these reasons, the majority of households filing under Chapter 13 do not receive one of the long-term benefits of bankruptcy, the discharge of unsecured debts. Figure 1: Institutional Procedures 14 Others have found even higher failure rates of Chapter 13 bankruptcies. See for example, Norberg and Velkey (2006). Norberg and Velkey (2006) also discuss how many households file repeat Chapter 13 bankruptcies in response to these high dismissal rates. 8

11 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Another benefit of filing under Chapter 13 is that debtors may save secured property (such as a home) by curing an arrearage (the value of missed payments) this benefit is discussed in more detail below. 15 Again, it is unlikely that this benefit will be realized if the case is either dismissed or converted to a Chapter Therefore, as this secondary benefit is primarily received upon completion of the plan (when a discharge is received) it stands to hold that the majority of households filing under Chapter 13 do not receive the long-term benefits that bankruptcy can provide. Finally, under Chapter 13 bankruptcy, the household retains the right to file under Chapter 7 should future economic shocks occur. Attorney Fees Both the timing of the attorney fees and the amount collected vary by chapter. Chapter 7 attorney fees are unregulated by the courts and households typically pay the entire fee upfront. 17 With Chapter 13, households usually pay some fees upfront; however, the majority of the fees are collected as part of the household s repayment plan. This can represent a substantial benefit to the clients as they are often hard pressed to assemble the legal and filing fees up front. 18,19 Individual bankruptcy courts set standard no look fees for Chapter 13 compensation--- only fees above a customary limit are subject to special scrutiny. Thus, attorney 15 Collateral retention may also be possible under Chapter 7. Before the BAPCPA, five circuit courts of appeal recognized the ride-through as part of the Bankruptcy Code. This ridethrough allowed bankrupt households to keep their property during and after bankruptcy by remaining current on their payments; it prevented creditors from imposing harsher terms on debtors during the bankruptcy process. The BAPCPA modified the Bankruptcy Code sections relating to the ride-through. While the ride-through was not eliminated, the BAPCPA also did not endorse its expansion. As a result, the courts are still split over the ride-through (Hogan 2008). 16 If the Chapter 13 repayment plan is breached, the creditor is no longer bound by the plan and the debtor is once again liable for all debts in addition to any accrued late fees. In cases where the arrearages are cured, the debtors may still be able to retain secured assets even if the plan fails. 17 In the Texas sample for which we have some repayment information, 60% of Chapter 7 lawyer fees are completely paid up front. For the other 40% it is not clear from our available data when the remainder was paid. For Chapter 13 filings virtually none are paid up front, with the vast majority of clients having paid less than 25% up front. 18 See Mann and Porter (2010) for a discussion of the difficulty households face in saving for bankruptcy. 19 In some cases, an attorney will convert an unsuccessful Chapter 13 case to a Chapter 7 without additional charge if adequate fees have been paid through the repayment plan. Published by The Berkeley Electronic Press,

12 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 fees are tightly clustered around this informal limit. 20 Courts also retain the right to review Chapter 7 fees for reasonableness. As we document above, Chapter 13 bankruptcies are substantially more difficult to file than Chapter 7 bankruptcies. Perhaps for this reason, the customary Chapter 13 fee is typically much higher than the equilibrium Chapter 7 fee set by the market. A 2008 U.S. Government Accountability Office (GAO) report found that before the BAPCPA, the median attorney fees for a Chapter bankruptcy was $2,400 compared to $712 for a Chapter 7 bankruptcy. Attorneys are able to collect higher fees from a Chapter 13 bankruptcy even if the case is unsuccessful (i.e., even if the household is unable to discharge its unsecured debts). Households begin making payments into the Chapter 13 repayment plan within one month of submitting the plan. Thus, even if the plan has not been confirmed, households must begin making payments into the Chapter 13 repayment plan. In the Chapter 13 repayment plan, attorney and trustee fees are given priority status relative to other creditors; their fees may displace payments to other creditors. Thus, even if the plan fails, attorneys collect the majority of their payment. 22 As an example of this, we drew a random sample of 54 dismissed (unsuccessful) Chapter 13 bankruptcies filed in Texas in For these 54 cases, the average lawyer fee was $2,055. The final trustee reports show that the average dismissed case lasted 13 months. Despite the rapid dismissal of the majority of these cases, lawyers still received $1,376 on average 66 percent of the nominal fee. In discharged cases (which we did not sample), we know the lawyer is fully paid; thus a lawyer s unconditional expected payment from a Chapter 13 filing was 80 percent of their fee. 23 The average unsecured debt in these dismissed cases was $19,803; households repaid $302 on average. Secured debt averaged $21,546 with $2,280 in principal repaid and $598 in interest. Thus, for this small Texan sample, even when a case was dismissed, the attorney received substantial compensation though a very small fraction of the other debts was repaid. In 22 out of 54 dismissals, the lawyer was paid more than all other creditors combined. 20 See Braucher (1993). Note that these fee limits increased in the aftermath of the 2005 law change. The averages we report here are from February 1, When we look at fees as an outcome variable, we restrict our sample to pre A discussion of how the BAPCPA altered attorney fees can be found below. 22 Braucher (1993) finds that in some jurisdictions, the price differential between Chapter 13 and Chapter 7 bankruptcies is much smaller. In these jurisdictions, lawyers only rarely file bankruptcies under Chapter percent of 2003 Chapter 13 cases in Texas were dismissed..4*1 +.6*.66 =

13 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Lawyer Specialization Lawyers vary significantly in the fraction of bankruptcies they file under Chapter 13. Figure 2 shows the distribution of lawyers in California, Texas, and Utah by the fraction of bankruptcies they file under Chapter 13. This histogram demonstrates the large variation in lawyer behavior. 42 percent of lawyers file fewer than 10 percent of their cases under Chapter 13 of the Bankruptcy Code. On the other hand, 20 percent of lawyers file more than 40 percent of bankruptcies under Chapter 13. If bankruptcy clients shop for lawyers who meet their specific financial situations, specialization may not indicate that lawyers play a significant role in determining chapter choice. On the other hand, if filing households are relatively unsophisticated when selecting a lawyer, the high degree of lawyer specialization suggests that the attorney is crucial in a household s choice of bankruptcy chapter. Prior research suggests that attorneys play a strong role in guiding households in their bankruptcy decisions. Braucher (1993) notes that consumer bankruptcy clients are not typically repeat clients. 24 First time bankruptcy clients typically have little information regarding the costs 24 Over time it has consistently been shown that repeat filers represent a small fraction of debtors. Sullivan, Warren, and Westbrook (1989) estimated the only 3 percent of their sample represented repeat filers. Similarly, the 1997 National Bankruptcy Review Commission estimated that only 8 percent of debtors were repeat filers. And in a sample of debtors who filed after the BAPCPA, Miller and Miller (2008) found that only 5 percent of debtors were repeat filers. Published by The Berkeley Electronic Press,

14 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 and benefits of each type of bankruptcy. Additionally, it is likely that the social stigma associated with personal bankruptcy reduces information about peoples bankruptcy experiences. Braucher (1993) effectively documents the role that attorneys play in guiding typically unsophisticated clients towards one type of bankruptcy. Some attorneys influence clients to file under Chapter 13 by emphasizing the benefits of retaining secured assets, the morality of repaying creditors, and access to future credit. 25 Other attorneys shift households towards Chapter 7 bankruptcy by highlighting the difficulty of maintaining the payment plan, the moral obligations to provide financially for their families, and the predatory nature of some creditors. 26 Braucher finds that clients' guilt and loss of self-esteem makes them highly vulnerable to lawyers' influence, whether exercised unwittingly or deliberately. She concludes that attorney practices have more effect on chapter choices than features of the law conventionally thought to be important. Similarly, Sullivan et al. (1988) find that among other factors, attorneys greatly influence the choice of chapter. Using data collected during surveys, the authors find that nearly 32 percent of debtors consulted attorneys specializing in bankruptcy while the remaining debtors sought counsel from more general practitioners. The authors discover a positive, moderate unconditional correlation (r = 0.27) between seeing a bankruptcy specialist and filing under Chapter 13 of the Bankruptcy Code. This suggests that attorneys may exert an important influence over whether debtors file Chapter 7 or Chapter 13. The heterogeneity in attorney behavior can be rationalized by a model in which firms engage in investments to lower the marginal cost of filing Chapter 13 bankruptcies. 27 The formal model is available upon request from the authors. As discussed above, attorneys collect higher fees from a Chapter 13 bankruptcy. Thus, while Chapter 13 bankruptcies are more difficult to file than Chapter 7 bankruptcies, they generate more revenue. However, firms can engage in investments to lower the marginal cost of filing Chapter 13 bankruptcies Though a Chapter 13 bankruptcy remains on a credit report for less time, in some cases Chapter 7 is a better choice for households seeking access to credit. Lenders are aware that households who file under Chapter 7 have fewer competing debt obligations than Chapter 13 filers. Also filing for Chapter 7 bankruptcy removes the option of doing so again in the near future. 26 The difficulty of assessing the relative costs and benefits of each chapter may also allow attorneys to rationalize either chapter choice decision. 27 Many others have considered the principal agent problem from a theoretical standpoint. Yet none have considered the problem in the context of personal bankruptcy. However, Carlin and Gervais (2008) and Inderst and Ottaviani (2010) are recent examples of theoretical papers considering principal agent problems in the context of retail financial markets. 28 Compared to Chapter 13 bankruptcies, Chapter 7 bankruptcies are relatively simple to file (at times they are even filed without the help of an attorney) which is why we do not consider the possibility that attorneys could engage in investments to reduce the cost of filing Chapter 7 bankruptcies. 12

