Charge-Offs and Bankruptcy Filings

There was some rumbling that it was odd that both credit card charge-offs and bankruptcies were down over the past year, or at least so I was told. (A "charge-off" occurs when a bank removes the account from its books as uncollectible and takes a loss.) Apparently, the reasoning was that charge-offs were alternatives to bankruptcy. A rise in one would correspond to a decrease in the other. It was not clear to me that this should be the case. It seemed more likely that charge-offs and bankruptcy filings were complementary. More bankruptcies meant more banks would charge-off credit card loans, and charge-offs could indicate future bankruptcy filings.

A wild thought occurred to me, which was to look at the data and see what the historical trends were. The bankruptcy filing data, although problematic, are readily available from the Administrative Office of the U.S. Courts, and the Federal Reserve tracks the charge-off data. The chart to the right tracks the year-to-year percentage change in the rates of total bankruptcy filings per capita and credit-card charge-offs. The red line represents credit-card charge offs, and the blue dots are for the bankruptcy filing rate. Click on the graphic for a slightly larger image.

A few comments about the data are in order. First, the data are for 12-month periods ending June 30 of each year. In my own analyses of bankruptcy filing rates, I have always used that period because historical government statistics were computed over that time frame, which coincided with the federal government's historical fiscal year. Continued use of that convention allows comparability with historical statistics. The credit-card charge-off data were taken from the Federal Reserve and represent seasonally adjusted data of charge-offs at all banks (click here for the data). The bankruptcy filing data are for total bankruptcy filings. It arguably might be better to use the government's data for nonbusiness filings, but those data are problematic as I have previously blogged and written about with fellow blogger Elizabeth Warren (The Myth of the Disappearing Business Bankruptcy, 93 Cal. L. Rev. 745 (2005)).

Looking at the graph, two things are readily apparent. First, I need to become more proficient with my statistical software, and second my university did not provide me with a decent graphics editing program (or I am just not any good with the software they did give me). Nevertheless, the graph reveals some interesting patterns. Most substantively, both statistics appear closely related and generally move in the same direction (r = 0.827). Also, credit-card charge-offs have much higher variance than do bankruptcy filings. The peaks and valleys of the charge-off line are much steeper than the corresponding peaks and valleys for the bankruptcy filing data. When charge-offs go down, they go down much more than do bankruptcy filings, and when they go up, they go up much more than bankruptcy filings.

The higher variance is difficult to explain. A working hypothesis that I have is that bank regulatory cycles somehow interact with bankruptcy filing rates. As regulators push banks to clean up their bad loans, the banks may push people toward bankruptcy. As banks' balance sheets look better, the regulators lay off, and consumers can borrow easier (and thus stave off bankruptcy for a while longer). That explanation would first show up (presumably) in banks' charge-off decisions. Right now, it sounds like a great theory, but without data it's only a theory.

Credit card delinquency rates should also be related to bankruptcy filing rates. In theory, credit card delinquencies should precede and be predictive of bankruptcy filing trends. I will post soon with data comparing credit card delinquency rates with bankruptcy filings.

Comments

[Mark T]

Major credit card banks do not collect chargedoff loans themselves. They usually sell them, because that a) enables them to eliminate expenses of maintaining staff to collect what are by definition written off assets, abd b) enables them to realize a return on them as quickly as possible which is good for reporting companies. Those that don't sell use collectors on highly contingent fee arrangements, such that the collectors determine the collection ratio not the bank. Thus, the statement "As regulators push banks to clean up their bad loans, the banks may push people toward bankruptcy." is very outdated. By cleaning up a bad loan, a bank writes it down or off on its balance sheet. Once that happens, it's logical to sell it to realize some income from it. The most one could possibly say is that as regulators demand tighter credit standards on new loans, the flow of easy money that keeps the weakest afloat for a period of timedries up and that might tip some into bankruptcy.

[Bob Lawless]

Thanks. That's helpful. Do we have reason to think that regulators focus on credit standards may cycle between periods of stringency and leniency?

[Mark T]

I would agree with that but would not know how to support it, other than anecdotally.

[Marie]

I have two charge off accounts on my credit history and they are a little over 3 years old. Is there a benefit in paying these accounts?

[Theresa Petersen]

I have two foreclosures on realestate dealings. I heard that once a bank charges it off as uncollectible that they can't 10-99 you because they wrote it off and it comes off their bottem line...is this true?