Tim Taylor has a recent post about Payday loans that I found interesting. I’ve written about the empirical evidence on payday loans previously, and now I want to make a point about under what conditions they can be beneficial and what conditions they can be harmful.
The first question to ask is how do people form expectations about their future income path? Is someone’s expected future income path dependent on their current income? My sense is that people’s expected future income path is relatively stable in the short term. When someone faces a short term income shock, they expect their future income to follow the same trend as their income before the shock. Because they have this belief, they will find it beneficial to take out a payday loan in the short term to help smooth out this income shock.
For people whose future income does indeed resume its previous path, the payday loan was undoubtedly beneficial. The problem with payday loans arise when an individuals future income path is lower than his expectation; this is the reason we see many people taking out 14 payday loans in one year. Since I do believe people often suffer from optimism bias, I suspect this dynamic is fairly common and would support some form of regulation of payday loans.
Critics of payday loan regulation often point out that a lot of the alternatives to payday loans could actually be worse than those loans. Tim Taylor writes about bounced checks,
They may bounce checks and face those fees. “In 2010, bounced check fees averaged $30.47. … One study calculated the median interest rate on these loans to be well in excess of 4,000 percent, or up to 20 times that of payday loans. … The highest rates result from bouncing multiple checks for small amounts, where a fee is charged for each bounced check. Further, knowingly passing a fraudulent check is illegal and could result in substantial civil and criminal penalties.”
This is true now, but given that the use of checks has fallen 5.7% per year from 2006 to 2009, and will most likely continue to fall, the problem of bounced checks will become less and less relevant in the future. It is of course still possible to overdraw a checking account and incur overdraft fees, but thanks to the recent financial regulations, consumers can now opt out of overdraft protection. Thus, I expect the negative consequences of these two alternatives to be less relevant over time.
Taylor also points out that Pawnbrokers are approximately as expensive as Payday Lenders. The difference is that a Pawnbroker loan is collateralized, and if the customer cannot pay he simply forfeits the collateral, whereas someone with a payday loan who cannot repay it, has to take out another loan and go further into debt.
Those are the only comments I have on Taylor’s list; the whole list is an accurate overview of the alternatives faced by payday borrowers. The whole post is informative, you should read it.
My new semi-fascination with payday lending began with this post by Adam Ozimek at modeled behavior.
I’m too tired to dig up a lot of papers on payday lending right now, but my sense about the welfare affects of these loans is that they are welfare enhancing for people who take a small number of loans out to protect against income shocks, but welfare destroying for people who have to roll over loans many times.
My read of the observational evidence is that there is still no clear answer: the evidence is mixed. This is based on my research of the empirical evidence which is well summed up in this article, in the last paragraph of section “2. Payday Lending.” The entire section is very informative.
I do not know why the results are so mixed; all the studies seem to use pretty credible methods, but if I had to guess, it has to do with the method of gathering the data. If my hypothesis is correct, and payday loans are welfare increasing for one group of people but welfare destroying for others, then methods that over-select from one group or the other will affect the results.
One of the few studies that I have seen that examines the effect of payday loans on heterogeneous borrowers, and not just the average borrower, uses a lab setting, and finds exactly what I expected.
Just to note, if you do read the last study, don’t get too caught up in the solid numbers, and focus on the qualitative results. I expect the number of borrowers whose demand for loans is too high to be welfare enhancing to vary from state to state.
Finally, there is a new book out called “Broke USA” by Gary Rivlin which chronicles some of the abusive practices by payday lenders, and although I haven’t read the book, I know it is filled with anecdotes of predatory practices by payday lenders. Stories like the ones that Rivlin tells, while sad and fascinating, are probably not representative of the industry as a whole. I agree that some of the practices in his book should be outlawed, but overall payday loans are not the great evil people think they are.
Exogenous Variation 