American Express Company ($AXP)

All right. Let's get started. Good morning, everyone. Thanks for joining. My name is Rob Wildhack. I cover the consumer finance group here at Autonomous. We're very excited to have Steve Squeri with us today. Steve, of course, needs no introduction. Even so, he's the Chairman and CEO of American Express. Those are roles he's held since 2018. And he's a 40-year plus veteran of the company.

41 years.

Great to have you back with us here, Steve.

Good to be here.

Just a quick note for the audience, we'll be using pigeon hole for the Q&A again, submit the questions there. I can see them up here. You can vote on them and then I can get them over to Steve. And now we can start. Steve, in your annual letter, you wrote a lot about AI and Agentic commerce, which we'll get into. But you also led off with a reference to the company's framework for winning, which is a term that I wasn't overly familiar with. So can you describe the framework, what is it? And how does that guide the strategy you're currently rolling out right now?

Yes. So when I took over in 2018, we were coming through some tumultuous times. We had -- we had lost Costco, prior to that, we came out of the financial crisis. And it always appeared to me having been there that long, that the company wasn't as focused as it needs to be, number one; and number two -- it was about catching up. It wasn't about winning. And so I spent pretty much a year almost coming up with this 1 page document. I called it my own manifesto. And basically, what it was, was to focus the animation on really what was important from a strategic perspective, how we were going to win, what we look for leaders, what the values were going to be and that has served us really well because I believe that people need a reason for being. And when you try to motivate an organization of 80,000 people -- what they need to know is why they come to work every single day and what they need to do. And you can't have it across a host of things. So we actually called it the place mat. And it's very simplistic, right? We wanted to focus on a premium consumer segment. We wanted to build on our leadership position from a commercial perspective. We really wanted to build our global integrated network and what's -- the other thing that we want to do is we wanted to take advantage of our globality. And if you think about American Express, American Express, like Visa and Mastercard is a global network. But unlike a lot of the issues we compete with, we provide services, and we provide cards in lots and lots of other countries, and we're a merchant acquirer in other countries. And so that globalness helps us compete. It helps us learn. It helps us leverage more. So what we really wanted to do there was to make sure that the organization was leaning into that. And we just added a fifth strategic objective about 6 or 7 months ago, what we really wanted to do was reimagine both the customer and the colleague experience. And we just was just announced today, we just won the J.D. Power Award for the best mobile app in the card industry, again, and we've been a J.D. Power leader for years but I think it was really important as technology continues to change, that we took more of a leadership position to think about not only how our customers are going to interact with us but how the -- our colleagues are going to interact with one another and how do we make it easier for them to get things done in today's environment. So it's good to test the time for us. We modify it as need be. And it allows us to focus on those things that I think are clearly important. And as you communicate throughout the organization, it gives you that framework for doing it.

Very good. Going from zooming way out to zooming way in, MX is going to get a nice windfall from the sale of [ Dynex ] GBT, 30% owner. I think it works out to a $975 million pretax gain. In the past, you've reinvested these gains. Sometimes you return them like with the certify. And it sounds like there's always a long list of potential investments at Amex that go unfunded. So walk us through the decision process, how you think about whether or not you reinvest something like this, give it back to the shareholders or the bottom line and how you're thinking about this one.

Yes. I think the first thing is that this is not the consumer travel business. This is a business we spun off about 10 years ago. It was business travel business. So the consumer travel business very important to the value proposition, no intention of spinning that off. We own it 100%. It is part of the overall value proposition of our card business. As we look at gains like this. Traditionally, what we've done is we've taken advantage of investment opportunities, whether they be acquiring more cards, some more technology investment. We look at efficiencies. You've seen us take restructuring charges and things like that. And then we pretty much always return some to the shareholders. In the certified situation, we had some really good momentum, and it was going to be really hard for us to effectively ramp up, whether it was those card acquisition opportunities or more technology spend. We thought we had ramped it up enough at that point. And so we decided that given the business momentum that we had, we were going to just drop the whole thing to the bottom line, which made our shareholders very, very happy. And when we look at this gain, and we talked about this a little bit in the first quarter. We over delivered in the first quarter according to the pundits out there. But we had decided we were going to make even more investments in technologies and card acquisitions. And a lot of times we say, we're making investments in marketing, which sort of freaks people out because they think marketing, they think TV, they think maybe a sponsorship, when we say most of our marketing investments are actually card acquisition investments, okay? And so that's where we invested. So as we look and as we think about where we are in the cycle from an investment perspective this year, we'll probably do some more in technology. We'll ramp that up. We'll probably do a little bit more in card acquisition. We'll look for some efficiencies. The other thing I'd like to point out is that while we call out something like GBT here, 975, 945, whatever it to wind up being, every quarter, there are ins and outs. I mean in the second quarter, for example, we'll have a small gain, we're going to reinvest that back into the business. There are other quarters where either from a mark-to-market perspective, I just think back to the -- to 2023 with the [indiscernible] the valuation, when we looked at that, we absorbed that. And whether we absorb something or whether we decide to reinvest it, we don't look for a we make that part of our base case and off we go. So I think that we'll always look to invest. We have lots of great investment opportunities. And it depends on the timing and can we gear up the machine in time to take advantage of those. The other point that I want to make is that GBT gain was -- is not factored into our guidance in any way. So that's separate from the guidance that we provided.

