Bankruptcy Filings Over the Next 12 Months

Bankruptcy filings have continued to decline, and using data supplied by Epiq Systems, my analysis suggests the same trend will occur over the next twelve months.

First, looking back at March and April, bankruptcy filings declined 12.0% and 11.8% respectively on a year-over-year basis. The daily filing rate in March was just short of 4,900 and in April was 4,575.

If we extrapolate these numbers out over the rest of the year, bankruptcy filings should just barely surpass 1.0 million for calendar year 2013. If filings continue for the rest of 2013 at the same pace they have for the first four months, the number will be closer to 1.1 million, but the filing rate is typically strongest in the spring and then tapers off each year. Thus, a figure closer to 1.0 million is more likely and that rate of filing would represent an annual decline of 14% in the number of U.S. bankruptcies for the second straight year in a row.

These predictions are mere extrapolations, and I figured I could do better. Therefore, I fed a bunch of data into my computer, Deep Thought. After a few moments, Deep Thought decided the answer was 42. Upon my motion for reconsideration, Deep Thought predicted just over 1.0 million bankruptcy filings over the next twelve months. The basic point here is that both the simple extrapolation and a more complex analysis suggest about the same answer. 

For those of you that must know the details, I did a vector autoregression using bankruptcy filings, outstanding consumer credit, and personal income. This particular analysis has some advantages over more basic regression techniques, among them being that one can compute a dynamic forecast of all the elements in your analysis. The results do suggest that changes in consumer credit precede bankruptcy filing rates and not vice versa. Outstanding consumer credit -- both revolving and nonrevolving -- along with personal income seem to be strongly associated with bankruptcy filing rates. Adding other variables to the model does not improve its performance.

Having made a prediction, it is important to consider the ways it could be wrong. The analysis I did looked at historical data, and if the future does not look like the past, the analysis will be off. One possibility is that the historic relationship between consumer credit outstanding and bankruptcy filings has changed since the financial crisis of 2007, although I think that is unlikely given that the early returns suggest it has not.

Another possibility is the 2005 extension of the general ban on repeat bankruptcy discharges from 6- to 8-years will result in a slug of repeat filers entering the bankruptcy system later this year. Those persons who rushed to file to beat the October 2005 effective date of the bankruptcy reform law will begin becoming eligible for another bankruptcy discharge later in 2013. Although I think this scenario has some plausibility, my instinct is that there will not be enough repeat filers to drastically alter the predictions. There would have to be a huge percentage of the persons who filed in July - October of 2005 just sitting on the sidelines for the moment when they can refile. Social science research on bankruptcy filings suggest the bankruptcy decision does not work that way for most every household. Maybe practitioners who follow the blog can comment on whether these potential refilers from 2005 exist in large numbers. Are potential clients coming into the office now only to be given the advice to wait a few months?

The best guess for bankruptcy filings over the next 12 months is just over 1.0 million cases. Absolute numbers can sometimes be taken to mean absolute precision, and it should be kept in mind this is only a prediction. The more quantitative analysis suggests a 95% confidence interval of between 800,000 and 1.2 million filings. Circumstances can change, and past performance is no guarantee of future returns.