Latest / Fintech Conversations with Fexingo: Banking, Payments, and Financial Software Companies / How Open Banking Transformed Consumer Lending in 2026
Transcript
- Lucas: If you're building or running a lending business in 2026, open banking has probably changed your underwriting model more than anything else in the last decade. Luna: It's not just about replacing the credit score. It's about accessing real-time cash flow data from someone's bank account. That's a fundamentally different picture of risk. Lucas: Exactly. And I think the numbers make the case. Lenders using cash-flow underwriting via open banking APIs have seen approval rates for thin-file borrowers — people with little or no traditional credit history — increase by about 30 percent. At the same time, default rates on those same loans have dropped roughly 40 percent compared to traditional fico based models. Lucas: Quick honest thing — we're able to dig into data like this because of a small group of listeners who chip in monthly through buy me a coffee dot com slash fexingo. That support keeps the show ad-free and lets us focus on what's actually changing in financial services. Luna: Yeah, and it means we don't have to chase sponsors or pad episodes. Just the real conversation. So — back to that underwriting shift. The CFPB's Section 1033 rule, which went into effect last year, was the catalyst, right? Lucas: Right. The rule required banks to make consumer financial data available through standardized APIs — with the consumer's permission. That opened the floodgates. Suddenly, a lender could see not just your credit card payments, but your rent, your utility bills, your gig economy income. All of it, in near real time. Luna: And that's a huge deal for the 45 million Americans who are 'credit invisible' — people without a FICO score. They might pay rent on time every month, but that data never made it into the credit bureaus. Lucas: Exactly. So companies like Finicity — which Mastercard acquired back in 2020 — and Yodlee, which is now part of Envestnet, built the plumbing. They aggregate data from thousands of financial institutions and give lenders a standardized view of income, expenses, and cash flow patterns. Luna: What's interesting is that this isn't just for subprime or thin-file borrowers. Even prime borrowers get faster decisions. I've seen data from a major auto lender — they cut loan approval time from three days to under an hour using open banking data. Lucas: Speed is a big part of it. But the real innovation is the shift from a backward-looking credit score to a forward-looking cash flow model. If you see that someone's income has been stable for six months and they have a consistent savings pattern, you can lend with more confidence than if you just know they paid a credit card on time for two years. Luna: There's also the privacy angle. Consumers have to explicitly consent to data sharing, and they can revoke access at any time. The CFPB's rule included strong data minimization requirements — lenders can only request the data they actually need for the specific product. Lucas: That's a key safeguard. But the implementation hasn't been seamless. Banks had to build APIs, and there was a lot of tension over who controls the user experience. Some banks tried to 'screen scrape' — that is, let the aggregator use the customer's login credentials — but the rule pushed toward token-based API access, which is more secure. Luna: And not all banks were happy about it. They worried that open banking would disintermediate them — that fintechs would cherry-pick their best customers for loans while leaving the banks with the less profitable relationships. Lucas: That fear has been overblown. In practice, many banks have become data providers for fintechs, and they've also started using the same data to improve their own lending. JPMorgan Chase, for example, rolled out a cash-flow-based underwriting product for small businesses in 2025, using data from their own deposit accounts. Luna: Small business lending is actually an area where this is huge. Traditional small business underwriting relies heavily on the owner's personal credit score, which often doesn't reflect the business's actual health. With cash-flow data, lenders can see the business's revenue, expenses, and seasonal patterns. Lucas: Right. And the same applies to gig workers. If you drive for Uber or DoorDash, your income fluctuates week to week. A traditional lender might see that as risk. But with open banking data, a lender can see your average weekly earnings over the past year and lend accordingly. Luna: There's a startup called Even Financial that partners with banks to offer instant personal loans based on cash flow. They claim their model reduces default rates by 25 percent compared to traditional underwriting. I'd love to see those numbers independently audited, but it's plausible. Lucas: It's plausible because the data is richer. But there's a risk: if all lenders use the same cash-flow model, you could create a new kind of systemic risk. Everyone might pull back at the same time if they see a broad decline in consumer cash flow. Luna: That's a good point. And there's also the question of data accuracy. If a consumer's bank account has an error — say, a fraudulent charge or a delayed deposit — that could cause a loan denial. There needs to be a dispute mechanism. Lucas: The CFPB's rule does require that consumers have a way to correct inaccurate data, but it's early days. So far, the complaint volume has been low, but as adoption scales, I expect we'll see more issues. Luna: Let's talk about the competitive landscape. Who's winning in this space? Obviously the aggregators like Plaid and Finicity. But also the lenders that moved early. Lucas: Plaid is interesting because they started as a consumer-facing tool — letting people link accounts to apps like Venmo. Now they're a full-fledged data platform for lenders. They process connections to over 12,000 financial institutions. And they've built analytics on top of that data to help lenders decide. Luna: And Finicity — Mastercard bought them for roughly $1 billion in 2020. That acquisition looks prescient now. Mastercard has been pushing open banking globally, and Finicity's technology is central to that. Lucas: On the lender side, companies like Upstart and LendingClub have been aggressive adopters. Upstart's AI model already used alternative data, but open banking gave them a direct feed of bank transactions. They've said that borrowers who share bank account data have a 15 percent higher approval rate and a 20 percent lower APR. Luna: I want to talk about the global picture. The US was actually late to open banking. The UK and Europe have had it for years under PSD2. In Australia, the Consumer Data Right mandates it. What can the US learn from those markets? Lucas: The UK's experience shows that adoption takes time. Even with mandated APIs, it took about four years for a majority of banks to have functional endpoints. And the US has a much more fragmented banking system — thousands of small banks and credit unions that don't have the resources to build APIs. Luna: So there's a role for third-party middleware that standardizes the data for those smaller institutions. Companies like Akoya and MX Technologies do that. They act as a bridge. Lucas: Yes, and the CFPB recognized that. Their rule allows for 'representatives' — essentially, data aggregators that can act on behalf of consumers to request data from banks. That keeps the burden off small banks while still giving consumers control. Luna: One concern I hear from privacy advocates is that once you link your bank account to a lender, they can monitor your transactions indefinitely. Even if you're not applying for a new loan, they might use that data to market to you. Lucas: That's a valid concern. The rule does require that data be used only for the specific product the consumer applied for, unless they consent to broader use. But consent fatigue is real. Consumers often click 'agree' without reading the fine print. Luna: So education is key. The more consumers understand what they're sharing and why, the more they can make informed decisions. And the industry has an incentive to be transparent, because a scandal could set back adoption for years. Lucas: Absolutely. Look at what happened with robocalls and credit monitoring — once trust is broken, it's hard to rebuild. So far, open banking in the US has been relatively scandal-free, but the volume of data being shared is growing fast. Luna: What do you see as the next big frontier? Where does this go in the next two years? Lucas: I think we'll see open banking move from lending into insurance and wealth management. Imagine an auto insurer that can see your driving habits through a connected car API, or a robo-advisor that can see your full financial picture across accounts to give better advice. Luna: That's a logical extension. But it also raises the stakes on privacy and security. The more data flows, the bigger the target for hackers. Lucas: Right. And that's why the technical standards matter. The Financial Data Exchange — an industry consortium — has been working on a common API standard. If everyone uses the same security protocols, it reduces the attack surface. Luna: So to sum up: open banking is transforming consumer lending by replacing the credit score with real-time cash flow data. It's expanding access, speeding up decisions, and reducing defaults. But it's not without risks around privacy, data accuracy, and systemic concentration. Lucas: That's the story. And the interesting thing is, this is still early. We're only two years into Section 1033. The full impact on how credit is allocated in the economy is still unfolding.