Very good. Can you speak to what you're seeing with respect to the health of the consumer and is there any update you want to give us on spending as you see it through April?

Yes. I think through the conference, I think you've had Visa, Mastercard a bunch of other people here speaking. And I think it's relatively consistent, right? When we look at quarter-to-date in the second quarter, we're slightly ahead of where we were in the first quarter. And when we look at it, I think it's you look at credit, you look at card acquisition, especially in the premium space, we look at it's the same. I mean we haven't had -- our delinquency 9 out of the last 10 quarters has been like 1.3%. The other quarter it was 1.2%. So we've been pretty steady from a delinquency perspective, same thing from a write-off perspective. and card acquisition is still there, especially premium card acquisition. The other thing I think that was concerning to some people in the first quarter was airline spending and what we were seeing there. And I can tell you, in April, we had 9% airline growth. So people are on planes, they're flying. We had record travel bookings in the first quarter. And so we're seeing a healthy consumer. We're seeing a resilient consumer and that's been really consistent.

Excellent. Let's talk about the premium card category. You've characterized competition as intense there. I wanted to ask about that topic maybe in a little bit of a different way. The last real spike in competitive intensity was in 2016 when Chase rolled out the Sapphire Reserve card. And given where you are with the platinum today, I think it's fair to say that MX has carried that threat successfully. So what were the key learnings from that experience? And are they still part of your playbook or toolkit today. If a new entrant came to the space with a big bag of rewards and sign up a how would Amex respond to that?

Yes. So I think you've got to look at the -- the premium category has grown faster than any other category in the credit card space. And I think the TAM of this continues to grow because I think with all the competition, and I would say it's -- you referred to 10 years ago, I think it's as intense now or more intense than it was then. But what happened between whether it's Chase, whether it's city, whether it's Capital One, whoever it might be coming in, they've put a spotlight on the premium space. And what has happened here is that creditworthy consumers have now realized that premium cards can be for them as well. And so I think you've seen a TAM expansion. And so when you look at overall credit card growth -- it's growing -- I mean one thing great about this industry is that people will always spend money and it will continue. But from a premium perspective, that piece of the pie has grown even more, okay? And so when you look at this, and I get asked this question a lot in terms of, well, jeez, how did you react to chase back then and you had such a fast response -- we're not that good and we weren't that fast, right? And so what happened was our thinking on this premium space has continued to evolve and will continue to evolve. And when you think about it, you look at all the various cards out there. If you want to look from a points perspective, you could probably do better with lots of other cards. I think the other thing is from a premium perspective, it's just not platinum. It's gold. It's our co-brand cards that have premiumness to them and so forth. But what really sets us apart in our ability to provide people the spending power that they need, the ability to have their backs and provide service to them and the infrastructure that we have built over a period of time, whether that be a lounge network of 30-plus and more in the pipeline, plus we're one of the only ones that can actually use their partners lounges. I mean, we have Delta has over 50 lounges, and that's all part of it. So you have this lounge network of almost 90 different clubs you can use. We have a very successful travel business which provides services well. [ Resi ] and talk, you get over 25,000 restaurants. And so what we've learned here and you find hotels and resorts like 400 -- what we've learned is that as you're going to go out and to this premium space, it is a lot more than just points. It is about providing service. It's about providing spending power, and it's about bringing assets to bear that are going to make a difference for your cardholders. And the other asset that we bring to bear is we bring a very powerful merchant network that wants access to these customers on an ongoing basis such that they will fund offers to actually provide extra benefits. And so when I look at this segment, I see it continuing to grow and I see a flywheel that we have created here, which continues to differentiate us. And so we're going to continue to lean in. You go back to the Chase Sapphire -- we didn't have [ resi. ] We didn't have the lounge network that we have today, and we certainly didn't have the amount in Fine Hotels and Resorts, and we'll continue to do those things and add that value that is going to continue to differentiate us.

