RegFi Episode 40: Financial Innovation and Supervision

Orrick RegFi Podcast | Financial Innovation and Supervision

RegFi Episode 40: Financial Innovation and Supervision
24 min listen

RegFi co-hosts Jerry Buckley and Caroline Stapleton welcome podcast guest Tracy Basinger, former head of supervision at the Federal Reserve Bank of San Francisco and a senior advisor at the Klaros Group. Tracy brings a unique perspective on financial innovation to the conversation, informed by 30 years of experience in the Federal Reserve System, where she led the formation of the FSR's first fintech team and established initiatives to leverage new technologies for bank supervision.

The group explores regulatory developments affecting fintech and banking as a service (BaaS), the revolutionary potential of artificial intelligence in the financial services sector, and innovative supervisory approaches, such as real-time data monitoring and supervisory dashboards, to keep pace with the rapidly evolving financial technologies.

Links:

Episode Transcript

Jerry Buckley:	Hello, this is Jerry Buckley, and I am here with RegFi co-host Caroline Stapleton. Our guest today is Tracy Basinger. Tracy is a senior advisor with the Klaros Group and a board member of AIR, the Alliance for Innovative Regulation. Prior to joining Klaros, Tracy served as head of supervision at the Federal Reserve Bank of San Francisco from 2017 to 2021. Tracy formed the Fed's first fintech team in 2015 and led the Federal Reserve's initiative to study developments in fintech. She served as the Fed's representative on the Basel Task Force on Financial Technology, and she established and sponsored the Federal Reserve System's suptech initiative to leverage technologies to conduct bank supervision. And given that your experience is so relevant to our focus on the RegFi podcast series on how technology will transform financial regulation, we're really looking forward to this conversation. Everyone knows the role the Federal Reserve plays in setting monetary policy and its enormous impact on the economy. But our listeners may not be as aware of the role the Federal Reserve Banks play in supervising banking, both at the bank level and at the holding company level. Could you describe for our listeners the role the Federal Reserve Banks play in bank supervision? And, perhaps, Based on your experience, could you share your thoughts about the way in which the Board of Governors interact with the Federal Reserve staff? How much is decided in Washington? And how does this impact the way in which Federal Reserve Banks examine the banks that are under their jurisdiction, Tracy?
Tracy Basinger:	Thanks, Jerry. Thanks for having me. I'm glad to be here today. So the Fed's role in supervision: First, who does the Fed supervise? So the Fed supervises state-chartered banks that elect to become members of the Federal Reserve System. So they supervise all state member banks. They supervise all bank holding companies, regardless of who supervises the banking entity. So even banks supervised by the OCC and the FDIC, the Fed would supervise the holding company. And then they have a large portfolio of foreign banking organizations that they supervise as well. And typically, how the Fed goes about it is the entities are separated into four different portfolios. Community banking organizations, which are those with $10 billion or less in assets. Regional banking organizations are those between $10 and $100 billion in assets. Large and foreign banking organizations are in a portfolio, and those are institutions with $100 billion or more in assets for the large institutions. And then their fourth portfolio is called the LISCC portfolio, which is the Large Institution Supervision Coordinating Committee portfolio that was formed after the financial crisis, and that is essentially the largest banks in the country, those globally systemically important institutions. And how they go about that is typically, again, for the smaller institutions, it's what they call point-in-time supervision, which is every 12 to 18 months, a team of examiners will come in and conduct an examination. And then as you advance through the portfolios, they become more what's called ongoing supervisions, where it's a series of targets. It could be horizontal reviews on a particular subject across the institutions in that portfolio that they usually roll up on an annual basis to determine the actual ratings. And then there's what I would call local programs, which community banking is usually done at the Reserve Bank at the local level. And you work your way up to the LISCC, which is a national supervision program where you have people that are employed by both the Reserve Banks on the Board that work in this national program together. So it varies and it changes quite a bit as the size of the banks increases. And then in terms of how decisions are made and what gets done at the Reserve Bank and what gets done in the Board — I mean, it's changed significantly over the course of my time at the Fed, and I'm sure it's going to continue to evolve. I guess I'd start with Congress gave the authority to supervise banks to the Federal Reserve Board, and then the Board chooses how they want to delegate that authority to the Reserve Banks. And so, as I mentioned, it's changed over the years, and there's, I think, a host of reasons. But I would characterize it that supervision today is much more of a team sport with Reserve Banks and Board staff working together to supervise banks, especially for the larger institutions. And there also are certain decisions that are made by either the Vice Chair of Supervision, which is a relatively new position that came out of the financial crisis, or the Board of Governors. Things like certain enforcement actions, certain decisions on the largest institutions in the country, stress testing decisions, those all get escalated higher, where perhaps an informal action at a community bank would be something that would be something that a Reserve Bank could handle with less involvement of Board staff. So it varies over time. And I would say as a typical rule, the community bank portfolio is handled more at the Reserve Bank level. And then as you go through the portfolios, there's much more involvement from the Board or staff in terms of setting not only what the scope of activities are going to be, but how they're executed.
