Charitable contributions in bankruptcy: an empirical analysis

Charitable contributions in bankruptcy: an empirical analysis

Charitable contributions in bankruptcy


While viewed by some as a variation upon the “Robin Hood” (1) theme, others consider the right to tithe, (2) or to give to charitable causes, even while in the midst of a bankruptcy proceeding, as a constitutionally protected expression of their religious convictions. Much has been written on the constitutionality of tithing and charitable giving in bankruptcy. (3) The issue can arise when the debtor seeks approval of a Chapter 13 debt-repayment plan that includes future tithing. (4) The issue also arises in a Chapter 7 proceeding when a trustee in bankruptcy orders a church to turn over monies previously donated by a tithing debtor prior to filing a bankruptcy petition. (5) In either context, the constitutional issues are similar.

On June 19, 1998, Congress addressed the issue of tithing in bankruptcy by passing legislation known as the Religious Liberty and Charitable Donation Protection Act (hereinafter referred to as the Charitable Donation Act). (6) This law specifically allows debtors to make certain charitable contributions during a Chapter 13 proceeding, and generally prohibits trustees from requiring churches to turn over income previously received by the religious or charitable entity from a debtor in bankruptcy. (7)

Assessing the wisdom of this public policy requires objective facts. To this end, the authors collected data from bankruptcy courts in Alabama, Arkansas, Georgia, Louisiana, and Tennessee from the year 1997. Our analysis focuses on the incidences of charitable contributions (8) in Chapter 13 bankruptcy proceedings in these jurisdictions. (9) In particular, we collected data on the actual and presumed “characteristics” of these selected debtors.

Our empirical analysis generates a number of insights. First, while bankrupt debtors are giving, their level of giving (as it relates to the percentage of the population, average amount given, or percentage of income given) is significantly less than the general population. (10) Second, there are indications that the income levels for bankrupt debtors may be decreasing when adjusted for inflation. (11) Third, charitable bankrupt debtors have more income than non-charitable bankrupt debtors by every income measurement except at the median, where they are equal. (12) Fourth, at no income measurement did any bankrupt debtor, charitable or non-charitable, have a positive net worth. (13) Fifth, the differences in the debt/income ratios of charitable and non-charitable bankrupt debtors are insignificant

This begins with a background discussion of the social and legal history of tithing generally and the new federal legislation specifically. Then the empirical data are presented. The article concludes with reflections. We note that Charitable Contribution Act may have given churches reason to celebrate, but the war is far from over. The issue of tithing in bankruptcy is too emotional and the financial players too big for the issue to go away. As the issue evolves, the more empirical grounding public policy makers have, the better. Our analysis is presented in that spirit.


In the Judeo-Christian system, the word “tithe” comes from a Hebrew word which translates to mean a “tenth part.” The history of tithing began very early in biblical times, preceding even the Mosaic Law. (17) Abraham, the patriarch to all Jews, tithed to Melchizedek, King of Salem and priest of the God Most High. (18) Just as Abraham tithed to Melchizedek as priest, so too the Mosaic Jews tithed to the priesthood. (19) The practice of tithing plays a prominent role in a variety of denominations and transcends Judeo-Christian faiths. In the Hebrew Scripture, tithes included money, first fruits of the harvest, and gifts in kind. (20) Tithing functions as a commitment to the service of others, a token of faith, and a symbol of the donor’s belief that worldly benefits result from divine providence. (21) This tithe represented the wages for the Levites (22) for their work in the Tabernacle. (23)

The tithe is considered by many Christians to be God’s plan to finance the church and consequently the church’s work for Him. Most Christians take this charge extremely seriously, (24) as does Scripture. The Bible makes the failure to pay tithes tantamount to robbing God. (25) This illustration regarding robbery serves as an example for those Jews and Christians who look upon God as the benevolent “Resource-Giver.” Tithing for a Jew or Christian may be deemed a necessary expense since God is the reason for any increase in material wealth and provisions.


A. In General

In proposing a plan for debt repayment in a Chapter 13 proceeding, Code requires debtors to provide monthly budget figures detailing both net income and expenses in order to calculate the amount of “disposable income” available to the debtor for funding the plan. (26) Disposable income is generally defined in the Code as that which is not reasonably necessary to maintain or support the debtor’s household (and business if the debtor is so engaged). (27)

Among the “Official Forms” utilized for presenting the financial information of the debtor is “Schedule J,” which is used for listing the itemized expenses of the debtor. One of the line-item expenses which appears on this official form, among the other monthly expenses that a debtor may have, is an expense for “charitable contributions,” (28) of which “tithing” can be considered a subset. The amount a debtor may choose to allocate from his monthly income for this particular expense can be variable, just as the amount a debtor may allocate for other expenses, such as “recreation,” may be somewhat variable in amount–unlike rent, for example, which is usually a set amount. It is here that past debates have arisen and upon which the subject of this paper focuses.

Where debtors have budgeted an amount for “charitable contributions” greater than an amount deemed acceptable to the Chapter 13 trustee, (29) objections to court confirmation of the plan have arisen. The question is thus posed–“What is reasonably necessary to maintain the debtor’s household?”

B. Case Law Interpretations

What is “reasonably necessary” for the maintenance or support of a debtor is a matter that has been left to developing case law, (30) there being no clear legislative direction given as to what types of expenses would be considered appropriate. (31) With regard to tithing specifically, divergent lines of cases have developed, (32) as well as much journalistic discussion. (33)

The majority of courts faced with this issue denied confirmation of Chapter 13 plans that provided for tithing as an expense of the debtor, primarily determining that such expenses are not reasonable expenses for confirmation of a Chapter 13 plan in spite of First Amendment free exercise arguments or other defenses to denial of confirmation raised by the debtor. (34) A minority of courts have confirmed Chapter 13 plans in spite of objections to plans which included tithing as a “necessary expense.” (35)

C. The Charitable Donation Act

In passing the Charitable Donation Act, (36) Congress definitively set forth specific guidelines designed to mitigate against the “separation of church and state” opposition to charitable contributions in bankruptcy proceedings. (37) This law was provoked primarily by legal activity wherein the trustees in Chapter 7 bankruptcy proceedings successfully (in most cases) sued churches who had been the recipients of tithes of persons who subsequently found themselves debtors in bankruptcy. (38) These cases were based upon the Code provision that generally allows a trustee to recover transfers of the debtor’s property made by the debtor within one year prior to the day the bankruptcy petition was filed under certain circumstances.(39) Relief was afforded to the trustees in the majority of these cases in spite of vociferous defenses presented on various grounds, including the free exercise clause of the First Amendment and the then–recently–enacted Religious Freedom Restoration Act (RFRA). (40) Specifically, the new law provides that monies that would be expended for charitable contributions are not to be included within the debtor’s disposable income “in an amount not to exceed fifteen percent of the gross income of the debtor for the year in which the contributions are made.” (41) In other words, debtors could give up to fifteen percent, but no more. This is the first occasion in which any specific item of expense has been set forth in the Code as one that must automatically be deemed “reasonably necessary” for the “maintenance or support” of the debtor.

Thus, for the first time, debtors are now able to freely list a “tithe” on their schedules in Chapter 13 proceedings. By passage of this Act Congress has removed from the courts the authority to adjudicate the reasonableness of a debtor’s expenditure for charitable contributions so long as that contribution falls within these guidelines and is made to a religious or charitable entity or organization as that term is defined in Section 548(d)(4) of the Code. (42)

It is significant to note that by this law there is no requirement that the debtor show a past history of tithing in order to tithe while in the Chapter 13 proceeding. Nor does this law require that the debtor extend either the plan period or the amount of payments in order to tithe in a Chapter 13 proceeding.