15 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Suppose that firms can undertake investments in technology, personnel, or human capital to lower the marginal costs of filing bankruptcies under Chapter Firms that expect to file enough Chapter 13 bankruptcies would find it optimal to undertake such investments. Investments made to lower the cost of filing bankruptcies under Chapter 13 lead to law firm specialization. More specifically, investing firms find it optimal to steer a higher fraction of households into Chapter 13 than non-investing firms, even holding client mix constant. There are two testable implications of this model. First, larger (high volume) firms will file a higher fraction of bankruptcies under Chapter 13 than smaller firms. The second implication is that firms with a client mix better suited for Chapter 13 will be more likely to file any given bankruptcy under Chapter 13. Similar implications can be obtained from a learning-by-doing model in which attorneys become better at filing Chapter 13 bankruptcies with practice. Our theoretical framework suggests that attorneys will steer households towards the bankruptcy option that maximizes profits, even, in some cases, at the 30,31 expense of their clients financial interests. Of course, Chapter 13 specialization may also lead to lower costs or improved bankruptcy outcomes. We examine these possibilities empirically and find little evidence that insolvent households benefit from employing a Chapter 13 specialist. Description of Data Our bankruptcy data come from Public Access to Court Electronic Records (PACER), the bankruptcy courts centralized registration and billing website. 32 Our dataset is limited to the three bankruptcy courts which gave us access to their 29 Specialized software programs automate much of the bankruptcy filing process. According to attorneys, this software is more helpful for the filing of Chapter 13 bankruptcies than the filing of Chapter 7 bankruptcies. 30 Specialization likely also reflects different attorney beliefs regarding the relative benefits of Chapter 7 and Chapter 13 bankruptcy. Lawyers may also vary in how seriously they take their obligation to look after their clients best financial interests and ethical views. These hypotheses cannot be tested, however, with available data. 31 One might believe that market competition among attorneys would limit the ability of attorneys to steer their clients in a direction contrary to their best interests. However, there is strong reason to believe that attorneys have substantial market power. In particular, households are much more likely to go to attorneys located close to their homes than to more distant attorneys. This is manifest by the strong first stage power of our instrument, which we discuss later. Consequently, we believe that attorneys have a great deal of influence over their clients choices. Additionally, the disciplining power of reputation is limited by clients sense of embarrassment in discussing bankruptcy. 32 Each court operates its own database with case information. Launched in 1997, courts slowly began to use PACER s electronic database system. The last court adopted the PACER system in Published by The Berkeley Electronic Press,

16 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 electronic records: Utah, Texas Northern, and California Northern from the period 2000 to The Utah sample is a census of bankruptcies from early 2000 to late 2004; after cleaning the data, our Utah sample includes 68,123 cases. On the other hand, the Texas and California data are samples from 2003 to 2006; these samples include 29,188 of the 113,412 and 24,105 of the 78,263 cases filed in the Northern Districts of Texas and California, respectively, during these four years. 36 percent of the bankruptcies in our sample were filed under Chapter 13 relative to just under 30 percent in the U.S. prior to the 2005 bankruptcy reform and just over 20 percent in The filing rates for the states in our sample are also somewhat higher than the filing rate for the nation as a whole. 33 For each petition in our sample, we have information on the bankruptcy chapter, the filer s address, the lawyer s identity, and the lawyer s address. In addition, we observe whether the case was dismissed or received a discharge of debts. Because we know the lawyer used in each case, we can calculate the fraction of bankruptcies (excluding the reference individual) filed under Chapter 13 for each attorney. We do this separately by year to allow for changes in attorney behavior over time, though the results are robust to calculating this fraction for the entire sample period. Filer address information was used to merge in 2000 Census block group level data. 34 Census block groups typically contain 600 to 3,000 residents, with 1,500 being the target size. The Census data provide information on demographics, income, and housing values of the filer s neighborhood. Table 1 shows sample means of the filing information and block group level demographics. The Census data suggest that nearly two-thirds of the population is married and another 11 percent is divorced. While the large majority at the block group level has completed high school, about one fifth of the population has obtained an undergraduate degree. The average median income across Census block groups is $50,122, the average unemployment rate is 3.5 percent, and about 61 percent of households are homeowners. Finally, almost the entire sample, 93 percent, resides in an urban Census block. For a subset of 33 The national average was 15 filings per 1,000 households in For approximately the same time period, California s filing rate was 16, while Utah s filing rate was 28, and Texas s filing rate was 8. In terms of demographics, per capita income in 2006 was $25,287 in the entire United States, $21,016 in Utah, $26,974 in California, and $22,501 in Texas. The foreign born percentage was 13 percent in the United States, 8 percent in Utah, 27 percent in California, and 16 percent in Texas. The black percentage was 12 percent in the United States, 1 percent in Utah, 6 percent in California, and 12 percent in Texas. Median age was 36 in the United States, 28 in Utah, 34 in California, and 33 in Texas. We control for observable demographic differences in the regressions. 34 From 158,218 fillings, we were able to match 139,839 to latitude and longitudes, which enabled us to assign them to Census block groups. Of those, we drop another 10,254 because they filed with a lawyer who filed fewer than 10 filings in the year (and so we cannot precisely estimate their lawyer s filing propensity. We also exclude those who live more than 10km from the centroid of the nearest census block group, leaving a final sample of 121,

17 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Variable Census Block Sample Texas Detailed Sample Table 1: Sample Means Variable Census Block Sample Texas Detailed Sample Filed Median Household Income $50,122 $47,743 Lawyer Fraction HH Income Under $10, Urban HH Income $10-$20, Married HH Income $20-$30, Divorced HH Income $30-$40, Household of HH Income $40-$50, Household of HH Income $50-$60, Household of HH Income $60-$75, Household of HH Income $75-$100, Household of Fraction Homeowners Household over th Percentile of Log Housing Value Finished High School th Percentile of Log Housing Value Finished college Monthly Income 3,131 Black Fraction with no Income Hispanic Land Assets 76,239 Other Race Fraction with no Land Assets Age Below Personal Assets 30,990 Age 6 to Fraction with no Personal Assets Age 19 to Secured Debts 82,216 Age 25 to Fraction with no Secured Debts Age 30 to Unsecured Debts 47,036 Age 40 to Fraction with no Unsecured Debts Age 50 to Unemployed Observations 121,416 15,293 Self-Employed Published by The Berkeley Electronic Press,

18 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 households in our dataset, we were able to collect more detailed information. Specifically, for a sample of 15,293 Texas filings, the court provided additional machine readable documents. For households in this subsample, we could collect data on income, debt, asset, and household composition. 35 Summary statistics for this sample are also provided in Table 1. The means of the Census block variables are generally similar to those of our baseline sample. The additional financial information shows that households in this subsample earn about $3,100 in gross income per month, have secured debt levels of about $82,000 and unsecured debt levels around $47,000. Empirical Specification One of our goals is to document the role attorneys play in the chapter decision; in this section we outline our methodology. We later use analogous methods to examine the consequences of filing with a particular attorney on other bankruptcy outcomes including the dismissal rate, appropriateness of Chapter 13, and legal fees. Note that our sample consists of households that actually filed for bankruptcy; our empirical model will be misspecified if a client s decision to file for bankruptcy (regardless of type) depends on his lawyer s behavior. It seems plausible, however, that most households are in such dire financial straits that the decision to file for bankruptcy is insensitive to attorney behavior. Braucher (1993) reports that By the time [debtors] consult a lawyer, they may have been living on credit card advances for six months or a year after losing a job or undergoing a divorce. We therefore assume that clients inelastically file for bankruptcy and simply choose between Chapter 7 and Chapter We wish to estimate the extent to which a lawyer s propensity to file a certain chapter steers the household to file under that chapter. Thus our baseline empirical specification is the following linear probability model: 35 From this sample we exclude households with over $500,000 of unsecured debt. There were fewer than 20 of these outliers which were, in many cases, obviously erroneous. This cleaning did not substantially affect our conclusions, although in a couple cases it had a large impact on reported means. For example, one filing reported debts in excess of seven billion dollars for an unsecured lease. 36 The impact of lawyer behavior on the probability that a household files for bankruptcy cannot be measured using our data. However, it seems implausible that this would be an important concern, as nearly all lawyers file some bankruptcies of each chapter. To the extent that it is more profitable to file a client under the non-preferred chapter than to not file them at all, a lawyer has a strong incentive to provide the type of services that keeps the client from leaving his office. Additionally, a household that is pushed in a direction contrary to its best financial interest retains the option to consult a different attorney. Finally, to the extent that pushing households towards Chapter 13 bankruptcy, which typically yields the fewest financial benefits, causes them not to file for bankruptcy at all, our estimates will be biased downwards. 16

