The consumer provisions pending in the proposed reform legislation rely heavily on projected increases in chapter 13 filings as a vehicle to return large amounts of money to general unsecured creditors. There are four reasons to believe that this reliance is overly optimistic: (1) The pool of chapter 7 debtors who would move into chapter 13 don't have the capacity to repay at the levels reform proponents hope for; (2) Even if current chapter 7 debtors who were means-tested into chapter 13 performed at the level of the average current voluntary chapter 13 debtors (a very big "if"), they would not repay the amounts of unsecured debt that have been claimed as the fruits of means testing; (3) The various deductions and exclusions in the most recent draft bills set rather easy targets for pre-bankruptcy planning that will allow debtors to remain in chapter 7; and (4) Anti-lienstripping language in the legislation significantly reduces the chapter 13 incentives for debtors who might otherwise select the chapter.²

Given the confidence placed on chapter 13 as an ambulance to rescue unsecured creditors, it is a good time to kick the tires. Perhaps the most compelling characteristic of chapter 13 is its regional variability along almost every important dimension of the practice. Chapter 13 filing rates remain relatively stable over time at about 30 percent of total filings. Completion rates hover nationally at about one-third of confirmed plans, but this national average is a composite made up of extremely variable figures arising from different courtrooms, divisions and districts. For example, between 1989-1999, Tennessee displayed chapter 13 consumer-case filing percentages ranging between 55.5 and 65.9 percent, while Massachusetts ranged between 12.5 and 17.6 percent.³ Comparable variations can be found in almost every important part of the practice.

Wide variation in chapter 13 practice was a cause of concern for the National Bankruptcy Reform Commission. A strongly dissenting minority didn't dispute the findings about variability, but opposed the remedies proposed by the majority. The published dissents presaged the means-testing legislative proposals developed in the House and Senate during the last three years.⁴

Unless one takes the position that consumer bankruptcy practice is everywhere a reflection of a system perfectly attuned to local needs and abilities, large variations in chapter 13 performance should be a matter of continuing policy concern. The silver lining in such variability is that districts and states can be viewed as laboratories in which practices are being tested against the norms of chapter 13 success.⁵

This assumes that there is consensus on the norms for chapter 13 success. What are the appropriate measures of chapter 13 performance that allow districts to be compared with each other on all the important dimensions of the practice? Here, with one example, we illustrate the problem, show a solution, and indicate that the solution raises its own questions and points to other problems.

Disbursements to Creditors

It should be uncontroversial that one important norm of chapter 13 is to return as much money as is reasonable to creditors, including general unsecured creditors. The reports of the standing trustees to the Executive Office for U.S. Trustees (EOUST) contain meticulous records of these returns, so it is a fairly straightforward matter to aggregate the reports to arrive at a description of where the money comes from as the system now operates.⁶

For the 12 months ending Sept. 30, 1998, standing trustees disbursed approximately $2.9 billion, of which $2.5 billion went to secured, priority and unsecured creditors.⁷ Table 1 shows the five states with the largest disbursements, the five states with the lowest disbursements, and the six states in the middle of the distribution.

The table shows that the top five states contributed more than 45 percent of the $2.5 billion disbursed for the entire country. The mean amount per state was slightly more than $52 million, and the median, falling between the values for Massachusetts and Kentucky, was $27.5 million. Thus, Tennessee, with a population approximately equal to the population of Massachusetts, disbursed almost 11 times more money to chapter 13 creditors. Indeed, standing alone, Tennessee generated more than 10 percent of the national total disbursed to creditors.

Table 1: Total Payments to All Creditors by State, FY 98

High Five		Middle Six		Low Five
		Minnesota+N.D.1	$33,319,502
Tenn.	$303,424,262	P.R.+Virgin Islands2	$28,884,891	R.I.	$2,176,580
Texas	$255,751,205	Massachusetts	$27,921,556	Hawaii	$1,903,131
Georgia	$248,511,363	Kentucky	$27,161,966	Vermont	$1,400,215
Calif.	$231,785,864	Arizona	$25,078,583	Alaska	$1,262,719
Florida	$119,442,740	Oregon	$24,690,265	S.D.	$1,037,949
1 Administered jointly in FY 98. 2 Administered jointly in FY 98.

