Falling bank multiples, elevated risks: AI, credit, and the macro
The desk posits that banks are facing significant headwinds from private credit concerns and AI-driven competition, which are pressuring valuations. Per the full note from BofA Global Research, many regional banks are trading at pre-COVID price-to-earnings multiples, reflecting a market that is cautious about future earnings potential amidst these risks. The commentary highlights that while the macroeconomic outlook remains stable, any deterioration could further impact bank valuations. This context suggests that traders should remain vigilant regarding bank stock movements as they relate to FX positioning.
What the desk is arguing
BofA argues that private credit and AI-driven competition are re-emerging as key risks for the banking sector, weighing on valuations despite regulatory support. At the BofA Financials Conference, management teams flagged these concerns, but the desk notes that many large regionals already trade back to pre-COVID P/E levels, implying much of the bad news is priced in.
Supporting evidence includes the view that banks have defenses against AI disruption—through their existing customer relationships and potential efficiency gains—and that credit deterioration is not the base case for US economics. However, if macro conditions worsen, valuation multiples and earnings would face further pressure.
The desk implicitly rejects the idea that current valuations offer a compelling entry point without a clearer macro outlook. They see limited upside unless the macro backdrop improves, but they do not anticipate a severe downturn either.
Key takeaways
01Private credit and AI are renewed risk factors for banks, but many have defenses and AI could enhance efficiency.
02Large regional bank P/E multiples have fallen back to pre-COVID levels, suggesting valuations already reflect significant headwinds.
03A deteriorating macro backdrop is not the base case, but would pose additional risks to bank earnings and multiples.
Market implications
Bank valuations may remain range-bound near current levels unless macro conditions improve or the industry successfully navigates AI/private credit challenges. A negative macro surprise could trigger further multiple compression, particularly for regionals with higher loan exposure.
Risks to this view
Upside risk: banks successfully integrate AI to boost margins or a benign macro supports earnings. Downside risk: credit conditions worsen or AI disruption accelerates, pushing multiples lower.
Hello, and welcome to Global Research Unlocked, where we discuss what's rising from growth industries to rising risks and opportunities in global markets. I'm T.J. Thornton, Head of Product Marketing at BFA Global Research, and we're recording this episode on Monday, March 2nd, 2026.
When I look at where the regional banks trading at 10.5 times price to earnings, that's probably half to one-turn discount to where the large-cap regional banks were trading pre-pandemic. This is a group that was trading at a discount, came into 2026, where we've argued that the likelihood for regional banks to catch a bid was high on the back of the broadening out theme, that the economic growth broadens out beyond just investments in AI data centers and into a lot more on the back of tax incentives and the one big beautiful bill. There was a lot of optimism around bank stocks at the start of the year.
Overall sentiment in equities, according to our fund manager survey, was the most bullish since July 2021. Investors were long the boom and consistent with that idea. The same January showed that the overweight in banks relative to history was second only to commodities.
A shifting regulatory environment and M&A contributed to the positive views. But more recently, the mood on banks has changed. S&P banks are down mid-single digits on the year and have been beset by a range of concerns from private credit to the potential for widespread AI disruption to the conflict in the Middle East.
Today, we're talking to Ibrahim Poonwalla, head of North America Bank Research, who's been writing about all of this and who recently wrapped up the B of A financials conference in Miami, returning just in time for the blizzard. Thanks, Ibrahim, for joining us today. Thanks, TJ.
At the time of the financials conference just a few weeks ago, there were already various concerns swirling. AI disruption in software, the possibility of more private credit defaults and concerns about retail appetite for private credit going forward. What was the general take from bank managements around these concerns?
If you go back and the conference was held mid-February, what we heard from the banks was actually quite different to what we've seen play out in the markets over the last two months. What we heard from the banks generally was a relatively constructive start to the year when we think about just business momentum, think about loan growth for the regional banks, capital markets for the big Wall Street firms around M&A, IPO activity, debt issuance. All of that seemed to have started out quite strongly for 2026.
Then you contrast that with these concerns that you just mentioned. I think on AI, we heard during the conference and what we've heard consistently is banks view themselves as potential beneficiaries of AI technology and their ability to use AI to bring in more operational efficiencies that could probably help them on the cost side. If you think about a bank, there's a lot of duplicative processes, a lot of middle office, back office kind of work, onboarding clients, et cetera.