Further, the Charitable Donation Act allows a debtor to tithe up to a set amount of fifteen percent, which exceeds even ecclesiastically recognized minimum standards. (43) This percentage was likely set to take into consideration the debtor whose religious conscience dictates that he or she exceed minimum standards. Also, interestingly, the percentage is statutorily based upon gross rather than net income. (44)

In order to specifically address the concern which initially brought this statute into existence, i.e., the growing practice of trustees recovering from church organizations monies previously contributed to their churches within a year prior to filing bankruptcy, (45) the Charitable Donation Act exempts from the coverage of a portion of the “Fraudulent Transfer” provision religious or charitable contributions which do not exceed the fifteen percent guideline. (46) Indeed, the law goes a step further and exempts any contribution by the debtor during the relevant period, regardless of the amount, so long as that contribution was consistent with the practices of the debtor in making the contributions. (47)

This new law does not, however, limit the applicability of that portion of the “Fraudulent Transfer” provision which is designed to prevent actions by a debtor designed “to hinder, delay or defraud any entity.” (48) Thus, as stated by Senator Grassley: “Accordingly, a transfer of assets which looks like a tithe or a charitable donation, but which is actually fraud, can still be set aside.” (49) Further, debtors continue to be held to standards of “good faith” in proposing Chapter 13 plans. (50)

It would seem that the Charitable Donation Act, with its clear language and standard, would bring this matter to a swift and final conclusion. However, such obituaries would be premature, as exemplified by the Cavanaugh (51) and Buxton (52) cases. The Chapter 13 trustees in these cases asked the courts to declare that charitable contributions are not a reasonably necessary expense and, therefore, are excluded from expenses to arrive at disposable income under Section 1325(b)(2)(A). (53) The trustee in Cavanaugh was unsuccessful (54) and the trustee in Buxton was successful, (55) so the controversy continues. Further, there are those who believe that while transfers to charities or religious entities may be immune to attacks on the basis of constructive fraud (56) they are not immune to attacks on the basis of factual fraud. (57)


A. Objectives of the Study

The empirical portion of this article includes the analysis of both household (microeconomic) and national (macroeconomic) decision variables. (58) To provide data for the microeconomic portion of the study the authors conducted an analysis of debtors in Chapter 13 bankruptcy proceedings during the calendar year 1997. This was accomplished by sampling bankruptcy cases in eight bankruptcy jurisdictions coveting five Southern states. (59) Specifically, the microeconomic portion of the study was designed to provide a statistical analysis of several factors, including: (1) the number of debtors who have actually included charitable contributions (60) on their schedules as a “necessary expense,” (2) the types and amounts of expenses budgeted by charitable debtors as compared to non-charitable debtors, (3) the percentage of the total debt proposed to be paid by the charitable debtor as compared to the non-charitable debtor, and (4) the types of debts scheduled by charitable debtors as compared to non-charitable debtors. (61) In addition to these factors, other sub-factors emerge for comparison and all of the data is presented in Tables 1 through 9 collected in the Appendix.

The macroeconomic portion of the study employs econometric techniques to identify both the time series characteristics of the U.S. personal bankruptcy filing rate and the macroeconomic and financial variables that aid in the explanation of the level of bankruptcies. The variables identified as significant relate directly to both charitable and non-charitable debtors who have filed for bankruptcy.

The data indicated several significant findings: (a) bankrupts do give, but at rates and in amounts significantly less than the general population

B. How the Study was Conducted

As for the microeconomic portion of the analysis, after the jurisdictions were selected and those jurisdictions indicated a willingness to participate, the clerks’ offices were contacted for the range of docket numbers used in that district during 1997. These ranges of numbers for each individual jurisdiction were then given to the statistician who was requested to randomly select about 150 to 300 cases (69) for each jurisdiction. (70) Since cases are filed sequentially without regard to type of bankruptcy proceeding (i.e., Chapter 7, Chapter 13 or other type of case), Chapter 13 cases were manually culled from this listing, as case files were removed from the shelves leaving a variable number of cases to study within each jurisdiction.

As for what leads individuals to file for personal bankruptcy, it is apparent after conducting this study that macroeconomic consumer bankruptcy analysis suffers from a lack of a large amount of readily attainable data. Thus, most studies, as does this one, rely on annual observations for their analysis. (71) Annual personal bankruptcy filings obtained from the Administrative Office of the U.S. Courts, from 1960 to 1998, and the United States Census Bureau Statistical Abstract of the United States: 1997, are utilized for the model development section of the study.

C. How the Data was Collated

Once the cases were selected, bankruptcy documents were copied from each case including the Petition Page (which contains the basic filing information), the Summary of Schedules, Schedule D (List of Secured Creditors), Schedule E (List of Priority Unsecured Creditors), Schedule F (List of Non-Priority Unsecured Creditors), Schedule I (Current Income Information), Schedule J (Current Expenditures Information) and the Chapter 13 Plan.

A file number was then assigned to each of the cases by the authors separately from the case numbers assigned by the courts. The authors then went through each case and manually categorized each debt listed on the schedules of the debtors according to fifteen different categories. (72)

It is conceded that there may be a certain error rate associated with categorizing these debts due to the imperfection of this procedure and the fact that there is some overlap between types of debts. For example, credit card debt may obviously include clothing, entertainment, or other category of purchase. However, without interviewing each individual debtor or their counsel to get to the true nature of each debt, this was the best available mechanism. What the data does help to show is how the debt was incurred (i.e., credit card, personal loan, etc.) as much as it reveals to some extent what the debt was incurred for. Nevertheless, the process is validated to a large degree due to the experience of the authors as former bankruptcy practitioners. Further, the vast majority of the cases were prepared by bankruptcy counsel who were careful to precisely categorize the nature of each debt on the schedules. Data from the various schedules were then inputted into the spreadsheet program and the resulting variables analyzed for statistical differences. Each category was subdivided into charitable and non-charitable debtors. The mean for each variable corresponding to each type of debtor was then calculated and tested for significant statistical differences using a standard two-sample t-test assuming unequal variances (Tables 1 and 2). (73) Because means averages are many times distorted, the authors have provided twenty-five percentile, median and seventy-five percentile data so that certain deviations may be detected, such as whether there are high-and/or low-end aberrations which may induce conclusions that the mean average could not detect.

The macroeconomic section is a follow-up to the test for differences among charitable and non-charitable debtors. This analysis is performed so as to identify any key variables that may lead to differences in filing rates and corresponding differences in financial characteristics among debtors. Numerous explanatory variables are used by researchers to determine the importance of each in explaining the number of personal bankruptcy filings. The majority of these variables fall into one of three categories: key macroeconomic indicators, personal debt measurements, and demographic statistics. However, for the purposes of this study, only key macroeconomic indicators and personal debt measurements are used due mainly to the fact that preliminary testing reveals no statistically significant relationship between demographic statistics and the number of personal bankruptcies.

1. Macroeconomic Indicators

Basic economic intuition asserts that the normal ups and downs of the economy are important in determining the current and future financial well-being of household finances and their resulting ability to follow a payback schedule. Bishop notes that during recessions the likelihood of severe financial distress increases, leading to a greater probability of any particular household filing for bankruptcy. (74) Conversely, an economic expansion results in more fully employed resources, including labor, and brings with it the potential for more rapid income growth and a decreased likelihood of unmanageable financial distress.

2. Personal Debt Indicators

The most casual of observers in the legal or financial industry notes an increase in personal debt levels. Kennickell, Starr-McClure and Sunden detail findings from the 1995 Survey of Consumer Finances (SCF). (75) They note that the proportion of families with debt rose slightly between 1989 and 1992 and then more substantially by 1995. The amount of mortgage debt outstanding increased approximately thirty percent over the same six-year period. Additionally, the percent of families with outstanding credit card balances rose from forty percent to forty-eight percent reflecting the effective marketing campaigns of credit card issuers. Thus, it is posited that the level of consumer debt has a direct influence on the number of personal bankruptcies.

3. Description of Variables

To arrive at an accurate model it is necessary to select a set of’ explanatory variables that capture the major attributes of each of the factors cited above. While this process is admittedly arbitrary to some extent, the researcher must consider factors such as data availability and the characteristics of the data series itself. To complete the model specification the following variables were selected with the expected effect on the dependent variable (total bankruptcy filings) in parentheses:

GDPA = real gross domestic product in fixed 1992 dollars (-) (76)

EMPL_LVL = employment level civilian labor force age 20 and over (-) (77)

SAV = personal savings per person 20 years of age and over (-) (78)

TDSP – debt service payments as a % of disposable personal income (+) (79)

CIC_PI = ratio of consumer installment credit to personal income (+) (80)

The data used in the estimation of the model are of annual frequency over the period 1960 through 1998. It is important to note that although the data set is admittedly small by time series standards, this is all of the data available at this time to the best of the authors’ knowledge.