19 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? (1) chapter13 ij = X i β + α frac13 ji + ε ij where chapter13 ij is a binary variable equal to one if individual i filing with attorney j files under Chapter 13 of the Bankruptcy Code and zero otherwise. X i is a vector of household-level characteristics correlated with the decision to file under Chapter 13, and frac13 ji is the fraction of an attorney s clients who file under Chapter 13, excluding the reference household, i. Going forward, we will refer to this term as fraction 13. We use frac13 ji as a proxy for the lawyer s latent propensity to file households under Chapter 13. Due to potential nonrandom matching between firms and households, estimates of α may not reflect the causal effect of lawyer behavior on household filing behavior. Suppose, for example, that lawyers are chosen by filers to maximize an objective function that takes as inputs the lawyer price, p j, distance to travel to the lawyer, d ij, lawyer quality, q j, and, potentially, the lawyer s propensity to file a Chapter 13 bankruptcy, θ j. Filers may vary in terms of their preferences (or suitability) for Chapter 13 bankruptcy, which is indexed by ψ i. Thus, for each lawyer the agent computes U(p j, d ij, q j, θ j ψ i ), and picks the lawyer who maximizes this function. The OLS specification in equation (1) is inconsistent if ψ i, the agent s preference for Chapter 13 bankruptcy, is causing them to match to attorneys with a high θ j. In this case, the estimated α reflects variation in both ψ i and θ j. 37 Alternatively, distance to a particular lawyer, d ij, may be correlated with consumer preferences, ψ i, if attitudes toward Chapter 13 bankruptcy vary across neighborhoods. In this case, our estimate may reflect the preferences of the lawyer s clientele rather than a lawyer effect. One method to control for these unobserved preferences is to assume that ψ i = ψ k, where k is the neighborhood of household i. Then we can control for this unobserved preference using neighborhood fixed effects: (2) chapter13 ij = ψ k + X i β + α frac13 ji + ε ij where ψ k is estimated using Census block group fixed effects. Alternatively, we can define the neighborhood more broadly at the tract level, but include year interactions so that preferences are allowed to vary arbitrarily over time. This 37 A brief perusal of the yellow pages in Utah found plenty of advertisements for bankruptcy attorneys, but nothing that suggested lawyers seek to match on the basis of chapter filing preferences of the lawyer or client. One attorney, though, did mention their flat fee for Chapter 7 a possible signal of Chapter 7 preference. Braucher (1993) provides more systematic evidence that attorneys do little to signal specialization. Published by The Berkeley Electronic Press,

20 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 approach identifies the attorney s role based on the decisions of households that live very close to each other but choose to file with different attorneys. Of course, if unobserved filing preferences are idiosyncratic, and not defined at the neighborhood level, we will still get an inconsistent estimate of α. Our preferred approach is to refer back to the above utility function and use d ij, distance, as a source of variation in lawyer choice. We form an instrument, defined below, that proxies for the distance-weighted average of θ j in the surrounding area. This instrument is correlated with lawyer behavior and is, by construction, uncorrelated with the idiosyncratic preferences of households within a small neighborhood. Because neighborhood-level preferences may vary, we examine specifications with a variety of neighborhood controls and fixed effects. Even with these controls, our IV estimates could still be biased if there were substantial spatial differences in attitudes towards bankruptcy within a Census tract or block group. 38 We believe this is unlikely to be a significant source of bias, however. This weighted average instrument is defined as: (3) frac13 ı = j frac13 ji F j /d ij j F j /d ij where F j is the annual number of filings for the lawyer and d ij is the distance (in miles) from debtor i to lawyer j. Thus the instrument weights each lawyer by their proximity to the household and number of filings. F j increases the power of the instrument because it contains information about unobserved components of the filer s choice, such as lawyer price, advertising, and quality. Our results remain consistent if we weight by both F j and d ij or just d ij. Although one cannot directly test whether the instrument is correlated with our unobserved error term, we can examine whether it is correlated with observed covariates of the bankruptcy decision. If the instrument is highly correlated with the covariates, it raises concerns about what the unobserved variation may be correlated with. Thus for our Texas sample we regress the instrument, frac13 i, on the block group covariates used below and get an adjusted R-squared of When we add in household financial characteristics the R-squared is still only This is a positive sign that the instrument is not simply picking up neighborhood or personal characteristics. 38 For example, suppose preferences for debt repayment under Chapter 13 bankruptcy were stronger on one side of a neighborhood than another. In this case, the fraction 13 of attorneys servicing opposite sides of the neighborhood would differ due to the spatial variation in preferences. Our IV strategy, however, would attribute these differences to attorney steerage. 18

21 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Findings Table 2 reports the OLS estimates of the impact of lawyer fraction 13 on individual filing behavior. 39 The first column shows the OLS results for our full sample, without controls. Unsurprisingly, the coefficient on lawyer fraction 13 is This point estimate suggests that going from a lawyer who files 50 percent of bankruptcies under Chapter 13 to one who files 60 percent under Chapter 13 would increase an individual s own probability of filing under Chapter 13 by 9.8 percentage points. Furthermore, the R-squared is 0.25 suggesting that a quarter of the variation in filing rates can be predicted with the lawyer s identity. This result is consistent either with lawyers having a large causal effect on chapter choice or with strong sorting of individuals to those lawyers who specialize in their chosen type of bankruptcy. In the second column, we add a host of neighborhood covariates and state by year fixed effects. As detailed in Table 1, the neighborhood variables contain information about marital status, household size, education, race, employment, income, homeownership, and home value. Presumably many of these factors would be correlated with the financial or cultural factors that could drive 40 bankruptcy decisions. The addition of these covariates, however, has almost no effect on the lawyer fraction 13 coefficient and their partial R-squared is virtually zero. In the third column of Table 2, we include Census block group fixed effects. As discussed above, this controls for all time-invariant neighborhood characteristics and utilizes the variation in households who live very close to each other but go to different attorneys as well as the variation in attorney behavior over time. The inclusion of these fixed effects has virtually no impact on our coefficient estimates. Furthermore, the partial R-squared of the fixed effects is again miniscule, suggesting that neighborhood-specific attitudes or circumstances play little role in households choice of bankruptcy chapter, conditional on attorney fraction Recall that for our entire analysis we exclude the individual s own filing when calculating lawyer fraction 13. Appendix Table A1 gives the full regression output with coefficients for all covariates. 40 Lefgren and McIntyre (2009) show that demographics play an important role in explaining the level (if not type) of bankruptcies. Published by The Berkeley Electronic Press,

22 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 Table 2: OLS Estimates of Lawyer Impact on Chapter 13 Choice (1) (2) (3) (4) (5) (6) (7) (8) Census Census Census Census Texas Texas Texas Texas Lawyer Fraction ** 0.958** 0.955** 0.957** 0.905** 0.763** 0.745** 0.740** [0.006] [0.006] [0.007] [0.007] [0.017] [0.016] [0.019] [0.015] State By Year Dummy Variables No Yes Yes No No Yes Yes No Neighborhood Covariates No Yes No Yes No Yes No Yes Neighborhood Fixed Effects None None Block Group Tract* Year None None Block Group Personal Filing Covariates No No No No No Yes Yes Yes Tract* Year 2 Adjusted R Partial R 2 of Ch. 13 Filing Rate Observations 121, , , ,416 15,293 15,293 15,293 15,293 * significant at 5%; ** significant at 1%. Robust standard errors in brackets, clustered at the zip code level. "Neighborhood Covariates" are dummy variables covering Urban, Marital Status, Household Size, Education, Race, Age, Employment Status, Household Income, Homeownership, and Housing Value. Personal Filing Covariates, in Columns 6-8, include log values of Monthly Income, Land and Personal Assets, Secured and Unsecured Debt as well as dummy variables for those with zero values in any covariate. Full regression results are available in the Appendix Table A