Table 2 shows the top five, middle six and bottom five states in terms of disbursements to unsecured creditors. The total nationwide disbursements to unsecureds was $536.3 million, of which the top five states contributed 41 percent. Tennessee by itself contributed 10 percent. There is considerable overlap between Tables 1 and 2 at the top and bottom of the distribution.

So far, then, we see that a small number of states contribute the lion's share of disbursements to creditors, and that payments to unsecured creditors, though not a large percentage of payments to all creditors, nevertheless track those payments reasonably closely.

Table 2: Total Payments to Unsecured Creditors, FY 98

High Five		Middle Six		Low Five
		South Carolina	$8,419,155
Tenn.	$52,501,851	Massachusetts	$7,083,781	Conn.	$608,483
Calif.	$49,400,612	Oklahoma	$6,884,671	R.I.	$580,737
Texas	$43,726,556	Oregon	$6,600,858	S.D.	$506,284
Georgia	$41,037,131	Utah+Wyoming3	$5,666,161	Alaska	$253,737
Ohio	$31,446,399	Arkansas	$5,577,167	Vermont	$223,880
3 Administered jointly in FY 98.

Case Volume and Per-case Yield

Total disbursements to creditors reflect two variables: the number of cases paying during the year multiplied by the average payments per case—in other words, case volume times yield per case. Equivalently, per-case yield equals total disbursements divided by case volume. In the current systems of record-keeping and reporting by courts, trustees and U.S. Trustees, there is no publicly available national database in which payments per individual case and case duration are linked together with a case identifier. Because chapter 13 cases last for up to five years, the calculation of per-case yield requires an estimate of the appropriate case volume to use as a denominator.

Per-case yield is an important measure of chapter 13 performance because it permits comparisons between states with large and small case volumes. This is an essential step, because otherwise, we are faced with a serious confound between case volume and state population. As shown in Table 1, with the obvious exception of Tennessee, there is a strong relationship between the population of a state and the amount of chapter 13 money generated in the state. California, Texas, Florida and Georgia rank 1, 2, 4 and 10, respectively, and Tennessee ranks 16. The ranks of the middle six states ranged from 13 to 28, and those of the bottom five from 42 to 49. The predominant mediating variable is chapter 13 volume. Table 3 shows the average number of chapter 13 filings during the years 1995-1999 for the high five, middle six and low five states in the distribution. These numbers, calculated for all the states, served as the denominators for our calculations of per-case yield.⁸

Comparisons of Table 3 with Tables 1 and 2 show obvious connections between case volumes and total disbursements, particularly at the extremes of the distributions. But there are also some exceptions and details that, as is often the case in bankruptcy, may turn out to be more interesting than the rule.

Table 3: Average Chapter 13 Filings, 1995-1999

High Five		Middle Six		Low Five
		Utah+Wyoming	4,794
Georgia	36,765	Arizona	4,396	N.H.	272
Calif.	36,057	Minnesota+N.D.	4,285	R.I.	239
Texas	29,913	Oregon	3,474	Vermont	123
Tenn.	28,645	Kentucky	3,406	Alaska	121
Florida	14,412	Oklahoma	3,293	S.D.	110

Table 4 shows the top five, middle six and bottom five states in terms of per-case yields to all creditors. Table 5 displays the same rankings for per-case yields to unsecured creditors.