Ordinary least squares regression is used to determine explanatory beta coefficients. However, stationarity of the data used in estimation, as with many economic time series, is a concern. Thus, the Augmented Dickey-Fuller unit root test (81) is performed on the level, first, and, if necessary, second difference of each of the variables involved. Each variable is transformed using log differences to account for both nonstationarity and heteroskedasticity of the errors when only differenced variables are used (Table 1).

D. Empirical Results

1. Charitable Giving

In 1995, Americans in the general population who donated to religious or charitable organizations gave an average of $1,017 per household, and this represented sixty-nine percent of all U.S. households. (82) Further, Americans as a whole gave $116 billion or about two percent of their personal income? In our study the charitable donors gave a mean average of $606 per household, (84) and this represented sixteen percent of all households surveyed (150 of 957). As a whole the debtors in our study gave approximately $90,900 or about 0.46 percent of their personal income. (85) After adjusting for inflation the average contribution per household in the survey is 56.4 percent of the general population. In addition, the general population gives at a rate over four times our survey population in terms of percentage of households giving (sixty-nine percent to sixteen percent) and over four times our survey population in terms of percent of personal income (two percent to 0.46 percent).

In order to place these numbers in perspective, the median household in the general population would give about $739 ($36,953 (86) x 0.02), whereas the median charitable giver in our survey would give about $77 ($16,666 x 0.0046), or almost one-tenth the donation of the median household in the general population. So clearly while certain bankrupt debtors are giving, our data indicates that they are giving, at each of our measurement characteristics, at a much lower level relative to the general population.

2. Income

All debtors in this study are, for the most part, in deep financial trouble. The U.S. Census Bureau shows that the 20th percentile income for 1997 was $20,586, the 40th percentile for 1997 (87) was $36,000, and the median income for 1997 was $36,953 (88) for the general population of the entire United States. A cursory review of Table 3 shows that at the 25th percentile the income was $12,060 for all debtors and, further, that at the 75th percentile the income was $23,828 for all debtors in our study. The mean average household income in our study is $20,380 while the median average is $16,482. (89) This indicates that there are some higher income debtors who are influencing the mean average higher but the $23,828 figure at the 75th percentile indicates that such upward influence is not severe. (90) It is important to note that most bankrupt debtors are at the lower end of the income spectrum. This verifies the findings of studies which show that the average bankrupt debtor is in extremely poor financial condition regarding income. (91)

A significant development may be that bankrupt debtors in this study may have less absolute income than the debtors in the Sullivan study (the AWFOD study). (92) The AWFOD study determined that the median family income for bankrupt debtors was $15,000 in 1981 dollars. (93) Converted to 1997 dollars this 1981 median family income is $28,481. (94) This is almost $12,000 more than the median income ($16,690) in the current study and, further, it is over $4,500 more than the 75th percentile income ($23,935) in the current study. (95) There can be no absolute correlation of these years and incomes because different states were used in each study. (96) However, this factor is worthy of further study in that it may give greater insight into the current bankruptcy explosion. It is possible that there is a bankruptcy explosion because the real income of debtors has been constantly eroding, (97) or it may be that the cost of living variations between geographical locations explain the differences. At any rate, further study is warranted.

When we turn our attention to income comparisons between the charitable and non-charitable debtors in our study we find them to be equal at the median income level ($16,666), which seems to be quite remarkable. (98) However, the mean income of charitable debtors is $2,245 more than the non-charitable debtors ($22,592 to $20,347). (99) This demonstrates that both charitable and non-charitable debtors have some relatively higher income debtors above the median who are elevating the mean average

Simply looking at these income figures may lead to a conclusion that the charitable debtors are giving out of relative plenty in that the difference between the mean incomes is $2,245 or 9.9% of the mean charitable debtor’s income. However, before making such a conclusion, let us look at the at the asset and debt situations of both groups.

A point to note is that charitable debtors in this study have more income than the non-charitable debtors except at the median level where they are equal. (101) We believe this income difference may be significant in allowing charitable debtors to have additional free income with which to pay charitable donations. This will be especially relevant when we discuss child support and tax debts. (102) The difference between the charitable and non-charitable mean incomes is 11.6 percent. (103) Because the charitable debtors pay only 2.7% of income as charitable expenses, (104) their residue income after tithing is higher than the income of the non-charitable debtors

1. Assets and Debts

a. Net Worth. A cursory review of Table 3 of this study reinforces the idea that bankrupt debtors are in deep financial trouble. At every point of measurement, mean, 25th percentile, median, or 75th percentile, whether it is charitable or non-charitable, bankrupt debtors face a dire financial predicament. (105) They are loaded with debt relative to assets and income. This is the same bleak picture painted by the AWFOD study. (106) There is no category with a positive net worth (total assets less total debts). The true bleakness of all bankrupts’ financial position is readily seen when we contrast the net worth of the average family (107) with the bankrupt. The average family in the general population had a median net worth of $71,600 and a mean net worth of $282,500 in 1998. (108) In stark contrast, the average charitable bankrupt debtor in our study had a median net worth of negative $6,786 and a mean net worth of negative $5,320, and the average non-charitable bankrupt debtor in our study had a median net worth of negative $6,264 and a mean net worth of negative $9,807. (109) Further, we note that in the general population, even with families with incomes less than $10,000 per year, the median net worth was $3,600 and the mean net worth was $40,000. (110) There appears to be no distinguishing the financial positions of charitable and non-charitable debtors, except that charitable debtors have more of everything–income, assets and debt–to arrive at the same disastrous financial position. But, again, the fact that charitable debtors have greater income may be important in determining why they are able to devise a plan with payments to charitable interests. (111)

b. Debt/Income Ratio. The debt/income ratio which compares the total debt of the bankrupt to the income of the bankrupt can be a helpful indicator of the debtor’s current and future ability to repay debt and in our credit-intensive society, is the most critical measure of financial condition. (112) Debt/income ratio compares the level of debt to yearly incomes and shows the multiple of one’s yearly income that it would take to make a lump sum repayment of the type of debt one is analyzing. Simple income measurement is not very informative since mismanagement of resources occurs at all income levels. As the saying goes, “the person who is not satisfied with a little will not be satisfied with much.” (113)

The AWFOD study uses the debt/income ratio to compare the bankrupt debtor’s ability to repay with the average debtor in the general population. (114) Because income is the denominator, a lower ratio indicates a healthier financial condition. Therefore, if the ratio is 0.5 then it would take one-half of the debtor’s yearly income to pay off the debt and if the ratio is 1.5 then it would take 1 1/2 of the debtor’s yearly income to pay off the debt. At some point the ratio reaches a critical mass, where the debtor cannot service the debt and meet his or her current living expenses simultaneously, which brings about delinquent payments, repossessions, debt collectors, lawsuits, and finally bankruptcy. This is not a commentary on how or why the debtor reaches critical mass, for it can be involuntary (as with a major unexpected illness) or voluntary (as with lavish, unnecessary entertainment expenses). We are simply saying that once the critical mass is reached the problems begin to snowball out of control.

The debt/income ratio can be computed with (115) and without (116) the long-term mortgage debt included. Mortgage debt is generally long-term and thus distorts the short-term cash needs of the debtor. By eliminating the mortgage debt, the debt/income ratio is necessarily lower, but it gives a more realistic picture of the debtor’s short-term cash needs, (117) which is more relevant in the bankruptcy setting.

