Will 2014 Be the Year the Bureau Takes on Payday Lending? If So, Wha...


 To question number one, it looks that way. According to Kim Chapman and Carter Dougherty at Bloomberg, as well as other news outlets, Director Richard Cordray has announced his intention to move forward with regulations aimed at lenders that make small loans like payday loans, title loans, and installment loans with triple digit interest rates.  The question of course is what this “oversight” and “regulation” will look like. Will the rules look something like those in Colorado, which still permits loans of up to 200% in some cases? See Pew report on this issue. Will the new regulations deem certain practices of lenders unfair and deceptive?  Will regulation of online lenders be included? What empirical data will be considered in devising these regulations? Hopefully not that found in a paper I recently read on the Online Lenders Alliance website, entitled Assessing the Optimism of Payday Loan Borrowers.

The punch line of this study is that most payday loan borrowers expect to repay their loan on its due date and not borrow again.  Indeed, most of the borrowers in the survey did just that. Paid it back quickly.  The majority of the paper’s sample population of payday loan borrowers expected to repay the loan on its due date and not borrow again.  Only 40 percent expected to have the loan out for more than one pay period, and the mean duration of indebtedness that those borrowers expected was 36 days.  Interesting. 

This is far less debt and these are far shorter borrowing periods than anything I have seen in my over five years of scrutinizing these loans.  This is also far different from the CFPB’s own data, which looks quite credible and consistent with other studies on this topic,  found here, here, here, and here.

Wondering why this study reached such drastically different results, I took a look at the methodology. All of the data came from one national payday lender who operates in California, Florida, Kansas, Louisiana, and Oklahoma.  This study’s methodology eliminated the “regular” payday loan customers and focused on the occasional user. As the paper explains, “to ensure that the borrower was not in the middle of a borrowing cycle, borrowers who had borrowed during the preceding 30 days were not eligible.” The study has a super-high response rate overall, based upon this exclusion, but we have no idea how many people actually were eligible to take the survey compared to how many walked in. That would have been interesting right? To know how many (what percentage) of the people who walked in those stores were paying on a loan versus getting a new one without being in debt already? This information was likely available but was not included in the study.

We know that the profits in the payday loan business lie in having many regular customers. Lenders quickly acknowledge that the regular repeat customers are their bread and butter. Indeed, borrowers who use the loans as marketed, for occasional, unexpected bills, wouldn’t even pay the lender’s expenses. Moreover, getting new customers to become regular customers is a big important business practice, as evidenced by all the ‘first loan free” signs we see everywhere.

So what would it do to a payday loan dataset to eliminate all the current regular customers from the mix and focus only on the others, the occasionals? I’d reckon it would make the sample highly skewed in favor of those who use the loans less frequently than average, and thus who are in debt less. Heck, by definition, the sample is those folks who are less in debt. So…before we get too excited about these data and start regulating based upon them, let’s see if they can be replicated without the exclusions. The legitimacy of these data is important as the study claims to have direct implications for the proper scope of effective regulation of the product, a topic of active concern for state and federal regulators. So a payday loan is a two-week product and people end up in debt for half the year. But can the borrowers accurately predict how long they’ll be in debt? We don’t know, because this study excluded the regular borrowers who use most of the loans.