Thing about points because it's not as intuitive as a statement credit, like I spend $1,000, I get 5x, I get 8x. I don't know what that's worth, but a $300 fine hotels credit is $300.

Yes, it's $300. And the other thing is if you look at this network of hotels, it's not just the $300 credit that you can use when you book it. It's the early check-in. It's the late check. I'd always get that confused. And if I do it the other way, it's really not a benefit. Late check-in and early checkout doesn't really work so well and it's a free breakfast and it's also you get a credit at the hotel. And so there's a lot of value in that. And you probably talk about millennials later, but -- that's one of the aspects of the value proposition it really resonates with them.

And with earnings in April, you gave several usual metrics on the platinum itself and the refresh and the traction there. You also highlighted accelerating spending growth with Platinum Card members. Is that acceleration something that naturally is an outcome of the refresh cycle? Or is that a nice surprise and if it's the latter, what do you think is driving it?

Yes. I think this goes back to the framework for winning this whole refresh. We were not on a cadence of refreshing products. And one of the things that we did within our framework would say one of the ways we're going to win was to refresh products and refresh our value propositions on an ongoing basis. And so what does that do? . Well, what it does when you're refreshing products, as consistently as we have been here, and we'll talk about the platinum in a second, but when you're refreshing products, you're keeping them out in people's mind. You're making sure you're on a mission to understand what your customer actually wants out of your product on an ongoing basis because what they wanted 10 years ago and what they want today is very, very different. And what you also get out of this is as you're refreshing and investing in the new value proposition, that works in conjunction with your card acquisition dollars to really have a powerful 1 to a punch as you go to acquire. So let's talk about Platinum. So as we did platinum, what was important in platinum in any refresh is retention. And so what's happened is our retention has been exactly what we would have expected, probably a little bit better. I think people are really interrogating the value proposition and realizing that for $200 incremental fee, they're getting a lot more value. We've seen retention go up. We've seen acquisition numbers surpass our expectations. And we've seen -- and we said this in the first quarter, we saw a 6% lift in platinum spending growth above what we had expected. And that's a little bit of a surprise for us. And where it's happening is existing platinum card holders. And so why is that happening? I think even as these people -- our existing holders reengage with the product, one of the things we did with this platinum launch is we've led with technology. So if you -- when you got the Platinum card, and this is -- again, goes back to the J.D. Power award that we just won, when you got it, it was so easy to enroll in the benefits. It was so welcoming to look at that app and engage with that app. It caused engagement, not only with the new benefits that we had, but with benefits beyond what we had as well. And so we're getting -- I believe, and we'll do all the calculations. But I think we're just getting a higher share of our existing card member spending and that's not what we had expected. So I think you had a double edge here, right? Number one, you're always looking for a bump up from an acquisition perspective, and you want to make sure you keep the retention. But I think what was surprising was how much more our existing card members started to spend.

Existing members spending more new card acquisition really strong, those...

And the retention, right? Because if you look at it, the thing you want to do is retain as many customers as you can because it costs a lot to get a new customer. And then you don't want to have your new acquisition bucket of lost volume, right? You want that to build and you want that momentum to continue.

Yes. Okay. And MX has repeatedly talked about partner-funded value as it relates to the overall value prop of the cards. When I see a statement credit, has the amount that is partner-funded increased over time? And then are you going out and recruiting new partners? Or are they coming to you? And then lastly on this topic, is there ever a saturation point where there's just too many credits?

