Means Test Changes Won't Mean Much

Controversy abounds these days about whether government programs should adjust downward to reflect cost-of-living and income declines. I’d like to stir up a little controversy here at Credit Slips about Bob Lawless’ recent post that said the drop in median state income will "make it harder to file bankruptcy." First, I don’t quite follow the logic of the concern. Even if the income cut-off drops, "median" still means that half of all people are below the number. I would expect those considering bankruptcy to occupy the same places in the distribution of incomes in their state, regardless of median income fluctuations. So, it seems to me then that the fraction of potential bankruptcy debtors with above-median income would remain constant, even if the median income drops. The legal standard isn't changing, so I don't think it is fair to call the change in median income a "tightening" of the bankruptcy law. Second, even if bankruptcy filers don’t experience the general decline in income of the state’s entire population, the effect of a change in median income on bankruptcy eligibility is likely to be very, very small. Bob admits the change won't affect "a lot" of people but also says it won't be "a few." I think it really will be just a few. Why? Because the fraction of those made ineligible because of the means test is really tiny, and so even over an anticipated 1.5 million bankruptcy cases in 2009, we are looking at a minute change when we talk about adjusting the operation of the means test. In 2008, only 10% of chapter 7 debtors had above-median incomes. And nearly all of that 10% passed the means test once expenses are deducted. According to its report, the U.S. Trustee filed a motion to dismiss for abuse in 2,881 Chapter 7 cases--that works out to 4% of all above-median cases and .4% of all chapter 7 cases. Those numbers are hard to square with any fear that there will be any measurable change in the fraction of people made ineligible for chapter 7 this year. Importantly, these numbers don’t reflect how the very existence of a median income test may discourage people from filing a bankruptcy case or may push people directly to chapter 13 rather than risking an abuse determination. But again, that effect—whatever its magnitude—probably won’t change with median income fluctuation.

Comments

[FJP]

Katie, query whether the number of Ch. 7 debtors with above median incomes, and/or the number of debtors who faced a motion to dismiss for abuse, accurately measures "those made ineligible because of the means test." Where are we picking up the number of debtors who file Ch. 13, or who do not file at all, because of the means test? They are also people who were rendered ineligible for 7 by the means test, or who believe or were advised that they are ineligible.

[Raymond Bell]

I took a quick glance at the 2007 UST report that showed 2874 cases decided(484,162 chapter 7 cases filed) and 2006 where about 4163 cases were decided(833,147 chapter 7 cases filed) for abuse. Using Katie's assumption comparing each to total chapter 7 cases filed in 2007 the rate would be about .6% and .5% for 2006. I agree with her observations and adding the 2007-2006 percentages may suggest consumer lawyers became better informed and prepared before filing petitions.

[Bob Loblaw]

I thinks it's perverse that the federal government is using the cost-of-living numbers to make it harder for people to qualify for bankruptcy (even if just a little harder) while insisting that the same numbers absolutely require that a gift of $250 must be given to every social security recipiant.

[Bob Lawless]

Let's say this change affects 0.5% of all bankruptcy filers, which is a reasonable assumption given that we are in the "meaty part" of the distribution when we're talking about changes around a median. That's 7,000 filers (based on estimates for individuals filing bankruptcy). Even if it is half that number, that is more than a few. Those were the sort of numbers I had in mind when I wrote the change would affect more than a few people. Have we become so numb from our bad economic news that the financial problems of several thousand people are just a rounding error?

Also, FJP has it right. You can't just look at the people who have a motion to dismiss filed against them. The change will mean that there will be people who don't file bankruptcy, and other people who file chapter 13. Before I posted, I had heard from at least one bankruptcy attorney who was taking a second look at which cases needed to be filed before the change went into effect.

Finally, I'll stay away from a jurisprudential debate on what it means to "tighten" a law. For the people affected by this change, I'm not sure it will matter to them whether Congress initiated the change or whether it was done by an administrative agency pursuant to a previous congressional directive.

[Katie Porter]

I suspect the decline in the percent of cases challenged for abuse reflects more certainty about the standards, and not debtor attorney preparation. It's hard to be "prepared" and counsel if the standards are not clear. I predict we'll see a decline in the percent of above-median chapter 7 cases challenged for abuse in 2009. Why? Because higher unemployment is going to produce more cases with a big gap between the misnamed "current monthly income" from the past 6 months that put people over the median, and the on-the-ground reality of the debtor's current income--zero due to unemployment. I'm told the UST may ask a few questions in these cases but doesn't typically file motions to dismiss in above-median cases where the debtor has had a dramatic drop in salary because of unemployment. Stories from the field, as always, are very welcome.

[Katie Porter]

Bob-- Why aren't bankruptcy debtors as a group just as likely to have had their incomes decline as the general state population? Yes, the benchmark is going down. But shouldn't the would-be bankrupt population's income have declined in tandem with the general population of the state? There could be reasons a subgroup wouldn't reflect the general population but I can't think of any here. I am not convinced the change will affect the percentage of debtors who are above-median. There were some before, there will still be some. But if the distribution of incomes in the bankrupt population reflects changes in the general population, the number of those affected won't change (of course, the increase number of filings just means more of everything but that wasn't the point of your post).

And as you know, I absolutely acknowledge that the effect of the means test is not just who gets challenged for abuse. As I said in my post--echoing what you had said in yours, "Importantly, these numbers don’t reflect how the very existence of a median income test may discourage people from filing a bankruptcy case or may push people directly to chapter 13 rather than risking an abuse determination." We cannot know the size of this effect of the means test, and it could be quite meaningful.

[AMC]

"Why aren't bankruptcy debtors as a group just as likely to have had their incomes decline as the general state population?"

Because the Means Test is backward looking. Folks in bankruptcy are 6 months behind the population at large in terms of a declining median income level.

Plus, when you look at folks who are thinking about bankruptcy, you are likely to find a greater number of people who have experienced a job loss, or loss of overtime, or hour reduction, when compared to the general population. That is what pushes them "over the edge" from just making it into a bankruptcy filing.

Again - the Means Test is a backward looking test. The actual income of people who file bankruptcy is more likely to be less than median income going forward, but more than median income when looking backward.

The lowering of the median income threshhold is salt in the wound - and in a "Great Recession" the whole structure of the Means Test is the wound.

[Alex]

Let's think about this statistically. (Disclaimer: I am not a statistician, just a social scientist who uses statistics.) Medians are less sensitive to outliers than means, but this does not mean that that outliers have no effect on medians. If the median income has dropped because of a disappearance of outliers at the top (eg executives at bailed out firms whose compensation has been drastically cut) or an appearance of outliers at the bottom (newly unemployed), the assumption that the median divides the distribution evenly may not be justified. That is, we can't assume that "the fraction of potential bankruptcy debtors with above-median income would remain constant, even if the median income drops." And that doesn't even get into distortions that might be caused by multimodality in the distribution.

[Patches]

Well here in the 5th Cir. we apparently cannot just look at the means tests backward looking approach. 13 Trustees can also look forward and guesstimate what a debtor "might" be able to pay in the future in a plan of repayment. Say when a vehicle is paid off mid term. Even though it might not be the best thing in the world to pay a vehicle thru a 13 especially when there is only a year or two left, for the purpose of a 13 plan and keeping the payments the same throughout its better to put it in. "Forward looking approach" I think is what they call it..