There's a lot of opportunities that the banks should have in becoming more efficient, doing more with less or adding capacity to their bankers. I think what we keep hearing from the banks generally is, yes, it's going to have some impact for sure around hiring plans, looking out where the banks see the real benefit to them outside of taking costs out on the operational side is adding more capacity to their bankers, to their financial advisors. If you're running a wealth management firm, being able to do more with your clients, being involved in more high touch, complex tasks, as opposed to managing a lot of mundane paperwork.
And on the private credit side, what we heard from the banks was just high confidence or just relatively constructive messaging on the credit quality side. As you mentioned, we had about 40 to 50 banks attend the conference. And I can't recall a single bank really flagging any systemic issues on the credit quality front that they were seeing.
Everyone's watching private credit, worried about contagion fears. You're not hearing anything from a fundamental perspective. Banks generally lose money when there are blind spots.
Private credit, as far as bank lending into private credit is concerned, doesn't quite feel like a blind spot given we've talked about this definitely for the last six months, but also for the last couple of years. Okay. Got it.
I know you had a lot of discussions with clients following those bank presentations. And since, do investors generally agree with the assessment from bank management? As you mentioned, the market doesn't necessarily agree, but maybe that's just a short term thing.
And then the other question is just putting this together. Banks had a good year in 2025, multiples for many of the highest quality names did expand several turns. What's your view from here on the group?
Great point, EJ. It's worth reminding of what the starting point was. Banks have outperformed the S&P, especially the largest banks, both in 2024 and 2025.
We came into the year with premium valuations, especially the largest banks, then a lot of optimism amongst investors. The question we were getting even on January 1st was, can banks outperform for a third year in a row? When you combine that premium valuation optimism, we've been hit with all sorts of headline risks over the last couple of months, be it AI disruption risk, the private credit contagion you mentioned, and some of the other policy announcements from DC at the very start of the year that injected volatility in the group and bank stocks.
When you think about investors, we've had several dozen conversations over the last two or three weeks on these topics. I don't think investors truly do not believe the banks in terms of what they're saying on the fundamentals, nor would I say that the investors have a strong view around AI disruption risk or a big clear sight on contagion risk from private credit hitting bank credit quality. Investor discussions can be summarized as not wanting to fight the prevailing narrative and acknowledging the fact that these stocks came into the year.
I don't want to say price for perfection, but discounting a fair amount of optimism. You had to recalibrate the stock valuations to account for some of these new, what I would still call tail risk type of events. They were very low probability on January 1st.
Maybe they're a bit higher probability today. Fundamentally, I think investors still feel like this is more headlines than changing what anyone had assumed in terms of the growth outlook for the banks in 2026 and maybe even beyond. But some period of digestion of the move that we've had over the last couple of years in and of itself is probably not an unhealthy thing.
Understood. And a quick follow up, because it's true that for a number of the biggest banks, we have seen multiple expansion, but I was also looking for the other banks too, and we haven't necessarily seen it across the board. Are we getting back toward interesting valuations that might account for some of these risks?
You're right. When I said premium valuations, I was mostly talking about the big six U.S. banks. Think about the money center banks, the Wall Street majors.
Those were the stocks that were trading at premiums. Even today, they're trading at 14 to 15 times price to earnings. They've been de-rated to about 12 to 13 times price to earnings for 2026 on tangible book, 2.1 times price to tangible book for a return profile that looks closer to high teens return on equity.
For the regional banks that were generally being discounted and remain discounted, when I look at where the regional banks trading at 10.5 times price to earnings, that's probably half to one turn discount to where the large cap regional banks were trading pre-pandemic. This is a group that was trading at a discount, came into 2026, where we've argued that the likelihood for regional banks to catch a bid was high on the back of the broadening out team, that the economic growth broadens out beyond just investments in AI data centers and into a lot more on the back of tax incentives and the one big beautiful bill, on the back of stable to lower interest rates, and just the deregulatory push from the administration economy-wide that seems to be flowing through. We did expect regional banks to benefit from a relatively healthy-looking yield curve and an interest rate backdrop, loan demand and middle market activity finally picking up, which is also good for a lot of the regional banks that have capital markets businesses.
