Access, Identity, & the Cost of Complexity
Happy Wednesday, Fintech Listeners!
My wife has decided that our children are huskies. We spent all summer trying and failing to get them outside to play, but as soon as the weather turned and it started snowing, we couldn’t keep them inside!
Building snow forts. Getting into snowball fights. “Helping” us shovel the driveway.
It’s all happening now that winter has officially begun.
— Alex
P.S. — Quick favor. If you haven’t yet filled out this industry events survey, could you?
It’ll only take a minute.
I’m hoping to connect with as many of you as possible next year, and this helps us make that happen.
Should, for example, we host a 3x3 basketball tournament at Fintech Meetup next year? Tell us!
3 BIG IDEAS FROM THE PODCAST
This week’s episode of Fintech Recap is what happens when too many tabs are open.
Open banking updates. Digital identity experiments. An OG fintech bank splitting in two. Plus, internal memos from Meta and prediction markets for kids!
None of these stories are settled. But each marks a shift in how the industry is thinking about access, identity, infrastructure, and trust.
Jason and I did our best to follow the breadcrumbs.
Tune in for the full conversation here
And read below for my three big ideas...
#1: Cost RecoveryCopy anchor linkCopied
The CFPB’s revised open banking rule is expected to allow banks to charge fees for API access as long as those fees are reasonable for cost recovery. On paper, that sounds obvious. In practice, it’s anything but.
Case in point: JPMorgan Chase initially proposed high fees for data access. Then it negotiated lower ones with Plaid, while simultaneously agreeing to increase its investment in the API itself.
That sequence tells you everything. If a bank can charge less and spend more at the same time, cost isn’t the driving factor.
And it’s hard to dispute. JPMC is planning to spend roughly $18 billion this year on technology. What share of that qualifies as API infrastructure and related costs? That depends entirely on how the bank allocates it.
In other words, this isn’t about covering expenses. It’s about shaping incentives.
The CFPB may issue guidelines, but the rule is (in my estimation) unlikely to be overly prescriptive on pricing or provide a specific price that can act as a safe harbor for data providers. It’ll be too difficult to pin down. Instead, it’s likely to be determined by the negotiations between banks and data aggregators.
And the key thing to remember is that those negotiations will play out based on the market power of the companies negotiating. JPMC and Plaid will do well. Everyone else? Less so.
🎬 DIRECTOR'S COMMENTARY
One thing I’m hearing is that the CFPB is considering loosening up the secondary data restrictions in the current rule, as something of a carrot for fintech companies and data aggregators upset about the allowance of fees.
If true, it’ll be interesting to see how far the changes go and what objections, if any, banks or consumer advocates raise.
#2: Digital Identity Meets Reality Copy anchor linkCopied
Apple now lets users scan their passports and create digital identity credentials in the Apple Wallet. It works (Jason tried it!), but right now, usage is limited to 250 TSA checkpoints. And even there, success depends on the scanner functioning, the agent being trained, and the infrastructure actually recognizing what you’re presenting.
Apple’s been here before.
But unlike mobile payments, the barriers to digital ID adoption aren’t just technical. They’re also institutional.
That’s because Apple has no leverage here.
Payments are concentrated; identity is diffuse. That makes all the difference.
Apple Pay launched with fanfare, but its growth hasn’t been explosive. It’s been slow and steady. However, it’s important to remember that Apple had leverage and a centralized point to utilize it. Apple scared Visa and Mastercard into agreeing to a deal on Apple Pay that delighted Apple and pissed off issuers and merchants. But it didn’t matter that they were upset. In payments, the industry (mostly) goes where Visa and Mastercard tell it to go.
The digital identity landscape lacks this same leverage point.
Identity is scattered across 50 states, federal agencies, regulators, airlines, bars, pharmacies, websites, and on and on.
There’s no chokepoint to pressure, which means adoption will be slow and uneven in the best of times. Even if Apple nails the tech, they’re still competing with Google, Clear, Delta, banks, World (that weird Sam Altman initiative), and other identity players trying to build their own layers. No one wants to cede control. Everyone wants to be the gatekeeper.
And if you think payments have a low fault tolerance, try identity. A declined tap at the grocery store is annoying. A failed identity check at TSA can mean missing a flight or worse. People won’t trust a credential unless it works everywhere, all the time.
For banks and fintech companies building KYC or age-gated onboarding flows, this means digital ID won’t be a dependable replacement anytime soon.
Any product betting on ID portability, seamless verification, or mobile-native trust?
Until someone builds coordination, digital identity in the US will remain stuck (forgive me) at the gate.
#3: Green Dot(s)Copy anchor linkCopied
Green Dot is going private and splitting into two (as I originally wrote about in Monday’s newsletter here). Smith Ventures is buying the non-bank side. CommerceOne, also backed by Smith Ventures, is taking over Green Dot Bank and folding it into a new bank holding company.
Structurally, not much changes. From the outside (and inside!), it still looks like BaaS, with both parts tightly coupled. But the implied valuation is pretty interesting.
When Green Dot went public in 2010, it was valued at roughly $1.8 billion at the IPO price. As of late 2025, the public market values it at roughly $0.65 - $0.73 billion.
So what?
Public markets want categories. They want clarity. And when you operate across too many boundaries (regulated and unregulated, infrastructure and consumer-facing), it becomes easier for someone else to explain your value than for you to defend it.
Private equity didn’t dismantle the model. It just broke it into pieces that the public market could understand and price. That’s what private equity does best!
This is a signal to every publicly traded bank that is trying to do both. Product market fit may not be enough. If the market doesn’t understand what you are, it may not wait for you to explain it.
Even when complexity works, it isn’t always rewarded.
WHAT I'M LISTENING TO
#1: Did GENIUS Authorize the Creation of a National Uninsured Payments Charter? (Radically Clear) 🎧Copy anchor linkCopied
Is this episode one in a podcast that Michele Alt from Klaros is going to continue hosting? Or is it a one-off conversation that made its way to Spotify?
Unclear!
I hope it’s the former, but either way, I enjoyed hearing Michele, Roman, and Amias nerd out!
#2: The Winter of our Discontent: Generative AI Disrupts the Entertainment Industry Content Moat (Eye on the Market) 🎧Copy anchor linkCopied
This is one of my favorite podcasts on business and economics, and I found this episode particularly interesting.
Bonus: Engineering the SMB Capital Stack, Episode 4: The Role of Banks (by me, with Fundbox) 🎧Copy anchor linkCopied
In the finale of our miniseries with Fundbox (co-hosted by CEO Prashant Fuloria), Jackie Reses, CEO of Lead Bank, joins us to unpack how banks fit in a lending stack increasingly shaped by software. From embedded credit to re-bundled infrastructure, we dig into the collaboration between fintech companies and banks. Listen here!
Thanks for the read! Let me know what you thought by replying back to this email.
— Alex
Join 2,444 other finance and fintech leaders in the Fintech Takes Network
The conversation doesn't have to stop here
Keep learning and connecting in the Fintech Takes Network
EVENTS | FEED | LIBRARY | DIRECTORY