23 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? As a robustness check, in the fourth column, we allow neighborhood effects to vary over time; specifically, we include Census tract by year fixed effects. 41 This specification allows for arbitrary changes in neighborhood-level preferences over time. The results are identical, and once again the fixed effects add little in terms of explanatory power. In the fifth, sixth, seventh, and eighth columns we perform similar analyses on the subsample of Texas petitions for which we have detailed financial information. While smaller than our baseline sample, we can control for the household-level factors that are likely to affect the choice of bankruptcy chapter. In column 5, we examine our detailed Texas subsample, but do not include covariates. Consistent with our results in column 1, the estimated coefficient on lawyer fraction 13 is In the sixth column, in addition to the Census block covariates described earlier, and state by year fixed effects, we control for detailed household-level characteristics. In particular, we control for the log of each household s monthly income, real and personal assets, and secured and unsecured debts. We also include dummy variables for observations that report a zero or negative value for any of these measures. 42 The fraction of other cases the attorney files under Chapter 13 still remains a primary determinant of bankruptcy chapter, although including these covariates causes the coefficient to drop slightly to In the seventh column, we include Census block group fixed effects. The inclusion of these fixed effects has virtually no impact on our coefficient estimates; their partial R-squared is less than one percent. Finally, as seen in the eighth column, including Census tract by year fixed effects has no additional impact on our results. Thus, even after we control for the household s specific financial characteristics, we find that lawyer fraction 13 is a strong predictor of a household s chapter choice. Of course, this analysis only identifies a lawyer s causal effect if we have adequately controlled for the factors which are correlated to both the households choice of lawyer and of bankruptcy chapter. 41 By including tract*year fixed effects instead of block*year fixed effects, we conserve over 20,000 degrees of freedom. In unreported results, we consider the possibility that unobserved neighborhood preferences at the block group level may vary over time. We regress chapter filings on block group dummies and block group specific linear time trends for our Texas subsample, along with state level time dummy variables. We cannot reject the null hypothesis that the block group linear trends are jointly zero (p=0.54). When we then add lawyer fraction 13 to the regression (either instrumented or not) it gives statistically significant, though not very precise, estimates. Thus, our approach is robust to idiosyncratic time trends across neighborhoods. 42 A list of the additional household covariates can be found in Table 1. It is possible, of course, that there are important financial variables that are not included in our dataset. Additionally, due to state and federal bankruptcy regulations, the variables we have may enter the bankruptcy chapter decision in a non-linear manner. In this case, we might expect that household characteristics play a larger role in the bankruptcy decision than characterized in our regression. Published by The Berkeley Electronic Press,

24 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 IV Results As discussed above, OLS may insufficiently control for the financial or personal determinants of chapter selection. To address this concern, we employ the instrumental variables strategy detailed above; the results are shown in Table 3. The first column contains estimates from our baseline sample in which we include no covariates. With a first stage F-statistic of 2,372, we see that our instrument has considerable explanatory power. The estimated coefficient on lawyer fraction 13 is 1.00, consistent with lawyers playing a leading role in the bankruptcy chapter decision. Our identifying assumption is that household location is uncorrelated to propensity to file under Chapter 13. If this assumption failed to hold, our instrument would simply reflect neighborhood-level preference for one type of bankruptcy over another. In the second column, we include neighborhood covariates as well as state by year fixed effects. It seems plausible that neighborhood measures of income, family structure, and housing value would be among the leading candidate to explain why Chapter 13 bankruptcy is more prevalent in one locale than another. Our instrument retains considerable predictive power after we include these neighborhood-level covariates. The point estimate on lawyer fraction 13 is 0.95, similar to our other estimates. While these results are compelling, if unobserved neighborhood attitudes are a significant determinant of bankruptcy chapter, our estimates will remain inconsistent. In the third column, we include Census block group fixed effects. In this specification, identifying variation comes from two sources. First, like earlier regressions, there is variation over time with the entry, exit, and practices of firms. Second there are slight differences in location among households within a given Census block group. This strategy will yield inconsistent estimates only if neighborhood preferences (relative to the state) change at a high frequency. 43 While this strategy yields less precise estimates, the substantive findings are essentially unchanged. The coefficient on lawyer fraction 13 is As discussed in a previous footnote, we do not find evidence of block group time trends in Texas over the four years of the sample and in unreported results we find that estimates are robust to their inclusion. 44 These fixed effects also substantially reduce the F-statistic test on the excluded instruments in the first stage, thus one may be concerned about a weak instruments bias (see Stock and Yogo (2002) and Shea (1997)). Fortunately, one can see that the F-statistic remains above 20 in all regressions reported in Table 3, including the more demanding fixed effects specifications. For a regression with one instrument and one endogenous regressor, Table 2 in Stock and Yogo confirms that this is sufficient power to mitigate concerns about weak instruments. 22

25 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Table 3: IV Estimates of Lawyer Impact on Chapter 13 Choice (1) (2) (3) (4) (5) (6) (7) (8) Census Census Census Census Texas Texas Texas Texas Lawyer Fraction ** 0.950** 0.748** 0.849** 0.959** 0.936** 0.928** 1.199** [0.018] [0.028] [0.193] [0.234] [0.031] [0.076] [0.243] [0.344] State By Year Dummy Variables No Yes Yes No No Yes Yes No Neighborhood Covariates No Yes No Yes No Yes No Yes Neighborhood Fixed Effects None None Block Group Tract* Year None None Block Group Tract* Year Personal Filing Covariates No No No No No Yes Yes Yes 2 Adjusted R Partial R 2 of Ch. 13 Filing Rate F-Test for Excluded Instrument 2, , Observations 121, , , ,416 15,293 15,293 15,293 15,293 * significant at 5%; ** significant at 1%. Robust standard errors in brackets, clustered at the zip code level. "Neighborhood Covariates" are those used in Table 2 Urban, Marital Status, Household Size, Education, Race, Age, Employment Status, Household Income, Homeownership, and Housing Value. Personal Filing Covariates are also the same as Table 2 Monthly Income, Land and Personal Assets, Secured and Unsecured Debt. Robust standard errors in brackets, clustered at the zip code level. Published by The Berkeley Electronic Press,

26 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 In the fourth column, we include tract by year fixed effects. This specification allows for unobserved neighborhood-specific changes over time and takes advantage only of the variation in geographic location within a neighborhood as defined by Census tract. According to the Census bureau, tracts are designed to be homogeneous with respect to population characteristics, economic status, and living conditions. 45 It therefore seems plausible that the tract identifies a set of households who could be expected to have reasonably similar attitudes towards bankruptcy. Our thought experiment corresponds to one in which two households on different sides of a neighborhood frequent different lawyers on account of the slight difference in convenience. This estimator still yields estimates very close to our baseline OLS and IV results. The last half of the table reports estimates from the Texas subsample. These specifications are similar to those in the first four columns but allow us to check the sensitivity of our IV estimates to the inclusion of household-level controls. The corresponding estimates are uniformly close to our baseline specifications, though higher. This suggests that our IV strategy is even robust to 46 controlling for the detailed financial situation of the filing household. To summarize, OLS and IV estimates, with and without fixed effects, all suggest that lawyers play a central role in household decisions regarding bankruptcy chapter. The estimate is also robust to allowing for idiosyncratic changes over time in unobserved neighborhood preferences. The similarity in our coefficients provides evidence that our results hold under a variety of identifying assumptions. To put the findings in perspective, the 25 th percentile lawyer in our comprehensive sample files no bankruptcies under Chapter 13 while the corresponding measure for the 75 th percentile lawyer is 41 percent. Even if we rely on a relatively conservative estimate, a fraction 13 coefficient of 0.9, happening to go to a 75 th percentile lawyer instead of a 25 th percentile lawyer increases the probability of filing under Chapter 13 by 37 percentage points. Parameter Heterogeneity The ability of a lawyer to affect a client s chapter choice may vary across clients. In this section, we examine whether there are heterogeneous responses across observable groups. It should be noted that our IV strategy identifies the average influence of lawyers on households that choose lawyers based largely on geographic convenience. One might suspect that such households are particularly susceptible to filing under the bankruptcy chapter which is in the attorney s best interests. This may explain why the IV estimates, in some cases, are larger than 45 See (accessed on March 27, 2009). 46 Again, we cannot rule out the possibility that the financial characteristics of the household should enter the bankruptcy decision process in a complex non-linear fashion. 24