The AWFOD study found through the 1983 government survey (118) that only five percent of the general population had debt/income ratio (mortgage excluded) greater than 0.2

The 1998 Survey of Consumer Finances uses the threshold debt/income ratio of 0.4, which is almost five months of income, and found that 12.7% of the general population had a debt/income ratio exceeding 0.4. (119) Table 6A charts the same 0.4 threshold for our study and finds that 84.67% of the charitable debtors have a debt/income ratio exceeding 0.4 and 84.51% of the non-charitable debtors exceed 0.4. Since the percentage of the population with a debt/income ratio exceeding 0.4 in the 1998 survey is more than two and one-half times the percentage of the population with a debt/income ratio exceeding 0.2 in the 1983 survey, this indicates that the general population is carrying substantially more debt. This heavy debt load may be a foreboding of increased bankruptcy filings.

There is no significant difference between charitable and non-charitable debtors in the measurement of debt/income ratios. However, we do note a great divergence between the general population and the bankrupt population as a whole. While only 12.7% of the general population has a consumer debt burden greater than 0.4, 84.54% of the bankruptcy population had a consumer debt burden greater than 0.4. (120) To put this matter in proper perspective, a person making $20,000 per year with a consumer debt burden like 87.3% of the general population would have consumer debts less than $8,000. In contrast, a person making $20,000 per year with a consumer debt burden like 84.54% (121) of the bankrupt population would have consumer debts greater than $8,000.

As in 1981, the debt/income ratio illuminates the greatest difference between the general population and the bankrupt population bankrupt debtors carry a stifling amount of debt relative to their income. At some point and for varying reasons bankrupt debtors have reached the critical mass where income cannot carry the debt or the bankrupt debtor still has the financial wherewithal to weather the storm but he or she decides that the emotional and other personal costs are not worth the battle. Again, “[i]ncome compared with debt is particularly important in the context of financial collapse.” (122) However, when we compare the debt/income ratios of charitable and non-charitable debtors we find no significant differences. (123) This virtual equality of debt/income ratio means that while charitable debtors have more income, they seem to have used the extra income to acquire more debt.

4. Credit Card

Credit card debt (124) and defaults have been targeted recently as a major reason that debtors are filing bankruptcy

In our study, 544 of the 957 debtors (56.8%) had credit card balances. (131) When comparing charitable debtors to noncharitable debtors, 103 of the 150 charitable debtors (68.7%) had credit card balances and 441 of 807 noncharitable debtors (54.6%) had credit card balances. (132) Stephen Brobeck, Executive Director of The Consumer Federation of America, reports that fifty-sixty percent of all households in America carry bank card balances. (133) While we can form no direct correlation here, we do see that since bank cards represented 68.9% of all credit card debt, (134) the charitable debtors’ percentage seem to be in line with the national norm while the noncharitable debtors’ percentage seems to be low.

The AWFOD study, which used a 1981 bankruptcy population, (135) had eighty-eight percent of the wage earner population with outstanding credit card balances. (136) We do find this remarkable in that while it is commonly cited that credit card companies have become more aggressive in card distribution, the data of this study shows that only 56.8% of the debtors had outstanding credit card balances at the time they filed for bankruptcy in 1997, (137) while the 1981 study shows eighty-eight percent. If this data holds true in further studies, it could mitigate charges against credit card companies as to their impact on the rising bankruptcy filings. Further, we find that the 1998 Survey of Consumer Finances indicates that only 44.1% of families had a credit card with an outstanding balance. (138) This is actually a decrease from 1995 when 47.1% of families had a credit card with an outstanding balance. (139)

When the 1981 data is adjusted for inflation we find that although the 1981 mean average for credit card debt is $726 less than 1997 ($7,829-$7,103) and the 1981 75th percentile is $750 less than 1997 ($8,683-$7,933), the 1981 25th percentile exceeds the 1997 25th percentile by $758 ($1,751-$993) and the 1981 median exceeds the 1997 median by $725 ($3,982-$3,257). (140) This anomaly may be accounted for in that the region from which our study was drawn may be the region with the lowest average number of credit cards per family. (141) However, even when the current study’s 56.8% of debtors holding credit card debt is adjusted for the highest region in the AWFOD study (3.0-4.1) (142) the current study would still have only seventy-eight percent of its debtors holding credit card debt, a full ten percent less. It does not appear that increased credit card debt is a significant factor in increased bankruptcies.

5. Child Support and Taxes

We took a close look at the debt categories of child support and taxes for two reasons. First, our data indicated significant differences in average amounts paid for child support and taxes by charitable and non-charitable debtors. (143) Second, both of these types of debts are nondischargeable. (144) The Code, 11 U.S.C. Section 1328(a), provides for the discharge of all debts provided for in the plan or that had been disallowed under 11 U.S.C. Section 502. However, child support is a specific exception to discharge. (145) Furthermore, 11 U.S.C. [section] 523(a)(1) makes certain tax claims nondischargeable.

Nondischargeability of a debt is an extremely important bankruptcy concept because nondischargeable debt must either be provided for within the plan or paid outside of the plan. However, child support and tax claims take on even greater import under Chapter 13 as priority claims which must be provided for in the plan. (146) If the debtor fails to provide for the child support or tax claims in the plan, then the child support or tax creditors may object to confirmation of the plan. (147) This means that the significant differences in the child support and tax debts create a real effect on the cash requirements to fund a Chapter 13 plan.

Four percent (6 of 150) of charitable debtors listed child support debts and 26.0% (39 of 150) of charitable debtors listed tax debt. (148) In addition, 3.6% (29 of 807) of non-charitable debtors listed child support debts and 20.9% (169 of 807) (149) listed tax debt. Therefore, a greater percentage of charitable debtors than non-charitable debtors listed nondischargeable child support and tax debt. However, when we turn our attention to the average amount of child support and tax debt listed we see an entirely different picture.

At virtually all types and levels of measurement non-charitable child support debtors have significantly more of nondischargeable child support debt than the charitable child support debtors. (150) Table 7 shows that the mean average of non-charitable child support debtor is more than three times as much debt as the average charitable child support debtor ($9,629 to $2,832). Further, the 25th percentile is almost four times as much ($3,000 to $775), the median is almost two times as much ($5,750 to $3,100), and the 75th percentile is more than three times as much ($14,000 to $4,306). (151)

In regard to tax debtors, Table 7 shows that the tax debt for the mean average of non-charitable tax debtor is more than 2 1/2 greater than the charitable tax debtor ($8,816 to $3,371). The differences are not remarkable at the 25th percentile ($500 to $472) or the median ($1,261 to $1,060)

While the differences of the averages between charitable and non-charitable debtors in the tax debts is less than the child support, we found that the tax debts had a greater impact on cash needs because a significantly greater number of charitable debtors (39 and 6) and non-charitable debtors (169 and 29) had tax debts than had child support debts. (152) The mean average for charitable contributions in this study is 2.7% of annual household income or $606 per year. (153) We found that the effect of the difference in the child support and tax debt categories amounted to 2.3% of annual household income, or $468 per year. (154) Since the mean charitable contribution in our study is $606, (155) this $468 would account for 77.2% of the annual charitable contribution. It would be interesting to study whether charitable debtors simply know the law better and make conscious pre-bankruptcy choices, whether there is something in the basic attitudes of charitable debtors toward child support and taxes that causes them to prioritize payment of these debts, or whether there is some other factor which this study could not detect.

6. Macweconomic Data Analysis

With respect to the macroeconomic data analysis, the following model is estimated using least squares regression:

TOTBKRPT = C + [B.sub.1]GDP[A.sub.t-1] + [B.sub.2]TDS[P.sub.t-2] + [B.sub.3]CIC_P[I.sub.t-1] + [B.sub.4]EMPL_LVL + [B.sub.5]SA[V.sub.t-3 ]+ [e.sub.1] [where t represents a time index and e is the error term].

The estimation results suggest that each of the explanatory variables contribute significant explanatory power in the analysis of bankruptcy filings. An examination of the regression output in Table 2 shows an adjusted R-squared of .75, (156) which is quite respectable considering the results of previous studies. All coefficients have expected signs and are found to be significant at the. 10 level or less. The overall model displays a significant F-statistic of 22.25. (157) The variables utilized would appear to have the same effect on both charitable and non-charitable debtors. Thus, the aforementioned analysis of individual debtor attributes will hold regardless of economic conditions.

