Rigbi on Usury on Prosper

When Credit Slips started, we intended to feature new scholarly papers by the bloggers and others. I am going to attempt to revive that tradition by featuring a paper by Oren Rigbi, an assistant professor in the Department of Economics at Ben-Gurion University of the Negev. Rigbi’s paper, “The Effects of Usury Laws: Evidence from the Online Loan Market,” exploits a change in the lending rules that apply to Prosper.com to examine the effects of interest rate caps. Prosper.com is an online lending web site, as Katie Porter explained just after it launched. In April 2008, a change in the way Prosper is organized meant that the interest rate cap was raised to 36% where previously some borrowers had a lower cap (depending on the state where the borrower lived). Thus, Rigbi was able to explore the effects of raising an interest rate cap on the ability to borrow, the amount borrowed, the interest rate for the loan, and repayments.

There are certainly differences across borrowers, time, and states, but Rigbi uses careful empirical analysis to control for these differences. What’s left is a measurement just of the effect of the changing in the interest rate cap. Rigbi summarizes his findings as follows: “I find that higher interest rate caps increase the probability that a loan is funded, especially if the borrower is risky and previously been just ‘outside the money.’ I do not find that borrowers change the loan amounts they request or that their probability of default rises. The interest rate paid for all loans, however, rises slightly probably because online lending is imperfectly integrated with credit markets.” Rigbi concludes his paper by saying, “The main takeaway point from this inquiry is that interest rate restrictions do not seem to deliver the outcomes for which they were intended.” My description of the methodology and findings glosses over a great deal of detail. Rigbi was kind enough to indulge me in an e-mail exchange and even kinder to allow me to reproduce it here:

Lawless: At the end of your paper, you discuss whether your findings about Prosper.com would generalize to interest rate caps on day-to-day, face-to-face consumer lending. Frankly, I am very skeptical that they would. As you note, Prosper.com is highly competitive with large numbers of lenders competing for loans. It also uses what is essentially an auction system so the borrowers get an interest rate as determined by an auction among lenders. It is online, making transaction costs very low for borrowers to get a low interest rate. As an online site, I also wondered whether its clientele was typical of the low-income persons who are typically the subject of predatory lending and usury concerns.

Rigbi: An attempt to evaluate the extent to which the results can be generalized to other credit markets should focus on factors that are important according to economic theory and that differs between Prosper.com and other credit markets. Indeed, the result that interest rate restriction hardly affects the price paid by borrowers probably depends on the credit market being fairly competitive, and is less likely to hold when lenders have market power. The population of borrowers in the sample includes many lenders with credit scores as low as 520, and 15% of them self-report themselves as having their annual income being lower than $25,000. Therefore, at least some of those who should be protected by usury laws are represented in Prosper.com. Furthermore, the estimation procedure used allows the effect to vary with the borrowers’ credit score. The findings are presented for all the groups of borrowers including those at the bottom of the credit score distribution.

Lawless: In your abstract (quoted above), you write that lifting the interest rate cap did not increase the probability of default. Can you elaborate on that finding? It seems counter-intuitive because, assuming competitive markets, the lifting of an interest-rate cap should lead to riskier borrowers receiving loans with a concomitant increase in the default rate. Is this finding evidence of mispricing in the Prosper.com market, namely that lenders overestimate the probability of default?

Rigbi: As you correctly anticipate, the default probability of loans originated under a high interest rate cap is higher than the default probability of loans originated under a low cap. The reason for this is that more high risk borrowers receive loans under a high cap. The result of no change in the default probability, however, is conditional on the risk associated with a borrower. The thought experiment to have in mind is the following: imagine two borrowers with the same observed characteristics who receive loans such that one faces a low cap while the other faces a high cap. The question addressed in the paper is whether these borrowers have ex-post different probabilities to default. The answer to this question is no.

Since I find that similar borrowers pay more or less the same interest rate under the two interest rate caps, a finding other than no change in the default probability would suggest that the borrowers that are receiving loans under the different caps have different unobserved characteristics which affect their ability to repay the loan. In the language of economists, this is referred as selection on unobservables, which makes the results harder to interpret.

Lawless: One thing you cannot observe is the effect of repealing interest rate caps on the culture of the Prosper.com community. One aspect of interest rate caps that appeals to me is what they express about the expectations of the community in which they operate. In addition to the formal aspects of making an activity illegal—for example, one cannot collect a usurious rate of interest in the courts—illegality puts an activity outside social acceptance. Communities with fewer formal norms might be less cohesive with more self-interested behavior. Repealing usury laws might lead to perceptions that the markets in which they operated have become depersonalized and outside social conventions that otherwise constrain self-interested behavior. All of this might lead to more predatory behavior by lenders and a laxer attitude of responsibility by borrowers. As the interest rate caps on Prosper.com have been raised and as the online site has matured, do you think these sorts of dynamics have happened? You know the culture of the site and its online community as well as anyone.

Rigbi: In the first couple of years for its activity, Prosper has colored the borrowing activity it has facilitated with a social flavor. The idea was that borrowers who might be constrained from traditional credit were able to convey their personal, sometimes sad, stories and to convince lenders that they will be able to repay their loan. As the marketplace matured, the fact the borrowers’ stories were not verified and that a borrower can attach any story or photo to his loan requests, gradually reduced the importance lenders attributed to pieces of information that were not verified. Thus, one aspect of the marketplace maturity is the greater reliance of lenders on pieces of information that their content was verified.

Furthermore, I suspect that repealing interest rate caps would not make lending behavior more predatory mainly due to two factors. First, the supply of credit in Prosper is fairly competitive. Second, the way the web site operates, the borrower sets the maximum interest rate he is willing to pay while having the interest rate cap before his eyes. Most lenders are probably not even aware of the interest cap existence, and as such are less likely to change their behavior once it is repealed.

Thanks again to Professor Rigbi for indulging in this conversation and shedding some light on a working paper that Credit Slips readers might miss.

Comments

[mt]

Thanks for attention to that paper. I thought it missed the two main points of interest rate caps. It's not just balancing credit availability and default probability.

1) High interest rates transfer more wealth from borrower to lender than lower ones. In consumer finance, the wealth disparity is generally significantly in favor of the lender to begin with. So high interest caps foster more wealth disparity.

Second, it is implausible that borrowers are consistently obtaining returns on investment in excess of 36%. This is why consumers always feel they are falling further and further behind when they borrow at such high rates. Interest rate caps should be set at levels that permit positive returns at more plausible rates of return. While that is paternalistic, it is obvious that most consumer borrowere do not have enough financial literacy to understand the foregoing and protect themselves against the negative returns they are signing up for.