Well, let's start with the last one. I don't think there's ever a saturation point because what we do is we got -- so we have hundreds of products, right? And so you have statement credits that apply to Gold Card. You have statement credits for platinum, whether it's Delta, whether it's everyday cash, whatever it might be, small business and so forth. So I think what we try and do is not overwhelm it. The other thing is we're very selective. And I mean, look, this is not a statement credit, but when we did as we launched into 2026, we had 1,400 hotels that wanted to be part of the Fine Hotels and Resorts program. We only took 300. So we're not just taking everybody here, right? And those were hotels that wanted to give our card members a benefit, but we didn't believe that they met that standard. So we didn't do it. But when you look at these statement credits, the statement credits, the partner-funded value continues to go up. It's over $3.5 billion annually now. And that continues to go up, and it creates a flywheel effect. And so what happens is other people see these statement credits and they want to get in on them, and we do some category exclusivity for various products. But then if you're doing something with platinum, we can do something with gold, and we mix and match as we go along. But it's a little bit of a combination of us approaching partners and them approaching us as well. And the nice point about having the merchant business that we do and the merchant network that we do, it's not an issuing business just going out and asking. It's a merchant business that understands what the objectives are of some of those merchants, and they're able then to communicate that, bring it back to the issuing business and then come up with offers as opposed to going out coal to a merchant.

Arguably could make the offers more targeted to the given card base, right, because you know where they spend?

Well, we make the -- look, we make the offers a lot more targeted. We also can make the offers in such a way. And we do this with Centurian we do this with platinum a lot -- we can do an offer -- the one I've used, and this has been used for years, we did this years ago, by 3 or 4 years ago, we had a David Yurman offer, right? And we did this David German offer for Centurian cards. And so what they did is they said, look, tell us who buys jewelry and Telus who hasn't shopped in David Yurman in the last 2 years, and we'll do a targeted $750 credit off, no minimum purchase. So you're a centering cardholder, you see that credit come at $750. You go into David Yurman, and you got $750, right? And so that work -- the ROI works, jewelry has high margin. It's not going to work in a supermarket, but jewelry has really high margins. So that works. They get new customers and off it goes. And you'll see some of that stuff happen to platinum. Now some of you in the audience will say, how come I didn't get one of these. Well, you weren't in the target. That's all.

And then success with younger consumers has also been a growth driver. You talk a lot about millennials and Gen Z specifically, so what's the driver behind the increased penetration from Amex with this cohort? Is this an active choice from Amex to steer things towards a younger demographic? Or are they just kind of finding the cards on their own? And then why don't you think peers have been as successful chasing this demographic, too.

I can't answer the last piece of it because they just have not been as successful either from an acquisition or a creditworthiness perspective. But going back again to the framework. One of the things that we knew was we had to get younger. One of the headlines that when I took -- there were 2 headlines, and I talked about this a lot. There were 2 headlines that I still have in my mind when I was named CEO. The first headline was what an uninspired choice by American Express, that always sits well with you when you see that in print. Especially when you kids see it, but makes our work a little harder. And the second was, is the Amex customer base dying out. And that really resonated with me, right? Because the average age of our cardholder was a little bit older and now you look at -- and we talked about this at the earnings call, I think the average age of a new Platinum card customer is like 33 and gold cards like 27 around that. And you look at it and you say, well, we needed to get younger. And as you started to interrogate the value proposition, what you realized was there was a lot of things in that value proposition that really resonated with younger people. And that's why I said the fine hotels and resorts because -- and I just -- I have 4 girls age 21 to 35, right? So as my old 1 got out of school, she's traveling all over the place, and she's using the platinum card and she's staying at these places. And why wouldn't others? And so as we talked internally from a marketing perspective, we were targeting this cohort with a cash back card because we believe that they would not want to pay a fee. But what you realize is the millennials and the Gen Zs are our most educated consumers. By far, they do the most analysis and they get the most value out of the product, right? And so when you look at it and you say to yourself, okay, look, let's go after the best of the best, and you can see our credit numbers. Our credit numbers for millennial and Gen Z look better than the competitors' credit numbers for GenX and for boomers and forget about the disparity between the millennials that they have and the millennials and Gen Zs that we have. But the beauty of it is the lifetime value of this customer is extraordinary, number one. Number two, the share of wallet we get is also a lot higher. Because when I first got my American Express card, we didn't have the coverage that we have today. You did not have parity coverage in the United States. And I would always go out when I -- and even when I work, do you take American Express. A young person will not ask that question anymore because every place takes it. Now there'll be a place you'll find and then we'll go in and I think the cost is 5%, and we tell them it's the same as Visa, Mastercard, and we sign them up. But they don't have that hesitation and so not only are they not -- they're using it -- they're using it in their everyday lives. And that is a big thing. And so you look at it, 70% of our acquisition globally is millennial and Gen Zs or 66% to 70%. And our -- and it's premium products. And so these premium products, you get higher engagement, you get longer retention cycles on it, and we have a higher lifetime value. So we had to realize ourselves and once we started to market that way and then once we started to underwrite the right way, it all worked out pretty well for us.

