BANKRUPTCY BY THE NUMBERS
Bankruptcy Reform: Finding the Best Gross Income Test
Contributing Editor: Gordon Bermant
The major bankruptcy reform bills offered last year in the 105th Congress and this year in the 106th include a test of the gross incomes of chapter 7 filers to determine if they should be permitted to remain in chapter 7: a gross income means test. This test is combined with one or more other means tests (e.g., income after allowed expenses) to determine whether the use of chapter 7 by debtors with relatively high incomes is an abuse of the Code and therefore subject to motion for conversion or dismissal under 11 U.S.C.ï½§707(b). This article focuses on the rationale and demographic bases of the gross income test. It concludes that using county and Metropolitan Statistical Area median incomes may be an equitable and practical alternative to using national, regional, or state median income standards.
The details of the gross income means tests differ considerably among the various reform bills. This year's bills, for example, H.R. 833 and S. 625, place the means tests in different procedural and administrative contexts. Significantly, the bills also differ in the standards of gross income they select as the basis for determining who may not remain in chapter 7. At different times the bills use national median incomes, regional median incomes, and state median incomes as gross income standards. (2) The bills also take different approaches to supplementing the published income medians for larger families. (3) The rationale for using a variety of standards, or indeed for using any one of them, has not been explained.
A policy rationale for gross income means testing.
Establishing a gross income means test has an intuitive policy rationale with two parts. First, it seems fair to require high-income debtors in bankruptcy to reorganize their finances in order to repay at least some of their unsecured debt over time. Second, it seems both fair and practical to excuse or protect from this requirement debtors whose incomes are so low that they would be unable to repay a meaningful amount of money in any event.
Any means test uses some standard against which to compare the debtor's gross income. The standard should bear a meaningful rather than an arbitrary relationship to the debtor's income. In order to be meaningful, the standard should reflect the debtor's actual economic environment. In order to be practical, the standard should be based on readily available income data.
The standards in the current bills probably meet the criterion of practicality because national, regional, and state median incomes are already routinely collected.(4) These standards do not meet the criterion of meaningfulness, however, because the economic environments of debtors vary much more widely than can be reflected in accumulated national, regional, or state-wide figures. It would be preferable to select a readily available standard that is more likely to represent the debtor's actual economic environment. Data developed by the U.S. Department of Housing and Urban Development (HUD) for the nation's counties and MSA's appear to meet this requirement.
The HUD County-wide and MSA data.
HUD has developed a table of income limits for 3,188 counties and Metropolitan Statistical Areas (MSA) covering all 50 states, the District of Columbia, and Puerto Rico. HUD uses the information to determine eligibility for various assisted housing programs. These tables show promise as standards for gross income means testing in bankruptcy.
HUD starts with a median income figure for each county for a family of four. This figure is reduced by 30 percent for a family of one, by 20 percent for a family of two, and by 10 percent for a family of three. The median is increased by 8 percent for each member over four in a family. HUD performs further calculations for its own housing-assistance program purposes.
The median income figures range from a low of $11,600 (rural Puerto Rico) to a high of $94,300 (the Stanford-Norwalk Conn. MSA)(5). In most counties the median income for a family of four is between $30,000 and $50,000, with the median for all 3,188 counties-MSA at $38,600. The median incomes tend to be higher in the more populous counties. In all but four of the 106 counties with more than 500,000 people, the median income was above $38,600, and 75 percent were above $50,000.
Comparing the HUD County-MSA data to Other Data Sources.
The Bureau of the Census is the primary source of income data at national, regional, and state levels.(6) The following table compares Census data at these levels with County-MSA income figures from HUD. All numbers represent medians for four-person families except for the regions. State-level median incomes broken down by family size appear not to be published by the Census for all family sizes, but it is likely that these numbers could be made available if the new legislation required them.
For each level, the table shows the maximum and minimum incomes reported at each level of aggregation. Only data from the 50 states were included for purposes of the table.
MEDIAN GROSS INCOMES TAKEN AT DIFFERENT LEVELS OF AGGREGATION
(FAMILY OF 4)
(ALL FAMILY SIZES)
(FAMILY OF 4)
(FAMILY OF 4)
(STARR COUNTY, TX)
The Consequences of Using County-wide Income Data.
As the table clearly shows, reducing the size of the population on which the standard is based increases the range of the standard from top to bottom. Using the national standard sets the same income threshold for all debtors with the same family size, irrespective of local variation. Using regional standards provides a 13% higher threshold for debtors in the Northeast than for debtors in the South. It is clear, however, that the cost of living varies widely within each Census region (e.g., New York City vs. rural Maine). Using state-level data increases the range of standards against which debtors' gross incomes would be evaluated, but the range of incomes within states is substantial: for 32 states, the income in the most affluent county or statistical area is at least twice that of the least affluent county.
There are two major consequences of using county medians rather than an alternative based on a state or national medians. First, debtors whose incomes are below the county median but above the alternative would not be subject to a ï½§707(b) motion on the basis of gross income alone. Consider, for example, a couple with two children, filing jointly, who live in Santa Clara County, California. The couple has been steadily employed in large blue-collar service industries, and after many years, have arrived at an annual gross income $62,000. Because the national median income for a family of four is $53,350, if that median were the legislative gross income standard, this couple would be subject to dismissal or conversion for abuse under ï½§707(b); the details of their exposure vary between HR 833 and S 625. The median income for a family of four in Santa Clara County is $82,600. If county income were the gross income standard, the couple's exposure to conversion or dismissal would be reduced and perhaps eliminated, depending on other factors, by both the House and Senate bills.
