Treasury Must Not Forget That with Lending, Ignorance Isn't Bliss

The use of technology to better assess a potential borrower's risk profile and score borrowers who previously could not be scored-or were scored inaccurately-is one of the most exciting elements of marketplace lending.

The use of technology to better assess a potential borrower's risk profile and score borrowers who previously could not be scored-or were scored inaccurately-is one of the most exciting elements of marketplace lending. That is, of course, unless you are the U.S. Treasury Department, which according to its recently released report on marketplace lending worries that technology will make loan underwriting too good. That isn't a typo; the Treasury is worried that better underwriting may be "unfair" because it can disadvantage borrowers who are currently mischaracterized as low risk by making them pay higher (that is to say, correct) rates.

Such thinking is dangerous and itself unfair. It's unfair to lenders. It's unfair to borrowers who are currently and mistakenly treated as high risk. And it's unfair to the very borrowers who are currently miscategorized as low risk. Rather than something to be feared, more accurate pricing is essential to expanding capital access and protecting borrowers, and Treasury should not discourage it.

It's important to remember that the interest charged on a loan partially compensates the lender for the risk that the borrower may not pay the money back. The underwriting used to assess this risk and set the rate is inherently uncertain, since the lender doesn't know what the borrower will do and can only use information-about the borrower's circumstances, past acts, and people who are similarly situated to the borrower-to form a guess.

If the lender cannot be reasonably certain that it will be repaid, or be able to charge an interest rate that compensates for the risk of not getting repaid, the lender won't lend. Discouraging or denying lenders the ability to make the best possible guess is not only unfair to them, but it will also discourage people from putting their money at risk by lending it out, making less capital available for riskier borrowers.

Discouraging better underwriting will also be unfair to borrowers who are currently inaccurately considered high risk. The traditional tools used to inform underwriting can be incomplete and fail to capture relevant data from sources such as utility bills or rental history, which may provide insight into a borrower's likelihood of paying back a loan. Existing models may inaccurately assess risk, especially for members of groups traditionally underserved by the credit markets.

Newer methods, such as those used by marketplace lenders, may better incorporate and process relevant data, allowing for better scoring. This is obviously good for the borrower, who is now getting credit based on actual risk (or lack thereof), and is therefore paying a lower rate to borrow. These new methods are also good for lenders seeking to loan money to less-risky borrowers, because these tools enable lenders to identify potential customers and compete to serve them. Discouraging new and more accurate underwriting will continue to deny these creditworthy, but improperly scored, borrowers from accessing credit at prices they deserve.

Discouraging better underwriting is also unfair to the borrowers currently inaccurately considered low risk. Underwriting isn't prophecy. It doesn't say for certain whether the borrower will default; it just provides information about the odds of a default happening. This information, communicated through prices, informs borrowers about the risks they are taking and helps them make choices and tradeoffs. More accurate information will improve the quality of those decisions.

While the Treasury is worried about borrowers who are fallaciously considered low risk having to pay more for loans, it should be more concerned about borrowers playing with fire because they lack accurate information. If we care about borrowers' ability to repay, we should want borrowers to have as complete a picture as possible about the risks they face and make decisions informed by that knowledge.

The Treasury cites concerns that more accurate credit models may lead to stratification where those who are credit worthy get access to credit while those who are riskier are frozen out. While such a divided world is undesirable, the answer isn't to discourage accurate underwriting.

First, we should remember that the current system isn't perfect. It creates its own divisions that unfairly penalize people who are credit worthy but whose data doesn't fit the preexisting boxes. Second, inaccurate underwriting gives people at greater risk a false sense of security about their finances. Finally, lenders need information to risk their money, which is essential for credit availability, and better information should help lenders get the info they need.

Instead of discouraging accurate underwriting, the Treasury should advocate for policies that encourage entry and competition, including the use of more accurate and expansive underwriting, so that lenders and borrowers can make informed and mutually beneficial decisions.