Finally, having a period where you see a pickup in domestic APEX, when you think about S&P Equal Rate outperforming the S&P or the Russell 2000 outperforming the S&P to start out the year, we saw that in my world as well, where regional banks did outperform the GSIB banks. When you think about the valuations, I would say, given what has played out over the last two months, screen discounted relative to the growth outlook. What I like to say is, yes, the group is discounted relative to the 13 to 15 percent earnings growth that we have for this year and next year.
The group is not discounted if we are about to enter some sort of economic downturn or a material increase in job losses or a material slowdown in M&A activity. Any of these could cause a negative earnings revision cycle, and that's not quite discounted here. I do think there's a big gap between the discounted today relative to the growth outlook, but they're not discounted today or they're not discounting a high probability of a much worse macroeconomic outlook over at least the next 10 months.
Got it. That makes sense. That view is consistent.
I pointed to the fund manager survey at the start. Another one of the questions that is asked in the fund manager survey is people's views on hard landing, no landing, soft landing. The no landing view is high relative to recent history.
People are not really looking for a big downturn. Of course, there have been concerns recently, but that is consistent with your view about the stock. Certainly not pricing that in.
Another question on private credit. It's a bit ironic because over the last few years, the banks were seen to be losing some lending share due to the proliferation of private credit. Now, people are concerned maybe a bit about private credit, about flows into private credit and whether they can sustain the flows, especially from retail investors.
Is there a way in which slower inflows or slower private credit lending could be a positive for the banks from a competitive standpoint, or is it negative just because of what it may mean for credit? Great question, DJ. We've thought about that too.
I would break it down into two parts. One, the ability of the banks to lean in today, they are all well capitalized. Liquidity is no longer a concern.
Yes, I think when I talk to any of our banks, large banks, regional banks, they are looking to lean in and make new loans. To the extent these loans meet the underwriting standards for these institutions, think about credit quality, et cetera. I think banks would very happily take some of these loans on their balance sheet.
What I would call out is the one difference between private credit, there are many differences, but the one difference is private credit firms are very happy with a lending-only relationship where all they're doing is making a loan. You contrast that with the banks that we cover. Every bank I speak to, doesn't matter large or small, they are a lot more focused on the holistic nature of the client relationship.
If they are making a loan to a client or a borrower, their aim is to bring in cross-sell banking services, hedging, capital markets, bringing in deposits, doing treasury and cash management for these customers. The second piece, which I would call is bank lending on commercial side, is a lot closer to investment grade, whilst there are aspects to private credit that are closer to the high yield bond market. You heard other fixed-income and portfolio managers talk about the parts of the high yield market that probably gravitated towards the private credit and sit within private credit today.
That piece, I don't think is coming onto the balance sheet. For the banks, one, I think meeting the hurdles on credit quality, leverage levels, et cetera, and then having an opportunity to convert that into a wholesome client relationship will be two sort of governors on what kind of lending can come back to the bank balance sheets or where the banks would want to compete more heavily. Then obviously anything that feels a little more lower quality, maybe stays in private credit or maybe even moves into the debt markets.
Before we get into the broader discussion about AI and what it can mean for jobs and thus banks, there also has been interest from investors in identifying stocks that may be less at risk from being disintermediated by AI. AI is basically other than software, a lot of hard assets. You've mentioned that regional banks fit that description, so why is that?
I wouldn't limit this just to regional banks. I think what I would say is when you think about regulatory compliance, be it with banking regulations, be it with anti-money laundering, KYC, all of that where the banks are doing a lot to meet with the standards that the banking regulators have put out or the treasuries put out, I think that is one layer of protection which is harder to disrupt by writing up a smart model based on an AI native platform using cloud co-op, for example. I think regulations create a bit of a barrier when you think about disruption risk to the banks.
Balance sheets is another thing. You cannot AI away the balance sheet. If someone needs a loan, they need a loan.