Jerry:	And you say it's changed over your time. Has it become more focused in Washington over time, especially with the addition of a Vice Chair for Supervision?
Tracy:	Yeah, I think there's been much more involvement in Washington and much more engagement. And I think some of it is positions like the Vice Chair of Supervision. I think some of it is technology, access to information, access to data. It's a lot easier to communicate. And so that's caused this to be more of a, if you will, a team sport. I think there's also, at each of the various crises, you know, it's the main office, if you will, of all the agencies that is testifying before Congress. And that, you know, there's a view that having more engagement, more involvement, more authority and decision-making in Washington is probably a better thing for supervision overall.
Jerry:	Well, Caroline, let's get you into this conversation.
Caroline Stapleton:	Tracy, it's so great to have you. Thanks so much for being here. You mentioned technology changing the Federal Reserve supervision. It's, of course, also changing the banking industry more broadly. And we do a lot of work for clients that are entering the fintech space, they need advice on licensing, compliance, both banks and fintechs. One of the areas that is getting a lot of attention from regulators recently is banking as a service. I think what a lot of us colloquially refer to as BaaS has become, in many ways, a preferred route for fintech companies to get access to the payment system, lending, and to distribute their products in partnership with chartered banking institutions. So, I was hoping you could help our listeners understand how BaaS works at a high level, of course, and some of the benefits of these arrangements and the challenges associated with this model, including from the regulatory perspective in particular, just given your experience.
Tracy:	You know, it's interesting, you mentioned, you know, BaaS sort of growing and becoming more commonplace. And one of the things I've learned is that you have to be really careful using that term because different people use it with a different meaning. And so maybe for purposes of this discussion, I think about it as banks having relationships with fintech providers to provide financial services to end users with or without a middleware provider. So it's sort of that ecosystem of players. My experience is how this actually works probably varies based on the players, but that more generally speaking, I see banks contracting with fintech companies to allow them to use the bank's platforms and infrastructure to offer financial products and services. And some of this banks do via direct connect with the fintechs, while others use third-party middleware providers to connect the bank and the fintech. And generally speaking, you see community banks getting into this, and I'm sympathetic to their reasons as they're looking as sort of a way to innovate and a way to broaden their revenue streams. And they're doing this, I think, at a time when there's really relatively little regulatory guidance on these arrangements other than third-party risk management guidance. And as I'll sort of maybe get to, I think maybe one of the mistakes we've made in this place is focusing too exclusively on third-party risk management. Because we've seen a rash of recent enforcement actions against banks in this space, and they're increasingly, in my view, more far-reaching in their requirements for banks and certainly broader than simply addressing third-party risk management. The ones that I've looked at pretty closely include a lot of issues with BSA/AML violations or program deficiencies, concerns with the level of oversight and scrutiny by the board and senior management with regard to the BaaS activities, and most of them have some type of requirement to stop onboarding new fintech partners without regulatory approval, which can have the effect of essentially shutting down that business if they're not able to resolve their issues relatively quickly. The most recent consent order in this space is just the latest example of how difficult it is for banks and fintechs to manage these relationships, not only meeting the regulatory requirements, but also managing the operational complexity that they present. The regulatory scrutiny here has been increasing, and suffice it to say, it's likely to intensify significantly. I think regulators are realizing that these relationships are not just about third-party risk management, but also as much about operational risk management. And to this point, an order recently went as far as to impose capital planning provisions as it relates to the BaaS business. And my sense is that banks are starting to realize that they need to have a lot more oversight of not only the fintech, but of the data that comprises their books and records. And responsible fintechs are starting to realize that having a partner bank with strong controls is actually to their advantage in the long run, even if it does slow them down a bit in the short run. I don't know, I've heard a lot of people make comments to the effect that this is going to represent the end of fintech tourism. And I do think there is some separating of the wheat from the chaff among partner banks, middleware providers, and fintechs as we sort of see all of this play out.
Caroline:	So do you still think there will be a benefit to the BaaS model, you know, in light of the recent enforcement activity that you've referenced? Are there places to grow this model? Do you see some benefits to it going forward?