27 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? the OLS estimates. 47 All estimations are performed without neighborhood fixed effects, as the previous results suggest that IV estimation gives similar results, but with more precision. In Table 4, we examine the heterogeneous responses across observable groups; all regressions control for neighborhood characteristics. In the first three rows, we use our comprehensive sample to examine the implied impact of lawyers in the three different bankruptcy districts in our sample. In the following rows we compare urban and rural, low and high income, and low and high minority concentration. There is very little evidence of heterogeneity in the parameter, which is above 0.75 in all the above specifications and typically between 0.9 and 1. The Bankruptcy Abuse Prevention and Consumer Protection Act of 2005 (BAPCPA) made several changes to the Bankruptcy Code. First, the BAPCPA increased the paperwork involved when filing for bankruptcy. For example, the means test was intended to preclude households with high income levels from filing under Chapter 7. As stated earlier, this restriction rarely applies. 48 Nevertheless, because of the means test all debtors must now detail their monthly income and expenses households filing under Chapter 7 must complete form B22A showing that their current monthly income is below the state median and households filing under Chapter 13 must complete form B22C particularizing their monthly income and expenses. Each entry requires thorough documentation; without the proper clerical system, disorganized lawyers may find the task daunting. Due to the increased work involved, many bankruptcy attorneys left the field. However, those attorneys who stayed were able to charge higher fees after the BAPCPA. A 2008 U.S. Government Accountability Office (GAO) report found that after the BAPCPA, the median attorney fees for a Chapter 13 bankruptcy increased from $2,400 to $3,000. Additionally, the report found that attorney fees for a Chapter 7 bankruptcy increased, on average, from $712 to $1,078. It is important to note that the structure of attorney fees did not change. 47 Likely for the same reason, the OLS and IV estimates differ even more when, in the next section, we examine bankruptcy outcomes. See Imbens and Angrist (1994). 48 As discussed above, the means test only applies to debtors whose average monthly income is greater than the state median income level; approximately six percent of debtors have income levels above the state median income level (White 2006). According to the means test, these debtors are allowed to deduct standardized expenditures from their income. After these deductions are applied, 90 percent of the debtors with income above the state median income level pass the means test (White 2006). Therefore, the means test keeps less than one percent of households from filing under Chapter 7 of the Bankruptcy Code. Published by The Berkeley Electronic Press,

28 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 Table 4: Parameter Heterogeneity Specification OLS IV (1) Utah [0.009] [0.164] (2) Texas [0.009] [0.043] (3) California [0.011] [0.026] (4) Urban [0.007] [0.030] (5) Rural [0.023] [0.106] (6) Low Income [0.017] [0.063] (7) High Income [0.016] [0.101] (8) Low Minority [0.009] [0.095] (9) High Minority [0.019] [0.058] (10) Pre-BAPCA [0.007] [0.033] (11) Post-BAPCA [0.019] [0.051] (12) No Prior Filings [0.017] [0.147] (13) Low Assets (Texas Sample) [0.030] [0.098] (14) High Assets (Texas Sample) [0.017] [0.098] All regressions are significant at the 1% level and include year and state dummies (where applicable) and neighborhood census block covariates. The "No Prior Filings" row includes only 2004 Utah data. Robust standard errors in brackets, clustered at the zip code level. 26

29 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Attorneys still typically receive all their fees upfront for a Chapter 7 bankruptcy. In addition, attorneys still collect higher fees for a Chapter 13 bankruptcy with part of the fees being collected upfront and part of the fees being collected in the Chapter 13 repayment plan. In rows 10 and 11 we separately examine cases filed prior to and after October If the less principled attorneys left the market after the BAPCPA was enacted, or if the law changed the lawyer-client relationship, our estimates would be different in the two periods. However, our estimates are similar before and after the BAPCPA. Households that previously filed a Chapter 7 bankruptcy are eligible only for Chapter 13 bankruptcy and thus cannot be swayed by their attorney. Additionally, because Chapter 13 filers are likely to have their cases dismissed, the composition of clients may differ between high and low fraction 13 attorneys or neighborhoods. More specifically, high fraction 13 attorneys may serve a higher fraction of households that are seeking to refile. In row 12, we deal with both concerns by using a small subsample of our data from Utah in 2004 from which we can exclude households with any prior filings. The IV coefficient is smaller than our baseline estimates at However this difference is present in large measure for all the Utah 2004 observations for which we have prior filing information, even when we do not exclude prior filers. Additionally, the standard errors in the IV specification are sufficiently large that the 95 percent confidence interval includes 0.9, our baseline estimate. Using more detailed financial records, we use our Texas subsample to look at households with few secured debts (less than $20,000). Under Chapter 13, the debtor can retain secured assets and even cure an arrearage on secured debt through the bankruptcy process. By comparison, it is likely that Chapter 7 only offers temporary relief through the automatic stay. 49 Once the automatic stay is lifted the creditor is free to enforce whatever right he had before bankruptcy. Therefore, we would expect households with high levels of secured debts (more than $20,000) to file under Chapter 13 and households with little collateral to file under Chapter 7. Indeed, 24 percent of households with less than $20,000 in secured debt file under Chapter 13 bankruptcy compared to 48 percent of households with more than this amount. Comparing the OLS results from the first column of rows 13 and 14, households with little collateral are noticeably less responsive to lawyer filing behavior than those with more assets at risk. Since these low-collateral households are well-suited for Chapter 7, it makes sense that they are more difficult to sway (or, equivalently, lawyers are less willing to try and sway them). This makes the IV results all the more interesting. For low-asset filers the IV estimate is 0.94 compared to 0.68 in the OLS specification. The larger IV estimate suggests that households that match to their attorney on the 49 As discussed in an earlier footnote, only five circuit courts allow ride-throughs, a process in which households can keep secured debt in a Chapter 7 bankruptcy. Published by The Berkeley Electronic Press,

30 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 basis of geographical convenience are more easily pushed into Chapter 13 bankruptcy even when they receive few financial benefits from doing so. Table 5: Robustness Checks of Instrumental Variables Estimates Census Texas (1) Probit Marginal Effects [0.040] [0.102] (2) Instrument not weighted by lawyer's # of filings [0.038] [0.105] (3) Firm defined by address [0.030] [0.078] (4) Firm size instrument [0.104] [0.197] (5) Fraction 13 computed using data from all years [0.026] [0.082] (6) Lagged lawyer fraction used to compute instrument [0.030] [0.072] All estimates use instrumental variables. All results are significant at the 1% level and include state by year dummies (where applicable), neighborhood census block covariates, and personal filing covariates for Texas regressions. Robust standard errors in brackets, clustered at the zip code level. Robustness Checks While the results in Tables 2 and 3 are remarkably stable, it is worth considering additional threats to our IV identification strategy. One concern is that our instrument, the distance-weighted average fraction 13, is correlated with the unobserved characteristics of the filing household. Our model of lawyer specialization (and evidence we present later) implies that firms tend to specialize in Chapter 13 bankruptcy when their client mix is better suited for this type of bankruptcy. This suggests that lawyer behavior will be correlated with observable and possibly unobservable neighborhood characteristics, potentially invalidating our instruments. Our principal mechanism to deal with this is neighborhood fixed 28

31 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? effects, which removes all unobserved variation operating at the neighborhood level. Additional evidence comes from examining households with fewer than $20,000 in secured debts. As Chapter 13 protects collateral, these households should find Chapter 13 relatively unattractive. If our instrument simply reflects the matching of local neighborhood needs and lawyer specialization, the lawyer choice of households with few secured debts should be insensitive to our instrument. The first stage coefficient for this subgroup is 1.08 compared to 1.01 for our full Texas sample. This suggests that even for households clearly suited for a specific type of bankruptcy, the geographic considerations captured in our instrument play a strong role in whether they select an attorney specializing in Chapter 13. Another concern is that our primary specification is misspecified because our outcome is binary. In row 1 of Table 5, we present the marginal effects of lawyer fraction 13 on chapter choice using an IV probit specification. The results again are qualitatively similar to our baseline estimates. Next, one may be concerned that the instruments weight each lawyer by number of filings, which itself may be endogenous. Thus we form the instrument without weighting by the number of lawyer filings. Row 2 uses this instrument and finds comparable results. In row 3, we define the law firm by the address of the filing attorney as opposed to the name. Thus multiple attorneys located at the same physical address are treated as a single entity. The IV results are again similar to the baselines. In row 4, we consider an instrument based solely on firm size. Our theoretical model suggests that larger firms will file a higher fraction of bankruptcies under Chapter 13 than smaller firms. Thus an individual who randomly walks into a law office that files a large number of bankruptcies has a high probability of filing under Chapter 13. This new instrument is defined as: (4) ln(f) j ln F j /d ij ı = j 1/d ij where d ij is the distance (in miles) from debtor i to firm j and F j is the number of filings for the attorney. This instrument yields estimates similar to our baseline. In results not shown, this result holds even when we control for block group or tract by year fixed effects The relationship between the firm-size instrument and the probability of filing under Chapter 13 is, unsurprisingly, highly non-linear. We increase the statistical precision by also including the square and cube of the instrument in the first stage. This only becomes important when we add neighborhood fixed effects which decrease instrument power. Published by The Berkeley Electronic Press,