Although the bankruptcy model estimated could be used for bankruptcy forecasting, its main purpose here is to highlight the wide range of economic variables which contribute to personal bankruptcy filings and the lag-time needed for some of these variables to catch up to individual debtors in the United States.


While the Charitable Contribution Act may have given churches and secular charities a reason for celebration, this war is far from over. This issue is too emotional and the financial players too large for this issue to simply go away. However, this empirical study allows us to understand objectively who charitable and noncharitable debtors are and to begin to substitute empirical research for anecdotal evidence. This objective approach is of utmost importance when dealing with a subject for which people hold deep-seated yet dramatically opposed views.

This study reaffirmed the fact that debtors, regardless of income level, are in dire financial condition. They carry oppressive debt relative to their income and they overwhelmingly possess a negative net worth. However, despite this dim financial outlook, our study finds that debtors do give. The percentage of debtors giving and the amount and percentage of income given is significantly less than the general public, but it is remarkable that they give at all. A finding that alarms the authors is that income adjusted for inflation appears to be eroding. This issue needs further study, but it may prove to be a major reason for the bankruptcy explosion. The erosion of income as a major factor for the bankruptcy explosion is given support when it is noted that the debt/income ratios for the general population have increased substantially from the 1981 AWFOD study. Charitable debtors do appear to have more income than noncharitable debtors

An important finding is the significant difference between charitable and noncharitable debtors in the debt categories of child support and taxes and may provide at least a partial explanation for the additional cash charitable debtors have to make charitable contributions. Because noncharitable debtors carry substantially more nondischargeable child support and tax debt, they must provide additional cash to fund these nondischargeable debts in their plans. Our above-stated findings give objective information to a sometimes heated and emotional debate. Now we can begin to approach the issue of charitable giving in bankruptcy from a more studied perspective. The authors do not believe objective information will end the debate, but it will cause pure philosophical arguments to come face-to-face with facts on which such arguments will either rise or fall.


Tests for Unit Roots in Consumer Bankruptcy Variables

Using Augmented Dickey-Fuller Tests: 1960-1998

Variable Number of t Critical Value


Differenced t .01 .05 .10

TOTBKRPT Level -.0685 -4.0042 -3.5348 -3.1988

TOTBKRPT 1 -7.7629 *** -4.2324 -3.5386 -3.2009

GDPA Level -.2755 -4.2242 -3.5348 -3.1988

GDPA 1 -3.0984 -4.2324 -3.5386 -3.2009

GDPA 2 -5.3599 *** -4.2412 -3.5426 -3.2032

TSDP Level -3.4893 * -4.2242 -3.5348 -3.1988

TSDP 1 -3.4909 * -4.2324 -3.5386 -3.2009

CIC_PI Level -3.3626 * -4.2242 -3.5348 -3.1988

CIC_PI 1 -4.5509 *** -4.2324 -3.5386 -3.2009

EMPL_LVL Level -3.0831 -4.2242 -3.5348 -3.1988

EMPL_LVL 1 -4.4934 *** -4.2324 -3.5386 -3.2009

SAV Level 0.9808 -4.2242 -3.5348 -3.1988

SAV 1 -.5818 ** -4.2324 -3.5386 -3.2009


(1) Critical values are those computed/simulated by J. MacKinnon


(2) The appropriate number of lagged differences is determined by the

Schwarz information criterion (SIC).

(3) *** = significant at .01 ** = significant at .05 * = significant

at . 10


LS // Dependent Variable is DLOG(TOTBKRPT, 1)


Variable Coefficient Std. Error t-Statistic Prob.

C 0.145532 0.021645 6.723615 0.0000

DLOG(GDPA(-1),0) -2.805526 0.749959 -3.740905 0.0008

DLOG(TDSP(-2),1) 2.348598 0.416782 5.635072 0.0000

DLOG(CIC_PI(-1),1) 1.306479 0.282102 4.631223 0.0001

DLOG(EMPL_LVL,1) -2.765353 1.268207 -2.180522 0.0375

DLOG(SAV(-3),1) -0.145739 0.075651 -1.926466 0.0639

R-squared 0.793238 Mean dependent var 0.065264

Adjusted R-squared 0.757589 S.D. dependent var 0.118622

S.E. of regression 0.058404 Akaike info criterion -5.525950

Sum squared resid 0.098918 Schwarz criterion -5.259319

Log likelihood 53.04128 F-statistic 22.25158

Durbin-Watson stat 1.815558 Prob(F-statistic) 0.000000


Household Total Total

Income Assets Debt




Mean $20,698.94 $32,867.00 $41,979.98

Std. Dev. 35,685 37,772 44,608

25th percentile 12,222 5,793 13,252

Median 16,690 16,511 26,679

75th percentile 23,935 50,519 59,430

(N) 957 970 970


Mean $22,592.12 $40,775.37 *** $46,095.31

Std. Dev. 12,227 44,398 41,988

25th percentile 14,345 7,876 15,731

Median 16,666 30,151 38,389

75th percentile 26,666 54,809 64,898

(N) 150 150 150


Mean $20,347.05 $31,420.34 *** $41,227.18

Std. Dev. 38,496 36,274 45,055

25th percentile 12,056 5,537 12,865

Median 16,666 15,601 24,898

75th percentile 22,948 48,628 57,997

(N) 807 820 820

Secured Unsecured

Debt Debt




Mean $28,553.98 $13,426.00

Std. Dev. 31,677 24,001

25th percentile 5,860 3,058

Median 15,089 7,255

75th percentile 44,084 14,414

(N) 970 970


Mean $32,274.24 $13,821.07

Std. Dev. 35,507 13,694

25th percentile 7,019 4,260

Median 19,400 8,956

75th percentile 48,262 18,551

(N) 150 150


Mean $27,873.45 $13,353.73

Std. Dev. 30,901 25,444

25th percentile 5,579 2,749

Median 14,518 6,777

75th percentile 43,055 13,709

(N) 820 820

*** Significant difference at .01 level between charitable

and noncharitable debtors


Yearly Contributions by

Charitable Debtors

($ amount)

Mean $606

Std. Dev. 639

25th percentile 240

Median 360

75th percentile 600

(N) 150


Total Debt/ Total Debt/

Income Income

(mtg. included) (mtg. excluded)



Mean 2.06 1.19

Std. Dev. 1.61 1.24

25th percentile 0.90 0.60

Median 1.62 0.98

75th percentile 2.84 1.51

(N) 957 957


Mean 2.03 1.15

Std. Dev. 1.38 0.85

25th percentile 0.97 0.58

Median 1.88 0.94

75th percentile 2.70 1.55

(N) 150 150


Mean 2.07 1.20

Std. Dev. 1.65 1.30

25th percentile 0.89 0.60

Median 1.56 0.99

75th percentile 2.89 1.50

(N) 807 807


% of Debtors % of Debtors

with with

Debt/Income > Debt/Income >

20% (mtg. 20% (mtg.

included) excluded)


All Debtors 97.73% 92.37%

Charitable Debtors 98.00% 94.67%

Non-Charitable Debtors 97.68% 91.95%


Debt/Income > Debt/Income >

40% (mtg. 40% (mtg.

included) excluded)