Second, debtors whose incomes are above the county median but below the alternative would become subject to conversion or dismissal. Consider for example a second couple with 2 children living down the California coast a short distance, in Monterey County, where the median income for a family of 4 is $49,400. This couple is employed similarly to the couple described above, but in Monterey County their work is worth $51,000 annually. This couple becomes exposed to conversion or dismissal by the gross income test, even though their income falls below the national median for their family size. As before, differences between the pending bills would affect whether the motion would in fact be filed.
We have information about how a move to county-MSA based gross income standards would affect debtors who now file in chapter 7. In 1998 the Executive Office for United States Trustees completed a study of almost 2,000 chapter 7 debtors to determine, among other things, the income distributions of debtors who had no assets for distribution to creditors. (More than 95% of all consumer debtors fall into this category.)(9) It turns out that chapter 7 debtors are more likely to live in high income areas than the population in general, even though the median incomes of chapter 7 debtors, overall, are less than the national median. Many debtors with incomes greater than the national median live in counties with median incomes that are still higher than the debtors' incomes. That is, they are like the couple from Santa Clara County. But there are also many debtors whose incomes are beneath national medians but greater than the median for their counties. These debtors, like the couple from Monterey County, would be subject to suit under ï½§707(b), absent other considerations.
There is an immediate intuitive appeal to basing an income standard on information closely connected to the debtor's actual economic environment. Moreover, because the HUD data are updated annually to meet HUD's own needs, the information would track changes in incomes reasonably closely.
There could be an additional benefit to using county income levels , not apparent at first glance, that would arise if the gross income test were placed first among means tests to be applied to each debtor. Many debtors with incomes that are greater than national or state medians but less than their county median (like the hypothetical couple living in Santa Clara County) will have insufficient disposable income to meet the minimum chapter 13 funding requirements. The high housing costs of high-income counties, among other things, will be allowed under any test that relies on actual housing expenses or IRS guidelines, and these costs may drive the debtor's disposable income beneath the means testing threshold. The application of expense measurements is complex and time-consuming relative to gross income means testing. High median county incomes are probably decent proxies of high housing costs and other costs. Therefore, by putting a county-based gross income test at the head of the queue of tests, trustees and debtors' lawyers can more efficiently assess the proper course of action in planning for or evaluating the debtor's choice of chapter.
It is worth noting in conclusion that the standard established in S. 625 allows the higher of the national or the debtor's state median income to serve as the gross income standard. This recognizes variation among the states and gives the debtor the benefit when he or she lives in a state with a relatively high median income. Of course, this degree of choice could also be applied to a test which used county-MSA income in addition to state or national medians.
* This article originally appeared in the American Bankruptcy Institute Journal, Vol. XVIII, No. 6, July/August 1999. It is reprinted with permission of the American Bankruptcy Institute.
1. This research was supported by a contract with the Executive Office for United States Trustees. I thank Joe Guzinski and Ed Flynn for their encouragement and support, and Peter Zorn of Freddie Mac for first bringing the HUD data to my attention. All opinions expressed in this article are those of the author, and do not necessarily represent the views of the Executive Office for United States Trustees.
2. Three different gross income standards are proposed in section 102 of H.R. 833. Two of these are in amendments to 11 U.S.C.ï½§707(b) and one is in an amendment to 11 U.S.C.ï½§704. Section 102 of S.625 proposes a gross income standard as an amendment to ï½§704. Among these four amendments, national, regional, and state-wide medians are proposed as test standards, with different treatments for large families and uncertainty whether the large-family supplement of $583 for every family member more than four is to be applied per month or per year. The Senate's amendment to ï½§704 allows for the larger of the state or national median to be the standard for each debtor.
3. See the previous note. The rationale for these supplements is that median incomes tend to peak for families with four members, even though expenses continue to grow with increasing family size. If supplements are not added to the gross income standard to account for this demographic fact, large families would be treated unfairly by the gross income means test.
4. There might be a problem in respect to state or regional data separated by family size. See the discussion below.
5. Some counties contain more than one MSA, and some MSA overlap county borders. But the 3,188 HUD counties and MSA are non-overlapping.
6. Census also subdivides incomes into 9 divisions throughout the country, which are not included here.
7. The four Census Regions are NORTHEAST - (CT, MA, ME, NH, NJ, NY, PA, RI, & VT); SOUTH - (AL, AR, DC, DE, FL, GA, KY, LA, MD, MS, NC, OK, SC, TN, TX, VA, & WV); MIDWEST - (IA, IL, IN, KS, MI, MN, MO, ND, NE, OH, SD, & WI), WEST - (AK, AZ, CA, CO, HI, ID, MT, NM, NV, OR, UT, WA, WY).
8. This Primary MSA comprises Darien, Greenwich, New Caanan, Norwalk, and Stamford, CT.
9. Bermant, Gordon, and Flynn, Ed., Incomes, Debts, and Repayment Capacities of Recently Discharged Chapter 7 Debtors. Executive Office of United States Trustees, January, 1999. The study has been reviewed by the General Accounting Office in its report Personal Bankruptcy: Analysis of Four Reports on Chapter 7 Debtors' Ability to Pay. GAO/GGD-99-103 (June, 1999).