Payday Lending Regulation: The Substitution Effect?

A common argument made against regulating small dollar credit products like payday loans is that regulation does nothing to address demand for credit, so consumers will simply substitute their consumption from payday loans to other products:  overdraft, title loans, refund anticipation loans, pawn shops, etc. The substitution hypothesis is taken as a matter of faith, but there's surprisingly little evidence one way or the other about it (the Slips' own Angie Littwin has an nice contribution to the literature).  

The substitution hypothesis is prominently featured in a New York Times piece that is rather dour about the CFPB"s proposed payday rulemaking. Curiously, the article omits any mention of the evidence that the CFPB itself has adduced about the substitution hypothesis. The CFPB examined consumer behavior after banks ceased their "deposit advance programs" (basically bank payday lending) in response to regulatory guidance. There's a lot of data in the report, but the bottom line is that it finds little evidence of substitution from DAPs to overdraft, to payday, or even to bouncing checks. The one thing the CFPB data examine is substitution to pawn shop lending.  A recent paper by Neil Bhutta et al. finds evidence of substitution to pawn lending, but not to other types of lending, when payday loans are banned. I'd suggest that we're more likely to see a different substitution:  from short-term payday loans (45 days or less) to longer-term installment loans. That's not necessarily a bad thing...if the regulations are well-crafted to ensure that lenders aren't able to effectively recreate short-term payday loans through clever structuring of installment loans. For example, a lender could offer a 56-day loan with four bi-weekly installment payments, but with a "deferral fee" or "late fee" offered for deferring the first three bi-weekly payments. That's the same as four 14-day loans that rollover, and the "late fee" wouldn't be included in the APR.  That's perhaps an even better structure for payday lenders than they currently have.) The bigger point here is this:  even if we think that there will be substitution, not all substitution is the same, and to the extent that the substitution is to more consumer-friendly forms of credit, that's good.

While I'm griping about the NYT article, two other points. First, the article seems to have imbibed anti-CFPB talking points, such as the line that the CFPB's payday regulations could be adopted "without outside approval", as if such outside approval is generally required for any agency's rulemaking. Second, the article gets that the proposed payday rulemaking will undoubtedly put a lot of lenders out of business, but it doesn't grasp the implications, either in terms of lending volume or in terms of the financial implication on the remaining lenders. One of the big problems for payday lenders already is that there is very low volume per store. That means that fixed costs have to be amortized over a relatively small borrowing base, which keeps up prices and holds down profits. The situation is analogous to that of an island park where the deer population has grown too large for its food source. The traditional forester's response to an overpopulation problem is to cull the herd down to a sustainable size. Payday reforms that reduce the number of lenders will have the same effect as culling the weak members from herd. This means that the surviving lenders will be doing better. Indeed, some of the more efficient payday lenders get this and are quietly supportive of the proposed rule. So there might be a reduced number of lenders, but that doesn't tell us anything about what their volume will be. Colorado's payday reform experience indicates that the volume might stay roughly the same, even as loan prices fall and lender profits rise for the surviving lenders.

Comments

[Ebenezer Scrooge]

I'm intrigued by one form of substitution that Adam didn't mention--to family. One could view many of these loans as "anomie loans"--people preferring to deal with Dracula lenders than to get more deeply involved with their extended families.

By the way, payday lenders do affect the demand for credit. They advertise, after all, which helps legitimate consumer debt. One thing you can say for loansharking--nobody goes there unless they feel truly desperate.

[Adam Levitin]

Good point Scroogie. The payday borrower depositions I have read often indicate a preference to go to payday lenders rather than deal with the embarrassment or power dynamics that go with borrowing from family and friends.

I'm less sure how much payday lenders (or advertising in general) really affects the demand for credit. Any studies you know of?

[Jim]

Why would prices fall? Colorado capped prices. I think larger lenders charge more, at least they do in Texas.

[FJP]

It would be interesting to know what goods are being pawned by people who use pawn shops when payday lending is restricted. I'm particularly thinking about firearms. Guns that are in hock aren't being used in domestic violence or suicide -- or crime. There could be a social benefit from pushing borrowers toward pawn shops. (As long as the pawnbrokers are careful who they resell to, of course.)

[not_a_lawyer]

Professor Levitin elides the considerable evidence that thwarted (by regulation) payday borrowers resort to overdrafts or overdraft protection. Bhutta et al cite the relevant evidence of substitution toward overdrafts and provide new evidence that involuntary deposit account closures (due to excessive overdrafts) increase after payday bans. At $35-$70 per event, bouncing a rash of checks and having ones' checking account is very likely not "more consumer friendly" than a payday loan.

Installment loans are cheaper that payday loans because they are usually secured, but that doesn't help the erstwhile borrower if he already has a microwave and actually needs a car battery or to pay a bill. If that was a viable substitute in general, it would be happening already.

As for your fond notion that cutting store numbers and aggregate volume by half will "cull the weak members of the herd" and somehow benefit borrowers is too ludicrous to take seriously (even by blog post standards). Firstly, payday borrowers regularly say they value the convenience of payday loans above all else. Does your "shrink store numbers-cull the herd-economies of scale" model take that borrower preference into account? Secondly, if there were significant economies of scale from consolidation in payday lending it would happen organically, as it has in banking; the unnamed "more efficient lenders ...quietly supportive" of the reform would buy up the weak members of the herd and profit.