 | |||||||||||
| |||||||||||
| | |||||||||||
 Happy Wednesday, Fintech Listeners! I hope your week is going well. I’ve been up in the mountains of Montana for the whole time, with limited cell reception, which has been just as wonderful as it sounds. More about that on Friday! In the meantime, podcasts! — Alex P.S. — Healthy skepticism around stablecoin card programs is warranted, which is what makes Western Union’s moves in that space so meaningful. On June 30th, I'm sitting down with the team behind them. If stablecoins or stablecoin cards are anywhere on your radar, this is the conversation to listen to. Register here to join us! Was this email forwarded to you? Sponsored by C&R Software Bad debt is inevitable. How you handle it isn't. 3 BIG IDEAS FROM THE PODCAST  This week on Fintech Recap, Jason Mikula joined me from the Netherlands, where he had just survived a 90-degree day with no air conditioning. What fun. Speaking of fun, we covered the neverending fintech-to-bank charterfall, an unexpected return to BaaS Island, two White House executive orders, the Federal Reserve's convoluted master account situationship, and closed with two Can’t Let It Go rants that you need to stick around for. Trust me on that last part. Tune in for the full conversation here And read below for my three big ideas... #1: The Charter … Trap?Chime confirmed it will (at some point) become a bank. Mercury got conditional OCC approval on its national bank charter and raised $200 million at a $5.2 billion valuation in the same week. The narrative is that bank charters unlock better economics, lower funding costs, and a more defensible business model. All of that is basically true. And yet. Chime's revenue multiple went from roughly 30x in 2021 to 7x when it went public last year to roughly 3x today. Mercury's post-fundraise multiple sits around 8x. SoFi, which went the distance and became a bank in 2022, trades at 2.2x price-to-tangible book value. LendingClub (which became a bank in 2021) is around 1.25x. Both those ratios (which are very different from the simple revenue multiple ratio that we grade tech companies on) are good for banks (SoFi’s is quite good) but far away from the revenue multiples they used to trade at. The pattern is consistent: each gate a company passes through, from private to public, from non-bank to bank, compresses the multiple further, regardless of what it does to the underlying business. The business may get better, but the story gets trickier to tell. The reason why, as I’ve written about, is partly structural: bank investors don't ask what a business could be worth in ten years. They ask what it could be liquidated for tomorrow. That skepticism is baked in, which is why fintech companies (especially the public ones) are often reluctant to embrace the charter, despite its obvious advantages. #2: Deregulation by OutlierThe OCC has issued a consent order against Community Federal Savings Bank (CFSB), a single-branch institution in Woodhaven, Queens. The bank grew from $140 million in 2017 to ~$900 million in assets at the end of 2024 by running fintech partner programs for Airwallex, Wise, and Payoneer, among others. They had a very high percentage of transaction monitoring alerts closed automatically, sans investigation. Plus, CFSB couldn't confirm whether it held correspondent accounts for foreign financial institutions at all. The OCC took action, which is good. However, take note of what the order doesn't include. There are no board governance findings, no third-party risk management findings, no restrictions on bringing on new programs. Prior BaaS enforcement actions routinely covered all three alongside BSA/AML. This one didn't. That narrowness probably isn’t an accident. This consent order functions similarly to the CFTC's insider trading cases in prediction market land. You deregulate broadly, and then you point to an egregious outlier that’s so egregious nobody can argue otherwise. CFSB, a bank that grew 540% in seven years by running cross-border programs (while apparently unaware it was engaged in correspondent banking!) qualifies. What the OCC’s consent order doesn't do is draw a line. A bank with BSA/AML failures this severe has a third-party risk management problem by definition. The order doesn't say that. A bank onboarding new programs while being this deficient should probably need supervisory non-objection before doing so, but the order doesn't require that either. As Jason put it, the current regulatory leadership appears to believe prior agencies were too expansive in whacking banks on reputation risk and third-party risk management, alongside financial crimes findings. I think there’s some truth to that (see Blue Ridge Bank), but the narrowness of this consent order feels like an overcorrection. #3: The Fed’s Power ProblemThe Federal Reserve does three big jobs. It operates a payment system, it acts as a prudential regulator of financial institutions, and it sets monetary policy. Only one of these things requires absolute independence from the political process. The case for central bank independence in monetary policy is airtight. There are decades of examples, Turkey, Argentina, the list goes on and on. We know what happens when governments exercise too much control over interest rates. But the Fed's other two jobs don't carry the same logic. When it comes to prudential regulations, why does the Fed have the job at all? The OCC supervises national banks. The FDIC supervises state-chartered insured institutions. Perhaps the Fed's role as a bank regulator is one it shouldn't have. Jason's reporting on Evolve during the Synapse collapse, which included repeated FOIA requests to the Fed, illustrates why: there were no clear answers. There was no accountability. The Fed’s third job is operating the payment system, and what I’ve come to realize is that once you house the payment system inside an institution whose independence is justified by monetary policy, that independence starts bleeding into areas where it probably doesn’t belong. The master account situation is where this becomes most obvious. Master accounts are basically bank accounts for banks at the Federal Reserve. They give institutions access to the Fed’s payment rails, intraday overdrafts, emergency liquidity through the discount window, and interest on reserve balances, among other benefits. Who gets one, and who doesn’t, is decided by the twelve regional Reserve Banks with zero accountability or oversight. As documented in this tweet, the Kansas City Fed alone granted one account, revoked it after some deserved bad press, denied another applicant after years of silence, got sued, won, and then granted a third applicant a category of account that hadn’t even been formally defined yet. Unbelievably, none of those decisions required a public explanation. The Fed’s independence exists to protect monetary policy from political pressure. It was never designed to shield opaque decisions about who does and doesn’t get access to the financial system’s core infrastructure. Those are fundamentally different questions. Sponsored by Ocrolus Inc. Manual small business underwriting doesn't scale; different underwriters reach different conclusions from the same merchant data. WHAT I'M LISTENING TO #1: Here's how many Americans are cutting their food costs (Consider This) 🎧It’s wild how many great podcasts NPR is pumping out. This is one that is highly relevant to anyone operating in B2C financial services. #2: Is Washington Finally Rewriting Bank Supervision? (Banking With Interest) 🎧A great conversation on a question I have been quietly wondering about for a while: How is the Trump Administration reenvisioning the role of the prudential bank supervisor? *Bonus: Banking on Primacy, Episode 4: The AI Episode (Fintech Takes x Chime) 🎧When AI starts making financial decisions for consumers, whose side is it on? Episode 4 with Ryan King, technical Co-Founder at Chime, covers why AI is a slope change (not a step change), what that means for us, and how the business model answers the trust question. Listen here. *This rec is brought to you by one of our fantastic brand partners. Thanks for the read! Let me know what you thought by replying back to this email. — Alex | |||||||||||
|
{beacon}
Workweek Newsletter