32 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 In row 5, we calculate lawyer fraction 13 across all sample years. We use this long-run fraction 13 both for our regressor as well as our instrument. Again the results are very similar to our baseline. In row 6, we lag our instrument by one year to avoid contemporaneous shocks. Once again, the results support our prior conclusions. The Causes of Lawyer Specialization Some attorneys may specialize in one bankruptcy chapter to provide improved services to clients. Alternatively, attorneys may specialize because they have idiosyncratic beliefs regarding the relative merits of Chapter 7 or Chapter 13 bankruptcy. However, attorney specialization is also consistent with straightforward profit maximization. According to our model, firms specialize in Chapter 13 if they are relatively large. In Figure 3, we show the strong correlation between firm size and the fraction of bankruptcies filed under Chapter 13. Indeed, firms that file between 10 and 14 bankruptcies per year only file 15 percent of their cases under Chapter 13. On the other hand, firms filing more than 250 bankruptcies per year typically file over 40 percent of their cases under Chapter 13. A regression of fraction 13 on log firm size, controlling for neighborhood characteristics, reveals that as firm size increases by 10 percent, the fraction of bankruptcies filed under Chapter 13 rises by 0.8 percentage points. This coefficient is highly significant with a t-statistic of ,52 A second implication of our theoretical framework is that attorneys with a client mix well suited for Chapter 13 bankruptcy will specialize in Chapter 13. This will be mechanically true if we examine the correlation between fraction 13 and demographic characteristics (as measured by Census block level covariates). To address this issue, for our base sample, we regress a debtor s filing decision on lawyer fixed effects, θ j, and neighborhood covariates. A lawyer s empirical propensity to file under Chapter 13, holding client composition constant, is θ j. X j β represents the suitability of the attorney s client mix for Chapter 13 bankruptcy, where X j represents the average demographic characteristics of the neighborhoods around an attorney s clientele. To test whether lawyers with client mixes suited for Chapter 13 bankruptcy tend to specialize in Chapter 13, we regress θ j on X j β. The coefficient on X j β is 0.5 with a t-statistic of 6. This implies that if an attorney s case mix were 10 percentage points more likely to file 51 This finding is consistent with Sullivan, Warren, and Westbrook s (1988) result that attorneys specializing in bankruptcy law file a higher fraction of cases under Chapter 13 than attorneys who do not specialize in bankruptcy law. 52 An alternative explanation for this finding is that firms that promote Chapter 13 bankruptcy can attract more clients because less payment is required up front. See Mann and Porter (2010). 30

33 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? under Chapter 13, the attorney would be 5 percentage points more likely to file under Chapter 13, holding constant client characteristics. This is also consistent with our model of attorney specialization. The Consequences of Lawyer Specialization It appears that attorneys push individuals into Chapter 13 when the costs of doing so are low enough to justify the pursuit of the higher payment received for a Chapter 13. Although aggressive lawyers may induce households to file under Chapter 13 even when it is a poor financial choice, households filing under Chapter 13 may benefit from lawyer specialization through lower costs and/or lower dismissal rates. We examine these possibilities using a separate subset of our Texas data for which we have information about outcomes--- for these households, we have information on secured debt, case dismissal (if a case is dismissed, no debt relief Published by The Berkeley Electronic Press,

34 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 is granted), 53 reported monthly expenses, total legal fees, and upfront legal fees. Due to the higher data requirements, and the fact that we focus exclusively on Chapter 13 filers, the sample is noticeably smaller. Depending on the particular outcome, we have between 2,429 and 8,782 observations. Using this sample, we regress outcomes on the lawyer s fraction 13 and 54 the covariates used previously as controls. These results are shown in Table 6. We report both OLS and IV specifications for each outcome, providing progressively more controls in each regression. We include block group controls but not fixed effects. In the results that follow, it is important to keep in mind that the counterfactual is not the outcome if the household had filed under Chapter 7. Rather we compare the outcomes of a household that filed a Chapter 13 with an attorney who files many Chapter 13 bankruptcies as opposed to an attorney who files primarily Chapter 7 bankruptcies. Because some attorneys induce households to file under Chapter 13, the types of households found filing a Chapter 13 bankruptcy may differ across attorneys. Thus the coefficients in this table capture both causal effects on a given household as well as a change in the composition of households. As mentioned earlier, a major advantage of Chapter 13 bankruptcy is the 55 option to retain collateral (on which the debtor is current or overdue). Therefore, Chapter 13 bankruptcy would typically be an inappropriate choice for households with few secured debts. The first row of Table 6 examines the impact of going to a high fraction 13 lawyer on whether the household filing under Chapter 13 has less than $20,000 in secured debts. OLS estimates suggest that households that file a Chapter 13 bankruptcy with a high fraction 13 attorney are as likely to possess secured debts as those who file with a low fraction 13 attorney. The IV estimates, however, imply that Chapter 13 specialists file households under Chapter 13, even when their financial situation is poorly suited for this type of bankruptcy. Indeed, increasing a lawyer s fraction 13 by 10 percentage points is associated with a 4.3 percentage point increase in the probability that a household filing for Chapter 13 bankruptcy has few secured debts. Again, the IV results reflect the experience of those individuals who select their attorney on the basis of geographic convenience. Case dismissal occurs when the repayment plan is not accepted; as a result, the household does not receive long-term debt relief. To study the impact 53 This abstracts from the possibility that the household was able to pay back arrearages on collateral and thus retain the property even in the case of dismissal. 54 When examining whether or not the households have secured debt, we control only for blocklevel covariates to take into account neighborhood composition. The household-level covariates include measures of a household s actual financial situation, which are mechanically linked to the outcome variable. 55 As mentioned earlier, by comparison, Chapter 7 only allows households to retain collateral if they are current on the loan and live in a circuit that allows ride-throughs. 32

35 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? of attorney specialization on the likelihood of case dismissal, we restrict our sample cases filed prior to 2005; we restrict our sample to this time period so that the cases had time to be dismissed. OLS estimates suggest that households filing for Chapter 13 bankruptcy with high fraction 13 attorneys experience a slightly higher probability of dismissal than households that file with low fraction 13 attorneys. The IV point estimates are noticeably larger, although much noisier, and are only significant at the ten percent level. Increasing attorney fraction 13 by 10 percentage points is associated with a 3.5 percentage point increase in the probability of dismissal. This suggests that Chapter 13 specialists may steer households to file under Chapter 13 bankruptcy even when the probability of long-term debt relief is small Each bankruptcy filing includes a statement of monthly income and budgeted monthly expenditures. The household must pay the difference between these two numbers into the payment plan. 56 Thus, controlling for income, households that report a higher set of monthly expenditures may end up with smaller payments. While the proposed plans need not be accepted by the court, they generally are. There is no evidence that log expenditures vary much by lawyer. The point estimates are small and insignificant, though sometimes positive. Thus this metric shows no evidence of lawyer specialization aiding the client. Finally, our dataset contains information about the amount and timing of Chapter 13 attorney fees. As noted above, even when payments are paid through the plan, lawyer fees are paid in the early months of the plan. In row 4 we regress total fees on the lawyer s fraction Chapter 13. Lawyers who file many cases under Chapter 13 tend to charge slightly more on average about $50 more. This is not a large premium since, in the Northern Texas Bankruptcy District, the average Chapter 13 bankruptcy cost about $2,000 before the BAPCPA and $3, after. Examining the data, it becomes clear that experienced lawyers are charging the court-accepted maximum rate, while lawyers who file few Chapter 13 bankruptcies sometimes charge less. The only evidence of a benefit being passed on to filers comes in the last row---ols results show that Chapter 13 specialists charge less upfront than other lawyers. As most bankrupt households are short on funds and credit, this could be a substantial short-term benefit. However, there is little long-term value given that our random sample of As mentioned above, prior to 2005, households subtracted actual monthly expenses. After 2005, households with income above the state median subtracted standard deductions while households with income below the state median still subtracted actual monthly expenses. Thus, after controlling for income, both before and after the BAPCPA, households that report a higher set of monthly expenditures may end up with smaller payments. 57 We restrict our data for this regression to cases filed before the BAPCPA s enactment so as to eliminate the price change as a potential confounder. Published by The Berkeley Electronic Press,

36 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 dismissed filings suggests low rates of default on debt to the lawyer. Furthermore, the IV results are smaller, with the preferred specification estimating a drop of only $160 in up-front fees. Table 6: Does Lawyer Specialization Improve Bankruptcy Outcomes (Conditional on Choosing Chapter 13)? Specification OLS IV Dependent Variable Bivariate Neighborhood Controls Full Controls Bivariate Neighborhood Controls Full Controls (1) Few Secured Debts ** 0.43** n= 2,429 [.03] [.03] [.15] [.11] (2) Case Dismissed * n= 6,503 [.06] [.05] [.05] [.22] [.21] [.20] Log Monthly (3) Expenditure n= 3,865 [.03] [.03] [.03] [.08] [.06] [.06] (4) Lawyer Fees 61** 61** 69** 105* 77* 102* n= 3,865 [18] [19] [19] [37] [38] [39] (5) Up Front Fees 314** 313** 296** -226* -174* 160* n= 6,503 [26] [26] [25] [98] [72] [71] * significant at 5%; ** significant at 1%. Robust standard errors in brackets, clustered at the zip code level. All data are from Texas dismissal regression is only for 2003 and 2004 data; fee regressions are pre-bapca law change. Bivariate regressions include year dummies but no other controls. "Neighborhood controls" are those used in Table 2 Urban, Marital Status, Household Size, Education, Race, Age, Employment status, Household Income, Homeownership, Housing Value, and state by year dummies. "Full Controls" include the Neighborhood Controls plus the personal filing information included in Table 2 Monthly Income, Land and Personal Assets, Secured and Unsecured Debt. 34