Table 4: Per-case Yields to All Creditors, FY 98

High Five		Middle Six		Low Five
		Indiana	$7,874
Michigan	$12,010	Minnesota+N.D.	$7,776	D.C.	$4,685
Wash.	$11,796	Illinois	$7,514	Hawaii	$5,086
Okla.	$11,555	Louisiana	$7,438	Maryland	$4,496
Vermont	$11,347	Kansas	$7,427	N.J.	$4,080
W. Va.	$11,233	Mississippi	$7,408	P.R.	$2,942

Table 5: Per-case Yields to Unsecured Creditors, FY 98

High Five		Middle Six		Low Five
		Louisiana	$1,859
S.D.	$4,603	Idaho	$1,841	Arkansas	$919
Iowa	$3,527	Tennessee	$1,833	N.M.	$900
Ohio	$2,992	Vermont	$1,814	Pa.	$878
Kentucky	$2,947	Missouri	$1,814	N.J.	$763
W. Va.	$2,882	Illinois	$1,798	Conn.	$443

It is quite clear that this measurement changes the cast of characters acting in the ranks of the top, bottom and middle levels of chapter 13 performance. When the effects of caseload per se are removed from the equation, both large and small states can be found at both ends of the distributions as well as in the middle. South Dakota, the state with the lowest case volume, reappears as the national leader. Tennessee, far and away the national leader in total disbursements, is in the middle of the pack on a per-case yield basis to unsecured creditors.

What Does This Mean?

Readers will have their own explanations for why some of these states are positioned as they are or how case volume and per case yield relate to each other. In respect to per-case yield differences among jurisdictions,⁹ numerous explanations are theoretically possible and, in the absence of definitive data, plausible a priori. Here, for example, are a few among the possible explanations that are generally compatible with the data but not necessarily with each other:

• Chapter 13 filers in states with very high per-case yields have more disposable income than filers in states with low yields (likely to be false, given the demographics of the states in question);
• Filers in states with high per-case yields to unsecured creditors stay in their plans longer (see the next section for a test of this idea);
• States with high per-case yields for all creditors but low yields for unsecured creditors reflect a prevalence of plans that are dismissed or converted shortly after mortgage and other secured and or priority debt arrearages are cured (more information is required to determine this);
• States with high per-case yields for unsecured creditors reflect a practice of distributing unsecured payments across the entire duration of the plan, rather than beginning them after other expenses and debts have been paid (more information is required to determine this);
• States with high per-case yields for all creditors and unsecured creditors reflect more active management by standing trustees, including, for example, paying ongoing mortgage payments either with or without a fee attached (more information required to determine this);
• Some combination of the above factors.

Rates of Plan Completion

Beginning with their FY 98 reports to the Executive Office of U.S. Trustees, standing trustees have reported the percentage of terminating cases that were completed, converted, dismissed or granted a hardship discharge. This information allows an initial exploration of the relationship between the percentage of cases that complete and the per-case yield to unsecured creditors. Table 6 repeats the information in Table 5 and adds columns showing the percentages of cases terminated by completion during FY 98. Table 7 transposes the logic of Table 6, showing the top five, middle six and low five states in terms of percentage completions, and adds columns showing the related per-case yields to unsecured creditors. The two tables together give a fuller representation of a possible relationship between case completion rate and per-case yields to unsecured creditors.

For Table 6, the average completion rates for the top five, middle six and bottom five per-case yield states are 37, 31 and 22 percent, respectively. For Table 7, the average per-case yield to unsecured creditors for the top five, middle six and bottom five percent successful-completion states are $2,717, $2,117 and $1,161, respectively. Note that there are large and small states spread throughout the tables, both in terms of overall population and in terms of chapter 13 case volume. The relationships shown in Tables 6 and 7 may be related to factors endogenous to the debtors, case-management practices or both. But they strongly support the conclusion that returns to unsecured creditors are higher when plans are completed.