Tracy:	I do. I mean, I think that while we have seen a number of actions, we certainly still have a number of partner banks out there that have figured out how to do this well and how to do it right. And I think the ones that got into this space, you know, it's a huge investment to put the right kind of infrastructure in place to be able to do this well. And those that are willing to make that investment and continue to invest in the program as it grows, and those fintechs that are serious about being with the responsible partner, which I think so many got caught flat-footed with some of these banks that had to off-board them that I think you've got the players that remain are ones that want to do this right. And I think that's where the space will be. I think we're also going to likely have to see some sort of regulatory guidance come out at some point in the relatively near future, which I think will help people do it right or decide whether or not it's the right place for them.
Caroline:	So, on the theme of technology, another topic we're hearing a lot about at Orrick is artificial intelligence, AI. We have been experiencing a lot of requests from companies seeking to develop their strategies for implementing AI, in particular with respect to generative AI, just because it's so novel. I mean, in some ways, predictive AI has been around in one form or another for a lot of financial institutions for decades. But generative AI is really changing the game. And so we're even seeing our major tech clients — some of them in the forefront of manufacturing like the chips and the IP, powering generative AI — they themselves are looking for help in setting up their internal policies related to use of AI in compliance with law and evolving public policy. And then, of course, yeah, our financial services clients are wrestling with these same challenges. So, you took the lead in promoting fintech as a regulator. How would you advise banks and other financial services entities to proceed, particularly when it comes to using AI and interacting with customers and helping customers understand things like how their credit profiles impact their access to credit? What are your thoughts on that?
Tracy:	Yeah. I completely agree that generative AI is a game changer. I mean, I think the discussion of should I use AI or should I not use AI isn't the right question anymore because it's here. I mean, you do a Google search today, and a lot of times the first hit you get is here's the generative AI response.
Caroline:	Completely. I often say that I think that my children, who are eight and four, will one day look at me when they're in college and say, "You're telling me you started with a blank screen, and you typed every word?" And then I'll say, "Yes," and they'll say, "That must have taken forever." It did. So I agree. It's a complete game changer.
Jerry:	Sadly, those of us who learned to write in cursive were even before that, and that's a lost art as well.
Tracy:	We could go down a whole rabbit hole here. But to that point, I mean, I think the conversation now has to shift to how do we use it responsibly? How do we put the right controls in place? What are the parameters that we're going to have in place for this new technology? And I think, to your point, banks have been using AI in many ways for years, and they're going to continue to increase its usage. Most common that I see customer service, fraud detection, training, risk management. But my experience is that banks have been really hesitant to use any kind of AI in the credit decisioning because they're just not sure how the regulators are going to react to it. And the regulators, for good reason, are poised more to point out the risks than consider the benefits. And that's kind of the tenor in which the interagency guidance on the use of AI was issued. And that's so outdated at this point because things have moved so fast. But I also believe that in many cases, the regulators hadn't yet gotten, even if they have now, didn't fully understand the technology because they hadn't gotten their hands dirty in it. And so that's another reason that I think makes the guidance a little dated. And both the OCC and the Fed have formed novel supervision groups. So I'm hopeful that that's changing in terms of how there can be better dialog between the industry and the regulators on how to use AI responsibly. But if I were advising somebody today, it would be to proceed with caution when it comes to using it in ways that impact decisions about whether someone gets credit, what the cost of that credit might be, how to use it in collections, how to use it in personalized marketing that steers customers in a particular direction. And I would proceed with caution until we're at a point where we understand the data that the AI tools are using and until people are able to explain the results. I mean, to your point, I don't think we're at a point where we can explain the results of credit decisions, how credit decisions are made using AI, and then how it impacts the consumer's credit profile. And those would be two key things that would need to happen before I think it would be safe to use it in the credit space. And this is my personal view. Bias data in means bias data out. And to me, the data is the real risk here. And I believe that based on the history of credit allocation in this country, it's going to be a challenge to get to a point where you have good unbiased data that you can use for credit decisioning. And so having said that, though, I do think we should continue to test and innovate here so that we can ultimately improve on credit allocation. And I would also love it if from a regulatory perspective, we can get to the place where if we do have a tool that results in credit allocation using AI, that while not perfect, perhaps, but is better than what we have today, but that could be considered acceptable because I don't think it is right now. And the idea of using AI to make baby steps and improving the credit decisioning would be a great outcome in my mind.