All Debtors 93.30% 84.54%

Charitable Debtors 93.33% 84.67%

Non-Charitable Debtors 93.29% 84.51%


Credit Credit Card Credit Card

Card Debt/Income Debt/

Debt Unsecured





Mean $7,828.50 0.38 0.47

Std. Dev. 13,067.67 0.53 0.35

25th percentile 992.59 0.06 0.15

Median 3,257.30 0.18 0.44

75th percentile 8,683.00 0.50 0.79

(N) 544 527 540


Mean $8,475.56 0.41 0.48

Std. Dev. 11,047.39 0.58 0.35

25th percentile 984.00 0.06 0.15

Median 2,770.86 0.17 0.49

75th percentile 12,705.02 0.48 0.80

(N) 103 100 103


Mean $7,677.38 0.38 0.47

Std. Dev. 13,502.79 0.52 0.34

25th percentile 1,000.00 0.06 0.15

Median 3,275.00 0.18 0.44

75th percentile 8,212.37 0.51 0.79

(N) 441 427 437

Child Taxes





Mean $8,463.66 $7,795.19

Std. Dev. 9,628.46 28,076.03

25th percentile 2,190.94 500.00

Median 4,408.11 1,150.00

75th percentile 9,880.67 3,242.86

(N) 35 208


Mean $2,832.31 *** $3,371.16 **

Std. Dev. 2,376.75 9,338.85

25th percentile 775.00 472.03

Median 3,100.50 1,059.60

75th percentile 4,306.33 2,674.00

(N) 6 39


Mean $9,628.77 *** $8,816.12 **

Std. Dev. 10,166.75 30,756.04

25th percentile 3,000.00 500.00

Median 5,750.00 1,260.62

75th percentile 14,000.00 3,700.00

(N) 29 169

*** Significant difference at .01 level

** Significant difference at .05 level


1981 Credit Adjusted to

Card Debt (a) 1997 (b)

Mean 3,741 7,103

S.d. 7,606 14,442

25th percentile 922 1,751

Median 2,097 3,982

75th percentile 4,178 7,933

(N) 1,053 1,053

(a) See AWFOD, supra note 92, at 183.

(b) The Inflation Calculator,


Net Worth


Mean -$5,319.94 *

Median -6,785.50

Count 150.00

Std Dev 26,436.03


Mean -$9,806.84 *

Median -6,264.00

Count 820

Std Dev 34,215.24

* Significant difference at .10 level for means

(1)”Robin Hood” is a beneficent fictional character who, though his deeds were villainous, his causes were of a eleemosynary spirit. There is some parallel between tithing in bankruptcy and the “take from the rich and give to the poor” theme embodied by Robin Hood.

(2) For purposes of this paper, the tithe consists often percent or more of a person’s income which, for the most part, is to be given over to the religious order with which the tither is affiliated.

(3) See Alane A. Becket, Should Charity Begin with Bankruptcy?, 18 AM. BANKR. INST. L. REV. 12 (1999)

(4) Chapter 13 refers to a type of bankruptcy proceeding set forth in 11 U.S.C. Sections 1301, et. seq., and is one in which the debtor initiates a statutorily prescribed plan for paying his/her creditors a percentage of the debt owed over a three to five year period of time.

(5) See, e.g., In re Bloch, 207 B.R. 944 (D. Colo. 1997)

(6) Pub. L. No. 105-183, 112 Stat. 517 (codified as amendments to 11 U.S.C. [subsection] 544, 546, 548, 707 and 1325). The co-sponsors of this bill were Representative Ron Packard and Senator Charles Grassley, further supported by the Coalition for the Free Exercise of Religion, an alliance of some sixty religious and civil fights groups led by Steven T. McFarland of the Christian Legal Society’s Center for Law and Religious Freedom.

(7) See infra Part III (discussing the Charitable Donation Act).

(8) This study also separated out incidences of tithing as distinct from charitable giving generally

(9) Bankruptcy districts in the states of Alabama, Arkansas, Georgia, Louisiana and Tennessee were included in this study under the presumption that a greater number of tithers might be found within the so-called “Bible Belt” of the United States.

(10) See infra Part IV.D.1.

(11) See infra Part IV.D.2.

(12) Id.

(13) See infra Part IV.D.3.a.

(14) See infra Part IV.D.3.b.

(15) See infra Part IV.D.4.

(16) See infra Part IV.D.5.

(17) The Mosaic Law is so named to refer to the laws given by God to Moses on Mount Sinai according to Exodus 20:31 of The Holy Bible). These were general as well as specific rules of conduct.

(18) See Genesis 14:18-20.

(19) See Numbers 18:21-32.

(20) Bruce W. Megard, Jr. Tithing and Fraudulent Transfers in Bankruptcy: Confirming a Trustee’s Power to Avoid the Tithe After City of Boerne v. Flores, 71 AM. BANKR. L.J. 413, 413 (199 7).

(21) Id.

(22) According to The Holy Bible, the Levites were members of the Israelite tribe of Levi designated by God as the priestly tribe.

(23) See Deuteronomy 14:27-29.

(24) There are many Christians who claim that a tithe is not commanded under the New Testament according to 2 Corinthians 9:7, but for many the benchmark remains at ten percent. It is believed that one would not want to do less for God under grace than people did under the law of the Old Testament.

(25) See Malachi 3:8.

(26) 11 U.S.C. [section] 1325 (1998). “Income” includes all sources of income received by the debtor, and “expenses” consist of the average monthly expenses of the debtor and the debtor’s household. The difference between the income and expenses is considered the “net disposable income.”

(27) 11 U.S.C. [section] 1325(b)(2)(1998).

(28) Oliver Pollack states:

Under the [Chandler Act of 1938, where Congress first introduced Chapter 13 wage earner plans] Chapter 13 contained Official Form 13-5 which included a budget with fifteen designated expense categories. Form 13-5 was transformed under the Code to Form No. 10, otherwise known as the Chapter 13 Statement. In addition to Form 13-5% fifteen budget expense categories, Form 10 contained one additional budget expense category, Religious and other charitable contributions.”

Oliver B. Pollak, “Be Just Before You’re Generous.'” Tithing and Charitable Contributions in Bankruptcy, 29 CREIGHTON L. REV. 527, 546 (1996). He also notes that “[i]t is likely that tithing and charitable contributions occurred under the Act and the Code prior to 1984, but assumed a lower profile as the fifteen category budget may have masked the expense. Id.

(29) The statutorily defined job of the Chapter 13 trustee pursuant to 11 U.S.C. [section] 1303 is to protect the interest of the estate, which generally means to maximize the monies available to pay to unsecured creditors.

(30) See, e.g., In re Sutliff, 79 B.R. 151, 156 (Bankr. N.D.N.Y. 1987)

(31) See, e.g., In re Jones, 55 B.R. 462,465 (Bankr. D. Minn. 1985) (stating that the legislative history of Section 1325(b) is “singularly vague and unenlightening”).

(32) See infra notes 33-34.

(33) See, e.g., Michael M. Duclos, A Debtor’s Right to Tithe in Bankruptcy Under the Religious Freedom Restoration Act, 11 BANKR. DEV.J. 665 (1995)

(34) See, e.g., Waguespack v. Rodriguez, 220 B.R. 31 (W.D. La. 1998)

(35) See, e.g., In re McDaniel, 126 B.R. 782,784-85 (Bankr. D. Minn. 1991) (holding that the First Amendment would be violated to hold that tithing was not per se a reasonably necessary expense, but that the proposed level of tithing could be excessive)

(36) See supra text accompanying note 7.

(37) The Charitable Donation Act is secularly structured to include charitable contributions to any religious or charitable entity and is not limited to tithing. See 11 U.S.C. [section] 548(d)(3), (4) (1998).

(38) See, e.g., In re Bloch, 207 B.R. 944 (D. Colo. 1997)

(39) 11 U.S.C. [section] 548 (1994) (“Fraudulent Transfers and Obligations”). Specifically, trustees based their actions in these types of cases upon Section 548(a)(2) of the Code.

(40) 42 U.S.C. [section] 2000bb (1993). RFRA was the product of governmental wrangling between the executive and judicial branches, spawned by the apparent contraction in First Amendment protections induced by the U.S. Supreme Court in the case of Employment Division v. Smith, 494 U.S. 872 (1990). In Smith the Court dramatically departed from the “compelling interest” versus “religious burden” analysis established by the Court in Sherbert v. Verner, 374 U.S. 398 (1963), by relaxing the “compelling interest” analysis to inapplicability where a law is facially religion-neutral, regardless of the ultimate impact of the law. RFRA was enacted in direct response to Smith.