Table 6: Per-case Yields to Unsecured Creditors (% Successful Completions), FY 98

High Five		Middle Six		Low Five
		Louisiana	$1,859 (29%)
S.D.	$4,603 (17%)	Idaho	$1,841 (40%)	Arkansas	$919 (31%)
Iowa	$3,527 (41%)	Tennessee	$1,833 (28%)	N.M.	$900 (30%)
Ohio	$2,992 (43%)	Vermont	$1,814 (34%)	Pa.	$878 (17%)
Kentucky	$2,947 (37%)	Missouri	$1,814 (25%)	N.J.	$763 (15%)
W. Va.	$2,882 (47%)	Illinois	$1,798 (29%)	Conn.	$443 (15%)

Table 7: Percent Successful Completions (Per-case Yields to Unsecured Creditors), FY 98

High Five		Middle Six		Low Five
		Minnesota+N.D.	29% ($2,640)
W. Va.	47% ($2,882)	Louisiana	29% ($1,859)	D.C.	16% ($1,191)
Oregon	46% ($1,900)	Washington	28% ($2,635)	Conn.	15% ($443)
Nebraska	44% ($2,288)	Tennessee	28% ($1,833)	Alaska	15% ($2,094)
Ohio	43% ($2,992)	Nevada	27% ($1,648)	N.J.	15% ($763)
Iowa	41% ($3,527)	Oklahoma	27% ($2,091)	Florida	11% ($1,316)

In one sense, this is not a surprising result, given a prevailing view that unsecured creditors are served late in chapter 13, if at all. But if returns to unsecured creditors are a norm to be honored in chapter 13, and if case management procedures, beginning with plan construction and continuing with plan oversight, lead to higher plan completions, then the relationship demonstrated here counsels attorneys, trustees and courts to develop and practice such procedures. Given a national completion rate of only about one-third, it seems there is some distance yet to go.

---

Footnotes

¹ All views expressed in this article are those of the authors, and do not necessarily represent the views of the Executive Office for U.S. Trustees. Return to article

² See previous issues of this column for documentation of points 1, 2 and 3; see, also, Culhane, Marianne B., and White, Michaela M., "Taking the New Consumer Bankruptcy Model for a Test Drive: Means Testing Real Chapter 7 Debtors," 7 ABI Law Review 27 (1999). The anti-lienstripping language is contained in a pending amendment to 11 U.S.C. §1325(a) that renders §506 inapplicable to §1325(a)(5) for a purchase money interest in an automobile acquired within five years before filing. See Hildebrand, Hank, "Survey Shows Big Impact of Anti-Lienstripping Provision in S. 625," http://www.abiworld.org/legis/bills/99mayhildebrandsurvey.html (May 27, 1999). We recognize that it may be naive to assert that the intent of the legislation is tied necessarily to realizing high paybacks to general unsecured creditors. Taken all together, the credit counseling, means testing, debtor education, tax form reporting and auditing, anti-lienstripping and extended chapter 13 plan duration provisions create a climate that could significantly reduce consumer filings—and this might satisfy the fundamental intention of the proponents. Return to article

³ See http://www.usdoj.gov/ust/statistics/stats-new/05/statistics5.htm. Return to article

⁴ National Bankruptcy Review Commission, "Bankruptcy: The Next Twenty Years" (Oct. 20, 1997). See, especially, pages 233-302 and the several dissenting reports on consumer bankruptcy. See, also, Braucher, Jean, "Counseling Consumer Debtors to Make Their Own Informed Choices—A Question of Professional Responsibility," 5 ABI Law Review 165 (1997), and articles cited there. Return to article

⁵ As used here, "norm" means an ideal or aspired-to outcome. When quantified and put onto timetables, norms are expressed as goals. Return to article

⁶ The data here are based on all U.S. jurisdictions except for Alabama and North Carolina, which are not included in the U.S. Trustee Program. These are, however, very active chapter 13 jurisdictions. Return to article

⁷ Excluded from the $2.5 billion is $244 million to debtors attorneys through plans, $127 million back to debtors after their cases terminated, $2.8 million to the trustees as fees for unconfirmed filings and $1.7 million for miscellaneous noticing. Return to article

⁸ Readers who would like a full account of the reasoning behind this choice may contact us at [email protected] or [email protected]. Return to article

⁹ Of course, we recognize that several of the states shown in the tables and others comprise more than one judicial district. Differences among districts within states, divisions within districts and courtrooms within divisions are all sources of variations in chapter 13 practices, of which some are policy-relevant and will be the subject of subsequent research. Return to article