The Fintechs Are Banks Now (Matt Levine,


If you are going to securitize bank loans, you will need a bank. You don’t need to be a bank, but you will need a bank to make the loans and assign them to you for securitization. 1  If you are going to securitize bank loans and call them “peer-to-peer” loans—someone borrowed the money, someone else provided the money, maybe they’re peers, why not—then you will also need a bank. I suppose there may have been a brief period when people imagined that they’d just set up an online marketplace where people with money could lend it to people who need money, on a purely decentralized peer-to-peer basis, but actually existing peer-to-peer lending companies are in the business of (1) connecting individual borrowers to banks, (2) buying those loans from the banks, and (3) selling those loans to investors.

The reasons for this are essentially regulatory: U.S. law prefers personal loans made by banks and disfavors personal loans made by other companies, and is mostly pretty flexible about what happens a second before and a second after the loan is made. 2  So if you are a financial technology company you can build the website that markets and explains loans to people, and you can connect people to banks to get the loans, and you can buy the loans from the banks, and you can effectively be the entire consumer front-end and the entire financial back-end, but in the middle everything has to flow through a bank for a second to get the bank’s blessing.

Arguably this is good, because banks are heavily regulated and required to be prudent, so if you’re doing something weird and bad with your lending platform, the second that your loans spend at a bank will give the regulators information and jurisdiction to stop you doing the bad thing. Or arguably it’s bad, because banking regulators are conservative or captured or entrenched and they’ll use their jurisdiction to stop you from doing weird and good innovative things.

But mostly it just strikes me as sort of a technical feature. To make the loans you need a bank, but in 2020 you don’t need a bank in any sort of thick old-fashioned way. You don’t need to walk down to the local bank branch and shake hands with a banker to get your loans approved; banks have computers now, even APIs. You don’t need to partner with a big famous ancient banking institution; there are small and new and online-focused banks that don’t even have branches. If your goal as a fintech company is to disintermediate banks, to cut the traditional banking system out of your transactions, you kind of can. You just need a bank, though. It doesn’t have to be JPMorgan. 

A bank charter is a particular piece of technology, like a mobile app or a blockchain, only it’s a regulatory technology rather than a computer technology. It is just a part of the technology stack that you are engineering, in providing your online lending platform. Perhaps you could engineer around it, as a sort of Oulipo exercise in making life more difficult for yourself, but why? A bank charter is a simple ready-made modular solution for some of the regulatory problems in making loans, so if you are making loans you might as well just plug one in.

Of course to plug in a bank charter you have to rent one (by paying a bank to issue your loans). Or you could buy one:

LendingClub Corp. got its start replacing old-school bankers with machines that match borrowers and investors. Almost 15 years later, it’s planning to become a bank itself.

The online credit marketplace is buying Radius Bancorp in a cash-and-stock transaction valued at $185 million, according to a statement Tuesday. The acquisition of Radius, which has $1.4 billion in assets, will give LendingClub greater regulatory clarity and a less-expensive form of funding for its loans, the San Francisco-based company said.

“This is a transformational transaction that allows us to re-imagine banking in a way that is free from legacy practices and systems,” LendingClub Chief Executive Officer Scott Sanborn said in the statement.

LendingClub is the latest financial technology company to seek a bank charter, which allows firms to take deposits directly from consumers and use them to fund loans. Varo Money Inc., an online bank, has received approval from the Federal Deposit Insurance Corp. and the Office of the Comptroller of the Currency as part of its own pursuit of a charter, and On Deck Capital Inc. has said it will spend about $5 million this year to seek a charter.

Here is the press release, touting Radius as “a leading online bank founded in 1987” known for its “branchless digital banking platform” and “open APIs to offer ‘banking-as-a-service’ (BaaS) functionality to leading fintechs.” This all seems … yes, obvious, fine? Good? You run a fintech, there is a technology layer called “banking,” you use a specialist provider to implement that layer, eventually you bring them in-house, sure, right. It is thin banking, “banking-as-a-service,” banking reimagined “in a way that is free from legacy practices and systems,” banking as, you know, the split second of official blessing of the loans between your consumer front-end and your investor back-end. 3  

The weird thing is that anyone ever thought otherwise. “LendingClub Corp. got its start replacing old-school bankers with machines that match borrowers and investors.” It’s still doing exactly that. You don’t need old-school bankers. You’re still using machines to match borrowers and investors. It’s just that one component of the machine is a bank charter.

Leave a Reply