Okay. AI and Agentic commerce, certainly the latest frontier. Amex has been pretty vocal in highlighting areas where you have an advantage in an AI-led world. And I think the most notable would be in the data advantage that comes from the closed loop network and so I'm wondering how Amex data advantage here compares to the data that's generated in an open-loop transaction. And where I'm going with that is to try and gauge the extent to which an open-loop model could replicate Amex's capabilities and if it can't, does it tend to reason that AI and an agenetic commerce-led world could be an accelerant for the Amex flywheel?

Okay. So 2 points I want to make before I answer your question. The first point is when you think about agent commerce, I don't think we're in the first inning. I think we'll warm it up in the bullpen, okay? So this is -- we've got a long way to go here on this. That's number one. The second thing Anytime you're dealing with that AI, AI is only as good as the data that you have. We've been using AI at our company, machine learning, structured data, since 2010. So every credit decision we make, or fraud decisioning, all that. And when you look at our credit decisioning and you look at our fraud decisioning, specifically our fraud decisioning -- that is like the greatest example of the closed-loop network, right? It's one of the greatest examples of our fraud is about 1/3 of what the networks are because the data that we have, the insights that we have into a merchant -- the fact that we know that no cardholder has ever spent that amount of money at a particular merchant. An example I use all the time is if you had a $15,000 credit limit, and you go into a Bodega Brooklyn. And I'm not picking on Bodegas in Brooklyn, but you're going to about Bodegan Brooklyn, and you spend $13,000. You're trying to do a $13,000 charge. If your credit limit is good, that's going to go through. We're going to know that no person in the world has spent $13,000 on a Bodega. There's only so much Model you can buy. And the other thing is the reality is we don't probably even have $13,000 of spending in podge in a year. And so that timing of the data from a merchant perspective and the tying of the data from a card member perspective and having a network have that is very powerful. That's in an offline world. So now you take it to an online world when you start thinking about intent. So what's going to be really important in Agentic world. You're going to have an agent that you're going to tell that agent to go out and buy you a pair of converse sneakers, size 13, green, okay? High tops. That's what I want you to get -- and it's going to go out, and it's going to come back with size 6, yellow, low-top sneakers, right? Now one of the things that we say that we're going to be able to do is we're going to be able to capture intent so when you're buying today in an offline world and even in an e-commerce world today, I don't know what the intent -- you're not registering your intent with me, right? You're going out, you're going to that website, you're buying something. We're not registering 10 because you're right on the website doing it. If you give the agent instructions, and this is what we did with the ACE developer kit, we'll be able to capture that intent, match that to what actually was purchased. And if, in fact, the merchant will not take that back or we're going to make it -- we're going to either go after that merchant or what we're going to do is back that card member. Okay? And we announced that it's probably 5 or 6 weeks ago, nobody else has followed suit on that. And one of the things you have in an agent iCommerce world, your going to have to be really concerned about is fraud. It's trust, it's security. And so what we want to do is use the data that we have and use the relationships that we have from a closed-loop network to make sure when our card members decide to do an agent transaction that we have their back. And here's the reality. We have their back today, but we don't have the data. So it's going to make it more secure for our card members and probably more cost effective for us going forward. So look, would we be a winner in agent commerce. We think we're well positioned to be a winner. Will that happen overnight. That will not happen overnight because I don't know how many people are going to just turn their lives over to agents right away.

You've also emphasized embedding Amex assets into AI platforms and building your own proprietary AI or agentic experiences -- can you give us an example of what that might look like? And then what are the opportunities that you see here?

Yes. So look, from a payments perspective, and it is what the ACE developer kit comes in, and we've been working with Stripe and others to actually embed Amex within. And so that's what you see embedding and other platforms. We just announced with Anthropic last week that -- or a couple of weeks ago, that their agents will be able to operate within resi. And so you'll see that whether it's resort and we're bringing those 2 systems together. And then within resi, within to, within our own app within our travel business, will be developing agents within those ecosystems. And if you think about what we've tried to accomplish here between our resi and talk acquisition and our travel assets, we talk about the closed loop. The closed loop is matching consumers and merchants at a very macro level. From a resi perspective, we're operating within that closed loop and we're matching consumers with restaurateurs, creating another closed loop within the closed loop. We do the same thing in travel. Consumers with hotels, airlines and card rental within that closed loop. And so you'll see those flywheels get created within loops which is why as we think about ads as we turn to ads and offers from a monetary perspective, we see embedding those within those loops.