If you think about the good and the bad about banks is they are balance sheet heavy entities and you eventually need to keep your deposit someplace and you need someplace to borrow and banks do both of those. There is that aspect to it that is harder to disrupt as well to the extent we can tell. Then the final piece is the customer relationship aspect where it doesn't matter if it's corporate clients or if you're a financial advisor who's managing wealth assets for high net worth or ultra high net worth individuals, there is a client relationship aspect that also creates a moat for banking institutions.
The way we've thought about all this is the greater risk to banks is from incumbents, regulated financials or banks that are smarter and are more effectively using AI as opposed to banks getting disrupted because of AI or a tech native AI platform. What you also said is there could be aspects of a banking business where there are 100 things that a bank does, 75 things banks could emerge as winners. There could be 25 things that banks lose where maybe there are some legacy revenue streams or margins in terms of the fees that they get paid on wealth management assets or some of the lower cost deposits get managed a lot more efficiently.
But overall, when we think about what I mentioned earlier, the ability to take out costs, inject a lot of efficiency, and then regulations balance sheets, the human relationship aspect that does create better defensibility than at least what we've seen play out in the likes of software companies over the last month. The question about the broader impact from AI on jobs. What do you think about that?
Ultimately, employment is closely tied to the credit cycle. So I certainly understand why that would be an issue for the banks, because it would probably mean worsening consumer credit. We worry a lot about jobs, and I think we're already worried about the job market given some of the softness we saw towards the end of the last year.
When I think about the disruption or the downside risks to bank earnings and bank stocks because of job losses, it's very clear. We've all talked about the K-shaped economy for the last two or three years, if not longer, and the lower-income consumer being under pressure. What we worry about is if you do see a wider broadening in terms of the unemployment rate, lots of white-collar jobs being lost because they're being displaced by AI, you worry about all things consumer credit, credit card loans, auto loans.
And at that point, it doesn't really matter if you're a high FICO customer or low FICO, if someone loses their jobs, quite likely they could be late on their bills, et cetera. You could see more defaults. We worry about that as a direct impact of higher unemployment.
Then if that's where things are headed, again, U.S. economy, 70% of that is consumption-driven. If the consumer is weak, then it is going to, at some point, start weighing on a lot of businesses, especially when you think about middle markets, small businesses. We could see business bankruptcies go up on the back of that.
Anything that leads to a sharp and a sudden spike in unemployment in the United States, I think would be a negative factor for the outlook for bank stocks. At the start, you talked about what bank managements were saying about AI and how it could drive efficiencies. As a group, do you think there's a risk that, sure, these companies find ways to drive efficiencies through AI, but ultimately that just gets competed away and it's not much of a sustainable margin driver?
Absolutely, that could happen. It was one of our questions in our year-ahead notes for the top 10 questions to be published in the first week of January. Will bank stocks become a play on AI?
I think part of it is, will some of these benefits that I talked about earlier actually show up in growth and ROEs? That's debatable. As someone who's followed the industry for 20 plus years, I'm inclined to believe that some of the efficiencies, much like we've seen with other technology-driven productivity boosts that the industry has received, does get competed away.
It's not surprising at all. Banking is highly competitive, especially in a market like the United States, where you have around three and a half to 4,000 banks on both sides of the balance sheet, competition for loans, deposits, extremely intense. It is quite likely that some of the productivity boosts that we talk about become table stakes.
As a result, you need to do that to be competitive, but it may or may not actually fall to the bottom line in terms of an added boost to the profitability of the sector. Shifting gears, way back, like last fall, stablecoins were all the rage and there were fears about what they could do to banks and credit card interchange. There have been some regulatory developments since, even though that issue has moved to the back burner as other things have moved forward.
Where do we stand on stablecoins now, and do you think they're more of an opportunity or a threat for the banks? It's interesting. We just wrapped up our call on digital assets and stablecoins, and you're right when you think about stablecoins and more broadly digital assets, there's still a lingering question about exactly what are we trying to solve for that cannot be solved using existing systems.
In a world where we assume that blockchain technology is far superior and efficient than anything that's out there for the banks today, it's a bit more of a balanced view, as opposed to the fear that surrounded the sector a year ago, where you were worried about what stablecoins would mean. With all things regulations, what you're seeing out with stablecoins, the Genius Act was signed by the president into law last July. And you've seen the banking regulator, the OCC, the FDIC propose rules as mandated by the law.
