Peer-To-Peer Loans Remind The Cleveland Fed Of An Obscure Historical Episode Known As 'The Subprime Mortgage Crisis'

Except this time there's no profits.
Publish date:

UPDATE 11.20.17: The Cleveland Fed has now unpublished the peer-to-peer lending study that originally informed this article. Their mea culpa:


Since working paper no. 17-18 and related commentary on peer-to-peer lending were posted on our website on November 9, the authors have received several questions about the composition of the underlying data set they used in their analysis. In light of the comments received, the authors are currently revising their paper to further clarify the data sample they used in the study. Their revised paper will be posted as soon as it is completed.

The online lending industry has been kicking up a lot dust since the original paper came out. And if what the study's critics say is true – that the data used by the Cleveland Fed didn't distinguish between peer-to-peer and other types of lending – then the conclusions might be pretty flimsy. We'll keep an eye out for the revised paper.


Remember when everyone was wondering what JPMorgan and Citi would do to defend their market share against the rising tide of peer-to-peer lenders, whose big-data credit metrics and cost-saving disintermediation would prove an existential challenge to established banks? Haha, so do we. That was before SoFi took its heroicnosedive, before LendingClub got caught with its hand in the cookie jar, and the better part of the online lending industry began a race to the bottom.

Now that P2P is officially experiencing its growing pains, it's old enough to take a cold, hard look at its purported merits. Are consumers really better off skipping the local BofA branch and instead filling in a few online forms and having an algorithm spit out an interest rate? Are all these newfangled credit-rating tools helping to serve those traditionally brushed aside by the big boys? Is P2P good?

Researchers at the Cleveland Fed have an answer to all these questions: No. In fact, the answer is so resolutely no that the researchers blithely make the historical comparison that is to financial markets what Hitler analogies are to politics:

Signs of problems in the P2P market are appearing. Defaults on P2P loans have been increasing at an alarming rate, resembling pre-2007-crisis increases in subprime mortgage defaults, where loans of each vintage perform worse than those of prior origination years. Such a signal calls for a close examination of P2P lending practices.

Here's what that trend looks like charted:

 Cleveland Fed

Cleveland Fed

Not good!

The researchers compared a sample of 90,000 P2P borrowers against 10 million traditional borrowers to examine three P2P “myths”: that P2P helps borrowers refinance existing loans, that it helps borrowers build a better credit history, and that it banks the underbanked. Their conclusions:

We find that, on average, borrowers do not use P2P loans to refinance preexisting loans, credit scores actually go down for years after P2P borrowing, and P2P loans do not go to the markets underserved by the traditional banking system. Overall, P2P loans resemble predatory loans in terms of the segment of the consumer market they serve and their impact on consumers’ finances. Given that P2P lenders are not regulated or supervised for anti-predatory laws, lawmakers and regulators may need to revisit their position on online lending marketplaces.

Some of these points might seem counterintuitive. Take refinancing. Most P2P borrowers claim refi as their reason for taking out the online loans. Why would anyone refinance at a higher rate than they're already paying? No rational consumer would do this.

Still, the data suggest that even if P2P borrowers are refinancing, it isn't really helping their overall financial situation:

If P2P loans are used for refinancing and loan consolidation, we should not see differences in total debt balances between P2P and matched non-P2P borrowers during and after the P2P origination year. If anything, total debt balances of P2P borrowers could decline after P2P origination if (cheaper) P2P term-loans are used to refi nance (more expensive) revolving credit card debt. However, instead of this pattern, we observe total debt increasing after the P2P loan origination year....On average, the non-P2P debt balances of P2P borrowers grow about 35 percent more than those of non-P2P borrowers within two years of the P2P origination year.

The pattern is worse for those with credit card debt; compared to the control group, P2P borrowers see a 47 percent increase in credit card debt after getting an online loan.

Unsurprisingly, that doesn't do wonders for P2P borrowers' aggregate credit ratings. After taking out P2P loans, credit scores tend to fall and delinquency rates rise, the researchers found. “P2P loans have the capacity to worsen borrowers’ prospects for future access to financing,” they wrote. Here's a chart:

Cleveland Fed

Cleveland Fed

Finally, there's the question of whether P2P is serving a population that established banks have ignored, which is one of the big industry talking points. If this were the case, it might be worth looking past P2P's warts and blemishes and celebrating the model for expanding financial opportunities.

The data disagree. “We find nothing to suggest that P2P borrowers are different from non-P2P borrowers,” Cleveland writes. Zooming in on individual zip codes, the researchers find that P2P borrowers tend to be more diverse and less wealthy – signs that they might be underbanked – but also have similar debt-to-income ratios as their neighbors. Conclusion: “P2P borrowers are unlikely to be underbanked but are likely to be overleveraged even prior to obtaining their P2P loans.”

Put all this together and you get an industry that “strikingly resembles that of the subprime mortgage market before the 2007 subprime mortgage crisis,” a characterization that isn't exactly the kind of endorsement you want if you're LendingClub or OnDeck.

The saddest part of all this – from investors' perspective, at least – is that unlike the balls-out, coked-up subprime mortgage days, no one is really making money on this stuff. The backstory to SoFi's bonkers sex scandal earlier this year was a steady pressure to drum up business when internal models fell short. Meanwhile LendingClub has posted exactly two quarters of profit since going public in 2014. Its last earnings report sent shares down 20 percent. This is where the Cleveland Fed's comparisons to predatory loans break down. At least scuzzy subprime originators made money.

UPDATE: Unsurprisingly, the peer-to-peer industry isn't thrilled with this study. You can read some criticisms of the Cleveland Fed study here. If you pay a jillion dollars a month for an American Banker sub, you can read what TransUnion, the credit bureau that provided the data, has to say about the study (it's not great). There are some valid questions about which types of loans the authors grouped under the “peer-to-peer” heading. Nathaniel Hoopes from the Marketplace Lending Association adds:

When the Chicago and Philadelphia Federal Reserve researchers released an exclusive in-depth study of marketplace lending data just a month ago, they reached the opposite conclusions – finding that our loans reached the underserved and had a positive impact for borrowers.

Three Myths about Peer-to-Peer Loans [Cleveland Fed]