Lastly, on AI, you've laid out ways in which Amex is using it internally. So can you give us some more detail there? And then back to where we started, what do you do with the excess operating leverage that, that generates. Does that drop to the bottom line, get redeployed in marketing in BC? Or does it go through the same filter as any other sort of extra earnings, if you will.

Yes. I think -- so look, as we think about AI within, as I said, we've been using AI for credit and fraud decisioning for 16, 17 years, and it's all been structured data. So if you just start with where we've been with fraud and credit we'll use unstructured data to see if we can get better outcomes and better results. We're using agents within our marketing -- our Card Member acquisition engines, which takes out time and makes it more efficient. In tech, we're seeing 30% reduction in coding right now, coding and testing and so it allows us to get to more -- we've got a backlog. I mean, when you're as big as we are, and you've got a network, a merchant acquiring business, issuing businesses that are in different countries and you're operating corporate cards, small business cards and so forth, there's a lot of pent-up demand for technology and then in customer service, we're using it as well, whether it be in chat or in travel and coming up with itineraries. What we see it as -- listen, when you look from an operating leverage perspective, our OpEx to revenue ratios continually come down. But we want to grow. And when I took over, we were a $30 billion revenue company will get close to probably 80 this year. And the reality is when you're 30% and growing at 6%, you're looking to add $2 billion of revenue every year. When you're 80% and trying to grow you're now looking at $7.5 billion to $8 billion of revenue every year. So you've got to invest money to continue that engine to go. So look, look, our goal is -- our aspiration here is 10% revenue growth and mid-teens EPS growth. And I think that requires continued investment in the value propositions and efficiencies within the company as well.

Yes. And on the balance sheet, for several years, Amex has been pushing to do more lending with card members. The rationale being that they're going to borrow from someone might as well with us. Are you still gaining share of card member lending? Or is that effort "mature" at this point in time? Do you want to do more lending?

Yes. I mean, look, we want to serve our customers, right? And so the lending that we're doing is you're not going to see balance transfers, you're not seeing us with crazy offers out there just to bring everybody in. We're focused on premium lending, right? And so our focus is on meeting the needs of our premium customers. And Look, the reality is the reason that we've done this is that you don't want to turn over your customer to somebody else. And so we're trying -- it's the same reason we've had high-yield savings and checking and small business and so forth, we want to be able to meet more of their needs on an ongoing way. And so -- when you look at it, if you look at higher FICO scores, anything over 720, we're probably growing 2x the industry right there. But if you look at it overall, our lending growth is the same as our billings growth. But if you look at our NII growth, that growth is higher because we've been doing a better job funding because of our savings activities. We've been doing a better job pricing and our credit box is outstanding, right? I mean our write-offs are just we're around 2%, 2.1%. So we'll continue to lend that way. And what you're seeing for us is that -- you're seeing a lot of it from a pay-over-time perspective, you're seeing Planet and so forth. So a lot of it is short-term cash flow need lending to very high-quality customers.

And is that the way you continue to grow in lending, but keep the credit outcomes really good?

Yes, you're not -- our objective is not to grow lending. Our objective is to meet our -- we define who our customer is, okay? And then we look to meet their needs. And so if their needs are lending needs, we will meet those lending needs. And that's how we stay within the credit box. The other way you stay in the credit box, when you look at where the lending occurs, A lot of it is pay over time, and a lot of it is co-brand. And co-brand lending for us is very profitable, and it's also because there's an extra tie, right? I mean we don't want to lose status. And so a couple of years ago when we did our Investor Day, we showed the hierarchy of payments. And we showed American Express cards were always paid first. Part of that was not only because you didn't want to lose your points and your status would America Express, but you didn't want to lose your status with Delta. You didn't want to lose your status with Hilton. You didn't want to lose it with Marriott because if you default, you're kind of done.

The U.S. commercial segment. That 1 has been a little bit slower recently but you've started to roll out a slew of new products and services that are intended to reignite things. mean can you tell us how you see the consolidated offering, including expense management, stacking up to what has become a pretty competitive field?