You're going to have a very structured, regulated framework on what is allowed or not allowed for payment stablecoin issuers, from whether or not they can pay a yield or rewards on those stablecoins, or what these stablecoin issuers need to hold at the other side as collateral to comply with the Genius Act. We've seen several banks partner with some of the crypto firms to provide these solutions to their customers. When we also think more broadly about just tokenization of assets, think about money market funds, stocks, bonds, we are all waiting for the market infrastructure bill and watching whether or not that passes Congress over the coming months.
We have a world where the SEC, the CFTC are leaning in and want the growth of the digital asset ecosystem. In that world, I do think the big legacy financial institutions, be it asset managers or be it the investment banks, are all relatively well prepared and leaning in to adopt that technology with the view that if clients move that way, they want to be able to provide that solution. Now, do stablecoins or digital assets or blockchain technology create some risks to legacy revenue streams?
Again, the answer is yes. What does it mean? Maybe you get paid less for every fixed income bond that is traded on the desk, on Wall Street trading desk.
That's not a new phenomenon for the banks. What you would expect is at the back end, systems becoming more efficient would allow banks to provide the same service in a lot more efficient, low-cost manner. The conversation with experts, with the industry is a lot more balanced today when we think about stablecoins or all kinds of digital assets compared to a year ago.
Okay, and Yibi, last question. Another reason for the optimism to start the year was M&A and capital markets. Of course, a lot of that's going to come down to the economy.
What has happened in capital markets and M&A activity over the last two-plus months relative to expectations coming into the year? Expectations were sky-high coming into the year that we would see a record year for M&A activity after three or four years under the prior administration, under President Biden, where the Department of Justice, the backdrop was generally not conducive for dealmaking. And on the IPO activity side, what we are seeing in terms of the data and selectively where the industry has provided updates, we are still tracking about 15 to 20 percent up year-over-year in terms of overall activity, appreciating that we still have a full month of March to go as far as the first quarter is concerned.
At least, I would say until a week ago, this was something that we repeatedly heard during our financials conference in mid-February as well, was there's a lot of desire among strategics to follow through and get deals done, given policy backdrop, relatively constructive financing is available. On the sponsor side, when you think about private equity firms that have talked about monetizing investments for several years, and you saw that activity pick up in the back half of the year, from industry estimates, there's about three to four trillion dollars of assets that need to be monetized either through M&A or IPOs are tied to the private equity industry. When you think those two aspects, when you look at all the market volatility, that's actually fundamentally still holding in place as far as both trading and investment banking is concerned.
If you could have a couple of bad weeks or a more broad-based selloff in the S&P, I think much like we saw post-liberation day, there's absolutely a risk of activity getting pushed out or delayed, leading to some disappointment relative to what was fairly optimistic expectations across the street, including ourselves. Okay. EB, thanks for joining us.
That was a great discussion and going through my questions here just reminds me of how many issues the banks have had to deal with over the last few years. And I'm sure there's going to be a few more in the next few months with the pace at which these things are moving. Thanks, TJ.
Yep. I've done this for 20 years and it's never been a dull year for bank analysts. So thank you.
Though we've had two strong years of performance for the banks, the recent selloff coupled with continued earnings growth has meant multiples are off highs for the banks. EB mentioned that regionals are trading around 10.5 times earnings, a discount to pre-pandemic. Banks are fairly well insulated from AI threats given regulations and the need for a balance sheet, even though there are aspects to what banks do that can be impacted by AI.
Offsetting that are also greater efficiencies, the many back and middle office processes in banks that could be streamlined. We've seen credit issues come and go, but the risk is something that actually impacts the economy more broadly. This could be AI job disruption, which would likely create a vicious cycle of defaults and lower capital markets activity.
Our economists are not looking for this, pointing out the many revolutions that ultimately resulted in more jobs and more income from the industrial to internet revolutions. An interesting stat from an economics note last week is that 40% of the U.S. was once employed in agriculture. That's just 1% today.
As one sector saw its share shrink, many others took over. Thanks for joining. Bank of America and B of A Securities are the marketing names for the global banking businesses and global markets businesses, which includes B of A Global Research of Bank of America Corporation.