37 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? In total, there does not appear to be much evidence that lawyers who specialize in Chapter 13 bankruptcies secure better outcomes for their clients. They systematically push clients towards Chapter 13 bankruptcy even when the probability of dismissal is high as is the probability that the household would receive few financial benefits. We cannot rule out that the (negative) effect of pushing less appropriate clients into Chapter 13 may be masking smaller positive effects of the attorney being more skilled at Chapter 13 filings. However, we can estimate that the aggregate effect leads to higher dismissal rates for the average client. Of course, all of this is conditional on the household choosing to file under Chapter 13. The simple act of moving a household from filing Chapter 7 to filing Chapter 13 vastly increases their dismissal rate and their fees. Table 7 shows the impact of attorney fraction 13 on the dismissal rate and total attorney fees, independent of chapter choice. The instrumental variables estimates indicate that a 10 percentage point increase in attorney fraction 13 causes a 4.8 percentage point increase in the probability of case dismissal. Although up-front fees decline by over $1,100 dollars, total fees rise by $600 to $700. As mentioned above, our Texas sample suggests that lawyers recover the majority of their fees even when they do not collect them upfront. The OLS estimates give comparable but smaller results. Since a conventional endogenous matching story suggests the OLS estimate is biased upwards, one interpretation for the larger IV results is that our IV specification targets households that are particularly dependent on lawyer advice. Conclusion and Policy Implications Choosing between Chapter 13 and Chapter 7 bankruptcy involves critical tradeoffs central to a household s future financial welfare. While Chapter 13 bankruptcy allows households to keep secured assets, 58 discharge a broader set of debts, and preserve the option of a future Chapter 7 bankruptcy, 59 most Chapter 13 filings are ultimately dismissed or converted to Chapter 7. Because the failure rate of Chapter 13 bankruptcies is so high, it is important to understand what factors drive the household decision regarding bankruptcy chapter. Our analysis confirms earlier qualitative work that attorneys play a crucial role in helping their clients weigh the costs and benefits of each bankruptcy alternative. Ideally, a household would receive the same appropriate advice, 58 As already mentioned, Chapter 13 allows households to keep all secured assets (regardless of whether they are current or overdue). By comparison, Chapter 7 only allows households to retain collateral if they are current and live in a circuit that allows ride-throughs. 59 This would be particularly valuable if households experience income or expense shocks after the initial Chapter 13 bankruptcy filing. Published by The Berkeley Electronic Press,

38 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 regardless of which attorney they consult. Unfortunately, this does not appear to be the case. By far, the best observable predictor of a household s decision to file under Chapter 13 is the attorney they happen to consult. This is true even when we control for neighborhood fixed or time-varying effects and household-level Table 7: Does Lawyer Specialization Improve Bankruptcy Outcomes (Unconditional)? OLS Specification IV Dependent Variable Bivariate Neighborhood Controls Full Controls Bivariate Neighborhood Controls Full Controls (1) Case Dismissed n= 5,144 [.03] [.02] [.09] [.10] (2) Lawyer Fees n= 8,782 [29] [30] [30] [106] [111] [108] (3) Up Front Fees n= 8,782 [34] [34] [36] [126] [132] [134] All regressions are significant at the 1% level. All data are from Texas-- dismissal regression is only for 2003 and 2004 data; fee regressions are pre-bapca law change. Bivariate regressions include year dummies but no other controls. "Neighborhood controls" are those used in Table 2 Urban, Marital Status, Household Size, Education, Race, Age, Employment status, Household Income, Homeownership, Housing Value, and state by year dummies. "Full Controls" include the Neighborhood Controls plus the personal filing information included in Table 2 Monthly Income, Land and Personal Assets, Secured and Unsecured Debt. Robust standard errors in brackets, clustered at the zip code level. covariates include assets and liabilities. We obtain the same estimates when we instrument the fraction of cases the chosen attorney files under Chapter 13 with a distance-weighted average of fraction 13s of attorneys in the area. Even relying on intertemporal variation in attorney behavior within a neighborhood yields similar estimates. Identification which relies on residential location within a census tract also yields similar estimates. Moving from a lawyer at the 25 th percentile to a lawyer at the 75 th percentile in the distribution increases the probability that a household files under Chapter 13 by 37 percentage points. 36

39 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Attorney specialization can be explained in part by a model in which firms undertake fixed investments to lower the marginal costs of Chapter 13 cases. However, it seems that the benefits of specialization are enjoyed by the attorney as opposed to the clients. Households that file a Chapter 13 bankruptcy with an attorney specializing in that type of bankruptcy do not have higher success rates, more lenient payment plans, or lower attorney fees. Indeed, IV estimates suggest that Chapter 13 specialists steer households into Chapter 13 bankruptcy even when they are objectively poorly suited for this type of bankruptcy and have a high probability of case dismissal. A typical lawyer-instigated switch from Chapter 7 to Chapter 13 leads to a substantial transfer of wealth from insolvent households to specialized attorneys with a reduced probability of long-term debt relief. With these findings, our paper joins a growing literature on the nature of principal-agent relationships when the agent possesses specialized knowledge. It would appear that many professionals are willing to sacrifice their clients best interests to maximize firm profits. Due to the high cost and low success rate of Chapter 13 bankruptcy, we have focused our analysis on the adverse consequences of being inappropriately pushed into Chapter 13 bankruptcy. Chapter mismatch can go the other way as well. A firm that has not undertaken investments to lower the cost of Chapter 13 bankruptcy may encourage a household to file under Chapter 7 bankruptcy, when Chapter 13 better suits the households specific needs. For instance, a household may be able to retain an automobile or a house under a successful Chapter 13 bankruptcy, which they would give up under Chapter 7; however, this depends a great deal on having a successful Chapter 13 filing. Additionally, a Chapter 7 bankruptcy may conflict with a household s view of ethical behavior. The data to investigate such possibilities is not provided in publicly available court records, however. Additionally, Chapter 7 s low cost and low dismissal rate probably limit the adverse consequences of this type of chapter mismatch for most households. One potential solution to the mismatch between households and bankruptcy chapter is to better align attorney incentives with those of their clients. For example, spacing out the attorney payments more evenly over the course of the payment plan gives attorneys a smaller incentive to push households into Chapter 13 bankruptcy when the probability of dismissal is high. Another possibility is to simplify the bankruptcy law and allow only a single chapter for consumer bankruptcies. Under such a policy, all households would be treated identically, regardless of their choice of attorney. Braucher (2006) proposes this type of policy change. Published by The Berkeley Electronic Press,

40 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 Appendix Table A1: Full OLS Estimates of Lawyer Impact on Chapter 13 Choice (1) (2) (3) (4) (5) (6) (7) (8) Census Census Census Census Texas Texas Texas Texas Lawyer Fraction ** 0.958** 0.955** 0.957** 0.905** 0.763** 0.745** 0.740** [0.006] [0.006] [0.007] [0.007] [0.017] [0.016] [0.019] [0.015] Urban [0.007] [0.008] [0.017] [0.020] Married 0.078** [0.026] [0.027] [0.071] [0.074] Divorced [0.043] [0.044] [0.110] [0.113] Household of * 0.172* [0.030] [0.032] [0.077] [0.079] Household of [0.036] [0.038] [0.090] [0.092] Household of * 0.262** [0.039] [0.040] [0.095] [0.097] Household of [0.045] [0.049] [0.120] [0.120] Household of [0.049] [0.053] [0.155] [0.161] Household over [0.059] [0.067] [0.177] [0.183] Finished High School 0.073* ** [0.032] [0.034] [0.068] [0.072] Finished college ** **

41 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? [0.021] [0.026] [0.064] [0.073] Black 0.191** 0.203** 0.178** 0.216** [0.019] [0.020] [0.025] [0.027] Hispanic * [0.020] [0.023] [0.041] [0.047] Other Race [0.017] [0.024] [0.077] [0.085] Age Below [0.062] [0.065] [0.173] [0.173] Age 6 to [0.048] [0.051] [0.132] [0.132] Age 19 to [0.053] [0.055] [0.116] [0.114] Age 25 to [0.052] [0.053] [0.141] [0.142] Age 30 to [0.040] [0.044] [0.098] [0.106] Age 40 to [0.045] [0.048] [0.112] [0.113] Age 50 to [0.048] [0.049] [0.131] [0.134] Unemployed [0.069] [0.071] [0.145] [0.154] Self-Employed [0.028] [0.030] [0.064] [0.065] HH income Under $10, * [0.044] [0.051] [0.113] [0.114] HH income $10-$20, * [0.039] [0.042] [0.114] [0.121] HH income $20-$30, * [0.038] [0.043] [0.109] [0.109] Published by The Berkeley Electronic Press,

