Further debate about debt collection reform and credit availability

The Center for Responsible Lending has produced a nice, new empirical paper reflecting on and refuting the notion that certain debt collection reforms restrict the flow of consumer credit. The analysis is careful and impressive, and the natural laboratory experiment they found is fun and intriguing. In a nutshell, North Carolina in 2009 and Maryland in 2012 imposed new restrictions on debt buyers suing consumer debtors on purchased accounts (both states now require actual documentation of the debts and their ownership to support such suits). On cue, in the period leading to these reforms, the credit lobby predicted gloom and doom in terms of restricted access to credit, especially to sub-prime borrowers, if such liberal nanny-laws were adopted. Several years later, the CRL decided to look back and test this. Comparing the change in the number and dollar volume of new credit lines in North Carolina and Maryland in the two years before and after each of the reforms (coincidentally, periods of general economic contraction and recovery, respectively), and comparing these differences with comparable data for selected peer states and the nation as a whole, did the reforms seem to have a noticeable effect of reduced access to credit in these states?  The simple answer, of course, is no (i.e., less contraction in North Carolina than elsewhere during a recession, and more expansion in Maryland than elsewhere during recovery). The more nuanced answer means the debate will rage on.

The good: The reforms to North Carolina and Maryland debt-buyer practices evidently had no negative effect on credit availability. Indeed, sub-prime borrowers in both jurisdictions fared substantially better than their peer-state and national counterparts in terms of new credit extended post-reform. The authors concede that sustained (and even greater) credit extended to sub-prime borrowers may well not represent a positive in terms of consumer welfare. But reforms making it more difficult for debt buyers to collect on defaulted debt seem rather clearly not to have had the constricting effect on availability of credit that opponents of these reforms warned of.

The not-so-good: These findings are rather limited. The title of the paper is “Debt-collection reforms that protect consumers not found to restrict credit availability.” That’s a punchy, attractive headline. Unfortunately, what the paper really establishes would require a much longer, unwieldy title:  Debt-collection reforms that protect consumers from debt buyers (not necessarily general debt collectors) in litigation (not at all in the general phone-and-letter collection practice) not found to restrict credit availability. The paper offers some hints about how important these caveats are. First, while the nine largest debt-buyers purchased about $140 billion in defaulted accounts from 2006-09 (roughly $35 billion per year), the three largest buyers pursued litigation to collect only $650 million in 2010 and only about $1 billion in the three heaviest, later years (2012-14). It thus looks like buyers resort to litigation to collect at most only about 2-3% of purchased debts. It is not particularly surprising that the predictors of doom were obviously wrong when suggesting that reforms affecting only 2-3% of collections would severely restrict credit. Moreover, the reforms affected only debt buyers, not third-party debt collectors generally (and certainly not creditors collecting their own debts in-house). Again, it is not at all surprising that the credit industry would all but ignore a reform aimed only at a thin slice of credit extended (defaulted debt sold to buyers), and only at a small fraction of the activities of the debt-buying industry (litigation over 2-3% of purchased claims). I’m not making the standard critique of empirical research (“that was obvious!”); instead, I’m wholeheartedly agreeing with the authors that these marginal reforms could not possibly have been accurately predicted to have much effect on credit availability. But proving the lying liars wrong is a far stretch from establishing what I’d really like to see; i.e., that general debt collection reform, or better yet national bankruptcy law, has little or no effect on credit extension. I suspect this is true (at least over the long term), but solid research on the subject is sparse (though a nice collection of citations appears in the paper discussed immediately below).

Much to their credit, the authors acknowledge that another paper supports the restriction thesis in the context of general debt collection reforms. Writing for the Federal Reserve Bank of Philadelphia, Dr. Viktar Fedaseyeu reports that his econometric model seems to show that legislative restrictions on general debt collection activity lower the number of debt collectors, their recovery rates, and consequently the number of new credit card openings (by about 2% on average for every new restriction “point,” with a maximum of 8 points). I hate to level at this study the other standard critique of empirical research (“that can’t be true!”), but Dr. Fedaseyeu’s model is not particularly finely tuned (the “restriction” inputs seem wildly varying in importance and heavily weighted toward minutia like bonding requirements, though the model assigns them equal weight) and the result strikes me as modest at best (2% for a new restriction "point," with only a 6% difference between the 25th and 75th percentiles of point totals). It’s nice to simplify complex data, but I’m afraid this study crosses the line to oversimplification to cram a messy world into a nice, econometric model. One interesting statement buried late in the paper (p. 24) relates to the bugaboo about the effect of debtor protections on interest rates: “credit card interest rates do not appear to react to changes in debt collection regulations.” Hmmm. So does this paper help or hurt the credit lobby's standard script?

The debate will rage on, and these two papers make significant, unique, and valuable contributions to it.