Bankruptcy

The December Effect

By Bob Lawless

Most every month, I write up a post with some thoughts about the latest monthly filing statistics from AACER. I've always been a little hesitant about those monthly updates because a focus on the monthly ups and downs doesn't really tell us much about how, say, August was compared to July. If we measured filing data on a daily basis, we wouldn't say that Tuesday was a better day than Monday because there were fewer bankruptcy filings on Tuesday. The same is likely true for the monthly data measure too small of a time period to be able to say much about how one month compares to the next. Consumer financial distress unfolds over a longer time period than to be accurately captured by monthly financial data.

In addition to these concerns, one thing that drives me bats is when I see media reports about monthly bankruptcy filings that fail to adjust for the number of business days in a month. While it is true that electronic court filing means that bankruptcy filings can occur any day of the week, the number of business days in a month will be the primary driver of month-to-month changes in the filing rate.

Properly adjusted, the monthly data are useful for looking at longer trends and for trying perhaps to find recurring patterns. It has only been fairly recently that we have had good monthly data. I've only been able to get monthly data going back to January 2004, some from the Admin. Office of U.S. Courts and some from AACER. (By the way, if anyone has monthly filing data going back farther, I would be ever so grateful if you would let me know.) Looking at these monthly data, I am wondering whether we maybe don't have an annual pattern where filings fall in November and December, surge again in early spring and then plateau again in the summer.

In 2004, 2006, and 2007 (with December 2005 being anomalous because of the 2005 changes to the bankruptcy law), the December filing rate fell. The succeeding January filing rate shows a slight decline or slight increase. In February, the filing rate takes off again.

In 2004, bankruptcy filings per business day rose in February 2004 and then basically held steady or slightly declined through the spring and summer months. In 2005, the filing rate rose 13% in February, 14% in March, and 15% in April then held steady until the surge of filings to beat the effective date of the 2005 changes in U.S. bankruptcy law. Because of this surge, the fall of 2005 and spring of 2006 are too statistically anomalous to be of any use in trying to find a pattern. Filings trended upward steadily throughout 2006, although the summer months saw some of the slowest rates of increase. It was at this point, back in the early days of Credit Slips, that Ronald Mann advanced the "summer plateau" thesis in one of his guest posts. The pattern continued, with gains again in February and March 2007 with a summer plateau and the filing rate again picking up in October.

We don't have enough data points to make any definitive statements about the monthly pattern. I just wanted to throw the idea out there and make the point that these data bear watching. A December Effect does make intuitive sense to me. With the holiday season, December simply is not the time of year that people will eagerly rush into bankruptcy court. Many files that might otherwise happen in December are simply pushed into the early part of the following year. Thus, it is not surprising to see an end-of-the-year dropoff followed by a surge in the spring.

When I talked about this hypothesis with a colleague, it was suggested that credit card debt perhaps drove this effect as people loaded up their credit cards in November and December and then filed bankruptcy in the spring. Generally speaking, the consumer credit figures do show a peak for credit card debt (what the Federal Reserve calls revolving debt) in December followed by a paydown in January and February. That hypothesis just doesn't fit what we know about how most bankruptcy filers deal with financial distress, generally struggling with creditors for a year or more before filing bankruptcy. If the data continue to bear it out, the simpler explanation for the December Effect is a desire to avoid bankruptcy during the holidays or, perhaps even more likely, the time crunch that many persons feel during December.

For those who want are interested, here is an Excel file (Download MonthlyJan04toSep07.xls) with the AO's monthly filing data and the number of business days in each month as reckoned by me (using the list of federal holidays from the U.S. Office of Personnel Management).

Full post as published by Credit Slips on February 12, 2008 (boomark / email).