Not to be undone, however, the U.S. Supreme Court struck down RFRA in City of Boerne v. Flores, 521 U.S. 507 (1997), a zoning ordinance case, determining that RFRA was an unconstitutional exercise in legislative power. The Court also summarily vacated and remanded Christians v. Crystal Evangelical Free Church, 82 F. 3d 1407 (8th Cir. 1996), the only circuit case to have squarely dealt with the issue of RFRA – finding it constitutional – and a trustee’s efforts to recover tithing payments. Christians v. Crystal Evangelical Free Church, 521 U.S. 1114 (1997). In spite of this assault, the Eighth Circuit recovered and maintained its affirmation of the constitutionality of RFRA in the context of a federal law, i.e., the U. S. Bankruptcy Code, by reinstating the previous decision and thereby upholding the constitutionality of RFRA. Christians v. Crystal Evangelical Free Church, 141 F.3d 854 (8th Cir. 1998).

Finally, as an apparent attempt to affirm the viability of RFRA and its applicability to Federal law, RFRA was referred to in the Charitable Donation Act in a “Rule of Construction” to indicate that “[n]othing in the amendments is intended to limit the applicability of the Religious Freedom Restoration Act of 1993.” P.L. 105-183, [section] 6. See Steven T. McFarland, Testimony before the Judiciary Subcomm. on Commercial and Admin. Law of the United States House of Rep., at (Feb. 12, 1998) (addressing the reasons the applicability of RFRA should be reaffirmed in the Charitable Donations Act).

(41) 11 U.S.C. [section] 1325(b)(2)(A),As amended by the Charitable Donation Act.

(42) Section 548(d)(4) of the Code was added by the passage of the Charitable Donation Act and defines the term “qualified religious or charitable entity or organization” to mean “(A) an entity described in section 170(c)(1) of the Internal Revenue Code of 1986

(43) In the original version of the legislation, there was no statutorily prescribed percentage. Rather, the statute was written to contain a “reasonableness” qualifier so that a court would have to determine what amount was reasonable. With such language, the divergence of opinion and thus the legal debate as to what is “reasonable” would no doubt have continued. Through the lobbying efforts of the Christian Legal Society and others in the religious coalition, that language was replaced with specific percentages. See McFarland, supra note 40.

(44) The question is frequently asked whether a tithe should be based upon gross or net income. Many believe the tithe should be based upon gross income since this amount represents the “increase” to the recipient. Others have felt the tithe should be based upon after tax income. See The Issue: Tithing–Are We Obligated to Tithe on Our Net or on Our Gross Income at (Jan./Feb. 1990).

(45) Under state law, the applicability of which is permitted pursuant to Section 544(b) of the Code, the “reachback” period may extend back over several years time. See 144 CONG. REC. 3999 (1998) (statement of Rep. Gekas).

(46) 11 U.S.C. [section] 548(a)(2)(A) (1998).

(47) 11 U.S.C. [section] 548(a)(2)(B) (1998).

(48) 11 U.S.C. [section] 548(a)(1)(A) (1998).

(49) 144 CONG. REC. S4769 (1998). Senator Grassley cites as strong evidence of fraud “someone who is about to declare bankruptcy gives away all of his assets in donations of less than 15 percent of his income.”

(50) 11 U.S.C. [section] 1325(a)(3)(1998).

(51) In re Cavanaugh 250 B.R. 107 (B.A.P. 9th Cir. 2000).

(52) In re Buxton 228 B.R. 606 (Bankr. W.D. La. 1999).

(53) Section 1325(b)(2)(A) (1999) reads: (2) For purposes of this subsection, “disposable income” means income which is received by the debtor and which is not reasonably to be expended–(A) for the maintenance and support of the debtor or dependent of the debtor, including charitable contributions (that meet the definition of “charitable contribution” under Section 548(d)(3) to a qualified religious or charitable entity or organization (as that term is defined in Section 548(d)(4)) in an amount not to exceed 15 percent of the gross income of the debtor in the year in which the contributions are made.

(54) Cavanaugh, 250 B.R. at 712.

(55) Buxton, 228 B.R. at 611.

(56) 11 U.S.C. [section] 548(a)(2) (1998).

(57) Steven Walt, Generosity in Bankruptcy: The New Place of Charitable Contributions in Fraudulent Conveyance Law, 32 LOY. LA. L. REV. 1029 (1999).

(58) Macroeconomic decision variables are related to the performance of the overall economy, such as the Gross Domestic Product or national unemployment levels. Microeconomic variables utilize data obtained from individual sectors of the economy, such as individual household debt and income levels.

(59) These jurisdictions included the United States Bankruptcy Courts for the: Northern District of Alabama, Northern Division (Decatur, Ala.)

(60) The comparative analysis was expanded to include the broader category of debtors who have included “charitable contributions” as a necessary expense on their bankruptcy schedules, with those who actually tithe excised as a subset.

(61) This latter factor pertaining to completion of the plan will require long-term follow-up of these particular debtors and may be included in a later study.

(62) See infra Part IV.D.1.

(63) See infra Part IV.D.2.

(64) Id.

(65) See infra Part IV.D.3.a.

(66) See infra Part IV.D.3.b.

(67) See infra Part IV.D.4.

(68) See infra Part IV.D.5.

(69) The number of cases reviewed for any particular jurisdiction varied depending upon several factors, including the size of the jurisdiction, the percentage of Chapter 13 cases in a particular jurisdiction, and other non-specific variables.

(70) The random sampling for this study was done using Microsoft Excel.

(71) See infra note 74.

(72) These debt categories are as follows: Automobile

(73) The test performed is a simple two-sample t-test assuming unequal variances. This statistical technique allows for the comparison of equality of means from each sample utilizing the null hypothesis that the means of each group are statistically equal. Rejection of the null hypothesis allows for the conclusion that the means are indeed different at some specified significance level.

(74) Paul C. Bishop, A Time Series Model of the U.S. Personal Bankruptcy Rate, FDIC Division of Insurance, 98-01 BANKR. TRENDS 1 (1998).

(75) Arthur B. Kennickell et al., Family Finances in the U.S.: Recent Evidence from the Survey of Consumer Finances, FED. RES. BULL. Jan., 1997, at 1.

(76) Bureau of Economic Analysis, National Income and Product Accounts Tables–Table 1.1 Gross Domestic Product at Table=3&FirstYear+2001&Freq+Qtr (May 25,2001).

(77) Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey at (Mar. 23, 2001).

(78) Bureau of Economic Analysis, Personal Saving Rate, at (May 29, 2001).

(79) Federal Reserve Board, Household Debt-Service Burden, at (Apr. 19, 2001).

(80) Federal Reserve Board, Consumer Credit, at (May 7, 2001).

(81) The Augmented Dickey-Fuller model tests the stationary properties of time series variables. Nonstationary regressors invalidate many standard results of linear model and require special treatment so as to provide reliable estimates. See WILLIAM H. GREENE, ECONOMETRICAL ANALYSIS 847-851 (3d ed. 1997).

(82) Independent Sector, Giving and Volunteering in the United States — Findings from a National Survey (1999), at This site gives an overview of charitable giving in the U.S. for 1987-1998.

(83) Id.

(84) See infra Table 4.

(85) $90,900 charitable donations ($606 mean donation per charitable debtor times 150 charitable debtors) divided by $19,808,943 personal income ($20,699 mean income per debtor times 957 debtors). See also infra Tables 3 and 4.

(86) U.S. Census Bureau, Money Income in the United States: Current Population Reports — Consumer Income, at (Sept. 1999).

(87) U.S. Census Bureau, Historical Income Tables — Families, at (Oct. 2000). The year 1997 is used because the survey was taken in early to mid-1997.

(88) See supra note 86.

(89) See infra Table 3.

(90) Id.

(91) Marjorie L. Girth, The Role of Empirical Data in Developing Bankruptcy Legislation for Individuals, 65 IND. L.J. 17, 21 (1989)

(92) TERESA SULLIVAN, ELIZABETH WARREN & JAY LAWRENCE WESTBROOK, AS WE FORGIVE OUR DEBTORS: BANKRUPTCY AND CONSUMER CREDIT IN AMERICA (1999) [hereinafter AWFOD]. This landmark study analyzed over 1500 Chapter 7 and Chapter 13 cases from ten districts in three states: Texas, Illinois and Pennsylvania.