Yes. I think when you look at -- we look at commercial space, the commercial space is made up of 3 different segments. It's the large corporate and large global, which continues to grow at a normal 4% to 5% pace. I mean it's not anybody -- there's no corporation out there saying, get out and spend more. Go out and travel book premium hotels, heated fine dining and buy the most expensive bottle of wine that you can find. I don't think anybody's companies -- if you have a company that's telling you to do that, please introduce me to them. Maybe I'll go work there. But we don't have that. So we're really happy with that. From a small business perspective, we see small businesses continuing to grow. Where the struggle comes in is that middle market space. And that's where I think we're a little bit off in our offering from 2 perspectives. Number one, we would slow to the uptake. We decided to partner with expense management systems versus having our own expense management system. So we pivoted. And so center will come out. We think it will be competitive. And I think when you put -- the card aspects of higher spending service along with the basic functionality that we will offer and continue to improve on, I think we'll be able to regain our footing in that segment. And so -- and then you've got the cash back card. And so what you realize is that in that segment, you saw a need for some cash back. We saw a need from a corporate cash back card, which comes out in September or so. And so I think with the things that we're doing and the investments that we're making, we will regain some of that traction. But it's really in that middle market space that I think that's where I think we just haven't grown as quickly as we need to grow.

Does center plus maybe hyper put you back on offense in that category?

Yes. I think it does put us back on offense. I think because what happens today, and we see this, right? If we lose somebody to a ramp or to a brex, the owner still keeps their platinum card, okay? But it's the employees that get the other product. So I think it allows us to go back in from an offensive perspective and go back to some of the accounts that we might have lost. But I think also from a defensive perspective, it prevents future loss and it allows us to it allows us -- I think what the big opportunity for us is we are stronger than anybody in true small business. As these small businesses start to ramp up, we will bring them along on that product line. We will introduce them to center earlier. And so there'll be no reason to look any place else. And so when you look at the true small business space, the true small business space we still have I think, the best product, and it's not a group really that needs that technology yet, but they will over time. And so as we move then, we'll be able to embed center right in.

International is on the other side of the ledger. That's a segment that's been growing quite nicely for some time. How long can you continue to grow double digits there?

Long time. A long time.

A long time.

Long time. You could -- that would be the CEO said a long time. Look, the reality is, one of the things that we did again, going back to the framework was we also identified those countries that we really wanted to double down in and focus. And we identified what we really wanted to do from a coverage perspective. And so -- we identified 5 key markets, which was Canada, Mexico, the U.K., Japan and Australia as ones we would really invest very heavily in. Then we identified at the time was 35 cities, it's up to 50 cities outside the United States that we wanted to make sure we had 80% location coverage in. Not that we wouldn't focus on coverage in other places, but we wanted 80% location coverage. Why 80%? Because the 80%, we believe we can get $0.95 out of every dollar you spend, okay, with that kind of a number. And so at $0.95, you don't mind pulling your card out and every once in a while, you might -- somebody might say, hey, look, we don't take Amex for whatever reason. And so as we think about the constant improvement, we're not stopping at day, we want to get to 80 and continue on. As we think about the constant improvement that we've had from a coverage perspective, as we think about the investments that we've made and as we also think about the fact that we're probably only at about 6% market share in any of these markets overall, from both a business perspective and a consumer perspective, we believe there's a long runway for growth in international. And then we have other countries that we call growth countries and in others really where we have a presence, but really more of that presence for coverage so that when people travel they're able to use the card, right? And so I think there's a huge opportunity and a long runway for growth in international, and that has proved out. I mean, look, the only time international didn't grow was [indiscernible] But if you go back pre-COVID, international was growing at the same rate, and we continue to grow and we continue to add coverage. And so I think -- the opportunity for growth is terrific for us there.

So the coverage discussion is interesting because to me, building out international coverage sounds expensive. But despite the push there, you haven't seen -- we haven't seen much pressure on international segment margins over the last several years. And so why is that? Is it the kind of thing where the businesses operations are just funding the investment? Or is there something else to it?