Jerry:	Well, I think we're all trying to figure that out. And I think your advice is right: proceed with caution, but proceed. So, you know, in a prior podcast episode, we mentioned that the pro bono effort that our firm undertook with the Alliance for Innovation and Regulation, where you serve on the board. And we, along with Jo Ann Barefoot, who leads that organization, we interviewed the heads of innovation at each of the financial regulatory agencies at the time. The report identified legal, budgetary, personnel, hiring, and cultural hurdles that regulators face in implementing technology to enhance supervision. I've always thought that the Federal Reserve Banks, with their unique position in the bank supervision system, could be the ideal leaders in experimenting with rolling out supervisory technology because the Federal Reserve Banks are free from some of the constraints that the other regulators labor under. I can even imagine the Federal Reserve System allocating resources to establish a research arm like the Defense Department's DARPA, or Defense Advanced Research Projects Agency, that leads the way in innovation and national defense. Something like that in financial services led by the Fed and using AI and the other technologies that are being developed would be fascinating to me. So if I were a governor, I maybe would be making that an initiative. But things are moving quickly on the tech front, and the impact on banks and other financial players, as we've referenced, is going to be dramatic. You were among the first regulators to see this, Tracy. I'd be interested in your thoughts on what the Fed could do to harness technology to make its supervisory function up to the challenges that lie ahead.
Tracy:	Yeah, I very much recall talking to you on that project that you did for AIR. And you may recall when we spoke that at the San Francisco Fed, we were exploring how we might be able to engage in some of this experimentation at the Reserve Bank level since we weren't subject to all those same restrictions. And you mentioned earlier that I sponsored the Fed's first fintech initiative. And, unfortunately, I left the Fed just as the suptech initiative was restarting because we had to pause it for a time during COVID. So, my insights on where they are is a bit dated at this point, but I talked about the novel supervision group that was formed, and I think there is some more of this intent to experiment with that. But I couldn't agree more that experimenting with technology through the lens of conducting supervision is necessary and that the Reserve Banks have more flexibility to do this and more flexibility to engage with third parties and have the kinds of activities and events that you discussed. And I do know that since I left the Fed, they hired a Chief Innovation Officer that, as I understand it, is looking at how the Federal Reserve can leverage many new technologies more broadly. I really do think that leaning into experimentation and showing a willingness to test and adopt new monitoring techniques is a huge important step here. I mean, anytime you can get real-time access to data and use tools like AI to monitor that, to inform your supervisory activities, I think that's where we're going to have to go, because institutions are going to continue to move. The institutions that are supervised are going to use these technologies. They're going to be moving that quickly, and supervision is going to have to evolve in that way. And maybe since we've both mentioned it as a shameless plug for AIR, I would say that engaging with AIR and organizations like that can be really helpful for U.S. regulators as they're on this journey to sort of understand or to determine how best to use these technologies in their supervisory programs.
Jerry:	You know, we discussed the idea that Nick Cook advanced at the Financial Conduct Authority. Of course, Nick Cook works at AIR now. But when he was head of innovation at the Financial Conduct Authority, he did advance the idea of a dashboard that ingested in real time the information from financial services firms so that there could be a flashing amber when something showed up that was identified as a problem. The dashboard isn't going to control regulation, but it's going to empower a regulator. And I have been fascinated by that idea. Your thoughts?
Tracy:	I think it's a great idea. I mean, I have always been a great admirer of how the Financial Conduct Authority, in particular, has just leaned in on innovation. And they've shown a willingness to be at the forefront of learning about it, getting their hands dirty with it, and a willingness to take steps forward to do things like develop the dashboard that you suggested. So I think that's one of the things that puts them at the forefront of most of the conversations about innovation and regulatory authorities, that and probably the Monetary Authority of Singapore. And I get the differences in the mandates and banking landscapes between the U.K. and the U.S. that make it maybe a little more challenging here. But I fear in some ways that the U.S. is falling behind, and it's going to be a little difficult to play catch-up here, because, like I mentioned, as institutions continue to use these technologies in all aspects of their business, or at least many aspects of them, it's not an option for regulators not to do the same. And they're going to need something similar to or like this dashboard in order to just be able to understand the institutions that they supervise and to be able to monitor them in an effective way. So, I think that's the way to go. And since you mentioned Nick, I mean, I think it was when he was at the FCA, he was famously quoted when talking about innovation and the need to adopt these technologies that, “We should be aware that standing still is effectively running backwards.” And so that's what we need to caution ourselves about.
Jerry:	Nick, he was a podcast guest prior to this, and I always enjoy the conversations with him. He is so insightful, and AIR is so lucky to have him working with AIR. Well, Tracy, I think our time is running out, but I want to thank you for being willing to do this podcast, for your insights. It's always a pleasure to have a conversation with you. And I hope that we all, the alumni of the Fed and others, can urge the Fed to move in the direction that we're suggesting, which is develop a really strong leadership role in the use of technology and financial regulation. Thank you.
Tracy:	Absolutely. Thank you, Jerry. Thank you, Caroline.
Caroline:	Thanks, Tracy.