42 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 HH income $30-$40, * [0.034] [0.038] [0.096] [0.097] HH income $40-$50, [0.034] [0.038] [0.099] [0.099] HH income $50-$60, [0.035] [0.036] [0.104] [0.103] HH income $60-$75, * * * [0.034] [0.035] [0.095] [0.092] HH income $75-$100, * * [0.034] [0.034] [0.095] [0.089] Fraction Homeowners 0.053** 0.062** [0.012] [0.013] [0.025] [0.025] 25th Percentile of Log Housing Value * [0.008] [0.009] [0.017] [0.019] 75th Percentile of Log Housing Value [0.008] [0.009] [0.020] [0.021] Log Monthly Income 0.099** 0.105** 0.100** [0.008] [0.010] [0.008] Dummy- No income 0.602** 0.666** 0.620** [0.064] [0.086] [0.065] Log Land Assets * [0.005] [0.007] [0.005] Dummy- no Land Assets * [0.052] [0.074] [0.054] Log Personal Assets ** ** ** [0.004] [0.006] [0.004] Dummy- no Personal Assets [0.088] [0.123] [0.090] Log Secured Debts 0.056** 0.052** 0.056** [0.005] [0.007] [0.005] Dummy- no Secured Debts 0.316** 0.279** 0.325** 40

43 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? [0.044] [0.062] [0.046] Log Unsecured Debts ** ** ** [0.003] [0.004] [0.003] Dummy- no Unsecured Debts ** ** ** [0.038] [0.051] [0.038] State by Year Fixed Effects No Yes Yes No No Yes Yes No Neighborhood Fixed Effects None None Block Group Tract*Year None None Block Group Tract*Year 2 Adjusted R Partial R 2 of Ch. 13 Filing Rate Observations 121, , , ,416 15,293 15,293 15,293 15,293 * significant at 5%; ** significant at 1%. Robust standard errors in brackets, clustered at the zip code level. Published by The Berkeley Electronic Press,

44 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 References Allgood, P.C. and M.O. Bachman (2006) Effects of Specialization on Treatment and Outcomes In Screen-Detected Breast Cancers in Wales: Cohort Study. British Journal of Cancer 94, pp Bergstresser, D. B., J. Chalmers, and P. Tufano (2009) "Assessing the Cost and Benefits of Brokers: A Preliminary Analysis of the Mutual Fund Industry." Review of Financial Studies 22(12): pp Braucher, J. (1993) Lawyers and Consumer Bankruptcy: One Code, Many Cultures. American Bankruptcy Law Journal 67(4), pp Braucher, J. (2006) A Fresh Start for Personal Bankruptcy reform: the Need for Simplification and a Single Portal. American University Law Review 55(5), pp Carlin, B. I. and S. Gervais (2008) Legal Protection in Retail Financial Markets. Working Paper. Chandra, A. and D. Staiger (2007) Testing A Roy Model With Productivity Spillovers: Evidence From the Treatment of Heart Attacks. Journal of Political Economy 115(1), pp Chevalier, J. and G. Ellison (1997) Risk Taking by Mutual Funds as a Response to Incentives. Journal of Political Economy 114(2), pp Culhane, M. B. and M. M. White (1999) Taking the New Consumer Bankruptcy Model for a Test Drive: Means-Testing Real Chapter 7 Debtors. American Bankruptcy Institute Law Review 7 (1). Domowitz, I. and R.L. Sartain (1999) Determinants of the Consumer Bankruptcy Decision. Journal of Finance 54 (1), pp Epstein, A., J. Ketcham and S. Nicholson (2005) The Welfare Effect of Physician Specialization: Do We Want Physicians to Practice Alike? Working Paper. Flynn, E. and G. Bermant (2000) Pre-Bankruptcy Planning Limits Means- Testing Impact. American Bankruptcy Institute Journal 19 (2), pp. 22. Fuchs, V. R. (1978) The Supply of Surgeons and the Demand for Operations. Journal of Human Resources 13 Supplement: National Bureau of Economic Research Conference on the Economics of Physician and Patient Behavior, pp Gruber, J. and M. Owings (1996) Physician Financial Incentives and Cesarean Section Delivery. Rand Journal of Economics 27(1), pp

45 Lefgren et al.: Chapter 7 or 13: Are Client or Lawyer Interests Paramount? Grytten, J. and R. Sørensen (2003) Practice Variation and Physician-Specific Effects. Journal of Health Economics 22(30), pp Harrington, D. and K. Krynski (2002) The Effect of State Regulations on Cremation Rates: Testing for Demand Inducement in Funeral Markets. Journal of Law and Economics 45(1), pp Hogan, C. (2008) Will the Ride-Through Ride Again? Columbia Law Review 108(4), pp Imbens, G.W. and J.D. Angrist (1994) Identification and Estimation of Local Average Treatment Effects Econometrica 62(2), pp Inderst, R. and M. Ottaviani (2010) Intermediary Commissions and Kickbacks. Working Paper. Lefgren, L. and F. McIntyre (2009) Explaining the Puzzle of Cross-State Differences in Bankruptcy Rates. Journal of Law and Economics 52(2), pp Levitt, S. D. and C. Syverson (2008) Market Distortions when Agents Are Better Informed: The Value of Information in Real Estate Transactions. The Review of Economics and Statistics 90(4), pp Li, W. and P. Sarte (2002) The Macroeconomics of U.S. Consumer Bankruptcy Choice: Chapter 7 or Chapter 13? Federal Reserve Bank Working Paper # Mann, R. J. and K. M. Porter (2010) Saving Up for Bankruptcy. University of Iowa Legal Studies Research Paper No McGuire, T.G. (2000) Physician Agency. in Culyer, A.G., and J.P. Newhouse (eds.) Handbook of Health Economics, North Holland pp Miller, L.E. and M.M Miller (2008) Repeat Filers under the BAPCPA: A Legal and Economic Analysis Norton Annual Survey of Bankruptcy Law. National Bankruptcy Review Commission Report 90 (1997) Report of the National Bankruptcy Review Commission. Retrieved March 2010 from Nelson, J.P. (1999) Consumer Bankruptcy and Chapter Choice: State Panel Evidence. Contemporary Economic Policy 17, pp Neustadter, G. (1986) When Lawyer and Client Meet: Observations of Interviewing and Counseling Behavior in the Consumer Bankruptcy Law Office. Buffalo Law Review 35(1), pp Published by The Berkeley Electronic Press,

46 The B.E. Journal of Economic Analysis & Policy, Vol. 10 [2010], Iss. 1 (Advances), Art. 82 Norberg, S. F. and A. J. Velkey (2006) Debtor Discharge and Creditor Repayment in Chapter 13. Creighton Law Review 39(3), pp Shea, J. (1997) Instrument Relevance in Multivariate Linear Models: A Simple Measure. The Review of Economics and Statistics. Vol. 79(2), pp Stock, J. H. and and M. Yogo (2002) Testing for Weak Instruments in Linear IV Regression. NBER Technical Working Paper Series. No Sullivan, T.A., E. Warren and J.L. Westbrook (1988) Laws, Models, and Real People: Choice of Chapter in Personal Bankruptcy. Law & Social Inquiry 13(4), pp Sullivan, T.A., E. Warren and J.L. Westbrook (1989) As We Forgive Our Debtors. Oxford University Press: New York, New York. Sullivan, T.A., E. Warren and J.L. Westbrook (1994) The Persistence of Local Legal Culture: Twenty Years of Experience From the Federal Bankruptcy Courts. Harvard Journal of Law and Public Policy 17(3), pp Sullivan, T.A., E. Warren and J.L. Westbrook (2000) The Fragile Middle Class. Yale University Press: New Haven, Connecticut. Sullivan, A.C. and D.B. Worden (1990) Rehabilitation or Liquidation: Consumers Choices in Bankruptcy. Journal of Consumer Affairs 24(1), pp Tabb, C.J. and J. McClelland (2007) Living with the Means Test. University of Illinois Law and Economics Research Paper # LE United States Government Accountability Office (2008) Dollar Costs Associated with the Bankruptcy Abuse Prevention and Consumer Protection Act of 2005, GAO White, C.J. (2006) Oversight of the Implementation of the Bankruptcy Abuse Prevention and Consumer Protection Act. statement made on December 6, 2006 to the Committee on the Judiciary Subcommittee on Administrative Oversight and the Courts United States Senate. White, C.J. (2007) Making Bankruptcy Reform Work: A Progress Report in Year 2. American Bankruptcy Institute Journal 26(5), p. 16. White, M.J. (2005) Economic Analysis of Corporate and Personal Bankruptcy Law. NBER Working Paper #