No, there's something else to it. So the reality is, is that what you're seeing now is the fruits of our labor for decades. Building an international network is wildly expensive. Okay? Let me repeat, wildly expensive, and it requires many, many things. Number one, you can't build out an international network without having local card members which is why our GNS business is so important because our GNS business, for countries we don't operate in from a proprietary perspective, we have franchisees that are issuing cards. So you need to make sure you are issuing cards in every market or or locals will not see anybody. That's number one. You've got to go out and build infrastructure from not only a technology perspective but a regulatory perspective to do this. The second thing that you need to do is then go out and acquire a bunch of merchants. So we've done that for decades. We're at a point now where we're at scale. And so the reason you're not seeing sort of margin compression because we're not giving coverage away. The reason you're not seeing margin compression is because we have a scale business now from a regulatory, from a technology perspective, and we exactly what we want to do, and we have partners out there as well that are helping us. And so all of that investment in time, effort and money is really coming to pay off for us now, which is why the international business, I think, has huge upside for us as we continue to move forward.

I just didn't go far enough back in my 10-K.

You should have gone back a lot further -- you should have gone back a lot .

All right, 1 audience question. It's the only one that has any votes here. what leading indicators spend category mix, payment behavior would flag AI-related white collar pressure on platinum and gold before delinquencies and have any of those shown up? The question cutoff, so I'm just going to infer that.

To be honest with you, I think it's really hard to look at a spend indicator before sort of delinquencies. I think discretionary spend is the one that usually happens first, right? When people struggle, it's discretionary spend that goes down. So I guess my answer would be discretionary spending, whether that's travel, whether that's luxury goods and so forth. And I guess I'll take you back to COVID because during COVID, before anybody went bad, you saw this is one of the things that people were really confused about. Our balance sheet got really small from -- during COVID because our card members tend not to spend. So before they go delinquent, they don't spend. And so I guess we would see if you saw a lot spending come down, but I think what we would see before that is unemployment, right? So I don't think AI and genetic commerce and so forth is going to sort of disrupt our spend so much. I think the reality would be you would have to see unemployment. The counter to this is that our card base is not just a white collar card base. Our card base spans the gamut. We tend to over index actually from a health care perspective. And I don't see health care workers actually on a decline. I see, if anything, AI is going to add to longevity. I think we're all going to live a little bit longer, hopefully. But I think we're all going to hit a little bit more care. And so I think -- if I had a kid that was going to college today, I'd say get into the health care field because I think the health care field is going to be tremendous. The other thing is small businesses. So many of our card members are small business owners, not only from a small business perspective, but from a personal perspective. And we have lots and lots of entrepreneurs. Our card base sort of runs the gamut. And I think -- the conversation around AI has been white collar jobs going away, and we just had Sam and Dario yesterday come out in a Fortune article and say, well, we overstated the whole thing. And I think David Solomon has been pretty specific as well saying he doesn't think there's going to be armageddon. I've been in the same way. My view on sort of AI and its impact on the workforce is the technology view, which says, listen, we went from a farming country to a manufacturing company to a service country to a technology company. And every innovation that we've had has spurred GDP growth. And every innovation that we've had actually. [Audio Gap] And so technology has been able to put private capital into the public sector, and it's been able to make entrepreneurs out of people that just like to sit at home all day, okay? So I think to me, I think it's completely overstated -- that's not to say that AI will not replace jobs. It will replace some jobs, but I think ultimately, it will create new industries. It will create new jobs. And again, if history is our guide, technology has always spurred GDP growth and GDP growth spurs more wealth, more well spends more spending and off we go.

Off we go. Very good. Okay. We have time for 1 last question. I'll give you a blank canvas. When we're looking back and hopefully talking about this session, 5 years from today, your strategy, your framework has played out. What does Amex as a company look like then? And what will the outcome have been for shareholders?

Well, let me talk about the outcome, and I'll tell you what we look like. I think -- look, I think the reality is I think we've created a flywheel. And if you look -- take COVID out of it, we've been a high -- what I consider a high revenue growth company for the last 8 years. The CAGR has been almost 10% revenue growth. It's a mid-teens EPS company. It's a company with a high ROE. It's a creditworthy base in a growing segment of payments. Payments is growing itself and the segment and premium is growing as well. And we're growing with young people. And so if you're interested in a great global brand, that has those great financial economics, I think that's where you want to be. How the strategy plays out. Look, I think our framework continues to evolve, but I don't think we get too far off on knitting here. And I just talked about international as a growth engine for this company. I think premium the TAM is going to increase. And so I think what you're going to see is American Express being bigger internationally. I think we will continue to be a big player from a small business perspective, and we will continue to reinvent the premium space and the premium space that is a space that's open to people that want experiences, that people that want access and the people that truly value all the value that American Express can bring.

It's a great place to leave it. Steve, thanks again. Thank you very much everybody