Okay, in our first post we suggested that a behavioral model of consumer lending that hypothesized a rational lender and a heuristic borrower would predict three effects from BAPCPA:
- That the consumer bankruptcy filing rate would fall (no big surprise there).
- That the credit card charge-off rate would fall, but rebound.
- That consumers' ex ante borrowing decisions would not be altered by restricted availability of the bankruptcy discharge.
With regard to the first prediction, it is well known that bankruptcy filings spiked up in advance of the effective date of BAPCPA, and then plunged. According to the Administrative Office of the United States Courts, the number remains depressed, but it has been rebounding -- quickly in the second quarter of 2006 and a little bit more slowly in the third quarter.
It is important to recognize that it is still way too early to tell how things will settle in. Bankruptcy lawyers and bankruptcy judges are still adjusting to practice under the new Act. Over time, as practices stabilize in various districts, and lawyers regain an ability to handle cases efficiently, one might expect the filing rate to rebound. In short, we have no idea what the ultimate equilibrium rate of filing will be.
More importantly, the bankruptcy filing rate has never been a particularly good measure of the number of families in financial distress. A better indicator of how many families are having difficulty managing their debts is the percentage of credit card debt that is charged off each quarter (the lavender line below). Here, too, BAPCPA has had an effect. The charge-off rate spiked in anticipation of October 17, 2005 as families that thought they might need bankruptcy filed (and thus charged off debt) in anticipation of the new law's effectiveness. The difference here is that the rebound has been much more pronounced. While chargeoffs are not yet back to the level of late 2005, they are getting close.
Moreover, delinquency rates are already above where they were in the third quarter of 2005. Again, while we don't know where the new equilibrium will be, there is reason to think that even if the bankruptcy rate remains depressed, the effect on the level of consumer loan default rates will be much smaller.
What is most striking, however, and strongly suggestive of our behavioral model for consumer borrowing decisions is that, as far as we can tell, the massive changes in the bankruptcy law in late 2005 have not had any discernible effect on consumers' ex ante borrowing decisions. For the two key measures of consumer indebtedness published by the Federal Reserve system, the consumer Financial Obligations Ratio and Debt Service Ratio, October 17, 2005 was a non-event. Consumer debt loads showed no discernible change surrounding the effective date of BAPCPA.
Finally, the aggregate amount of consumer debt outstanding continues to grow as well.
Indeed, aggregate credit card debt has increased at rates that are higher (or at least as high) as the pre-BAPCPA rates of increase.
In short, a behavioral model of consumer lending suggests that tightening up on the bankruptcy discharge will do very little to change ex ante consumer borrowing decisions, and hence it will do little to change either the underlying level of credit card default or the number of families in financial distress. We see little in the above data to contradict this intuition.