(93) Id. at 64, Table 4.1.

(94) The Inflation Calculator, at visited June 3,2001).

(95) See infra Table 3.

(96) AWFOD studied Illinois, Pennsylvania and Texas while the current study uses districts from Alabama, Arkansas, Georgia, Louisiana, and Tennessee. Further note that the AWFOD study contains data from Chapter 7 and 13, whereas the current study contains data from Chapter 13 only.

(97) This hypothesis is given strength when it is observed that the debt/income ratios for the general population have increased substantially from the AWFOD survey in 1981. See infra Part IV.D.3.b.

(98) See infra Table 3.

(99) Id.

(100) Id.

(101) Id.

(102) See infra Part IV.D.4.

(103) See infra Table 3. Mean charitable income ($22,592) less mean non-charitable income ($19,976) is $2,616. The sum $2,616 is 11.6% of $22,592 (mean charitable income).

(104) See infra Table 4. The sum $606 (mean charitable contribution) is 2.7% of $22,592 (mean charitable income).

(105) See infra Table 3.

(106) AWFOD, supra note 92, at 64, Table 4.1.

(107) Id. at 81 n. 14. Here we follow the lead of the AWFOD study where they use the Survey of Consumer Finances, 1983: First Report, FED. RES. BULL., Sept. 1984, and Second Report, FED. RES. BULL., Dec. 1984, to make comparisons of the general public to their bankruptcy survey. Since the reports are published every three years, AWFOD used the 1983 survey, which includes survey results from 1980 and 1983. We use Recent Changes in U.S. Family Finances: Results from the 1998 Survey of Consumer Finances, FED. RES. BULL., Jan. 2000, which includes survey results from 1995 and 1998.

(108) Arthur B. Kennickell et al., Recent Changes in U.S. Family Finances: Results from the 1998 Survey of Consumer Finances, FED. RES. BULL., Jan. 2000, at 7, Table 3 (hereinafter 1998 Survey of Consumer Finances).

(109) See infra Table 9.

(110) Id.

(111) See supra note 69.

(112) AWFOD, supra note 92, at 73.

(113) “Some investment bankers making $200,000 live at the financial edge, and some clerks making $18,000 a year have comfortable debt-free lives. Income compared with debt is particularly important in the context of financial collapse.” Id.

(114) Id. at 73-76

(115) See infra Table 5.

(116) Id.

(117) AWFOD, supra note 92, at 74-75.

(118) Id. at 75.

(119) 1998 Survey of Consumer Finances, supra note 108, at 25, Table 14.

(120) See infra Table 6A.

(121) Id.

(122) AWFOD, supra note 92, at 73.

(123) See infra Table 5.

(124) Credit card debt is all revolving credit, including credit cards, retail store, oil companies and the like.

(125) Lawrence M. Ausubel, Credit Card Defaults, Credit Card Profits, and Bankruptcy, 71 AM. BANKR. L.J. 249, 2 70 (1997) (demonstrating a direct correlation between credit card defaults and personal bankruptcies)

(126) Industry Trends, INT’L CREDIT ASSOC. CONSUMER TRENDS, Mar. 1997, at 4.

(127) Sullivan et al., supra note 91, at 133.

(128) Governor Janet L. Yellen, Testimony Before the Subcommittee on Financial Institutions and Regulatory Relief of the U.S. Senate Committee on Banking, Housing, and Urban Affairs, July 24, 1996. The complete text may be found at A survey by the Consumer Federation of America has the 1996 figure at $499 billion and 1997 at $531 billion. Stephen Brobeck, Recent Trends in Bank Credit Marketing and Indebtedness Table 1 at (July 13, 1998).

(129) Yellen, supra note 128.

(130) Id.

(131) See infra Table 7.

(132) Id.

(133) Brobeck, supra note 128.

(134) Id.

(135) AWFOD, supra note 92, at 349.

(136) Id. at 183.

(137) See infra Table 5.

(138) 1998 Survey of Consumer Finances, supra note 108, at 21

(130) Id. at 20

(140) See infra Table 7.

(141) Nat’l Network for Family Resiliency, Average Number of Credit Cards (1995), at Of the eight bankruptcy jurisdictions studied five were from Alabama and Tennessee which had, according to this 1995 survey by the Federal Reserve Board, the lowest number of credit cards per family, which was 3.0. In contrast, the region which included Illinois was 3.7, the region that included Pennsylvania was 4.1, and the region which included Texas was 3.6.

(142) Id.

(143) See infra Table 7.

(144) For nondischargeability of child support, see 11 U.S.C. [subsection] 1328(a)(2) and 523(a)(5) (1998). For nondischargeability of taxes, see 11 U.S.C. [section] 507 (1998).

(145) 11 U.S.C. [section] 1328(a)(2) (1994).

(146) 11 U.S.C. [subsection] 1325(a)(1), 1322(a)(2), 507(a)(7), (8)(1998).

(147) 11 U.S.C. [section] 1324 (1986).

(148) See infra Table 7.

(149) Id.

(150) Id.

(151) Id.

(152) Id.

(153) See infra Table 4.

(154) We multiplied the difference between the non-charitable and charitable means in the tax debt ($8,816 – $3,371 = $5,445) by the number of non-charitable tax debtors (169) to get the total amount of tax debt for non-charitable debtors ($920,205). We then divided $920,205 by the total number of non-charitable debtors in the entire study (807) to arrive at an average tax debt per non-charitable debtor ($1,140). We then assumed a 36-month plan and divided $1,140 by 36 months to arrive at an average tax debt per month of $32.

We then multiplied the difference between the non-charitable and charitable means in the child support ($9,629 – $2,832 = $6,797) by the number of non-charitable child support debtors (29) to get the total amount of child support debt for non-charitable debtors ($197,113). We then divided the $197,113 by the total number of non-charitable debtors in the entire study (807) to arrive at an average child support debt per non-charitable debtor ($244). We again assumed a 36-month plan and divided $244 by 36 months to arrive at an average child support debt per month of $7.

The total of tax debt and child support debt per month is $39 ($32 + $7). For a 12-month period the total is $468. The annual mean household income for non-charitable debtors is $20,347. By dividing $468 per year by $20,347 we arrive at the percent of tax and child support debt to annual household income, which is 2.3%.

Please note that student loans are also nondischargeable debts

(155) See infra Table 4.

(156) An adjusted R-squared of .75 means that 75% of the variance in the dependent variable, total bankruptcies, can be explained by the variables included in the regression, after adjusting for the appropriate degrees of freedom.

(157) An F-statistic allows us to test whether the variables included in the regression, when taken as a group, contribute to any predictive value. The probability value of the F-statistic, shown in Table 2 as 0.000000, gives the significance level of which we can reject the hypothesis that the set of explanatory variables has no predictive value. Thus, an F-statistic of 22.25 has an associated p-value of 0.000000, and we are most definitely able to reject this hypothesis.

Gloria Jean Lidell * Pearson Liddell, Jr. * Stephen K.. Lacewell ***

* Assistant Professor of Business Law, Mississippi State University, J.D. (Howard University), M.A. (Antioch University), B.A. (District of Columbia Teachers College). The authors gratefully acknowledge the J.W. Criss Fund administered by Mississippi State University for the funding of this project with the support of the College of Business and Industry of Mississippi State University. The authors further acknowledge the Administrative Office of the United States Courts and the Clerks of the U.S. Bankruptcy Courts and their staffs (Alabama, Arkansas, Georgia, Louisiana and Tennessee) without whose generous and kind assistance this study could not have been conducted. Additionally the authors thank Dr. Alireza Tahai for providing the statistical model for random selection of cases and our graduate assistant, Linnea Hall, whose hard work helped to bring this project together.

** Assistant Professor of Business Law, Mississippi State University, J.D. (Howard University), B.S. (Central State University).

*** Visiting Assistant Professor, Murray State University, Ph.D. (Mississippi State University — Finance), M.S.B.A. (Mississippi State University — Finance), M.B.A. (Murray State University — Finance), B.S. (University of Tennessee, Martin — Economics).