Richmond Fed President Tom Barkin on the Fed's next move — 11/05/2026

[00:00:00] Speaker 1: I'm happy to introduce our first session. Steve Leisman is CNBC's senior economics reporter covering all aspects of the economy, including the Federal Reserve, and Tom Barkin has been president and CEO of the Richmond Fed since 2018 and a CFO before that. From day one, he's been on the road across our district listening to businesses and you might say recalculating our route. So please welcome them. [00:00:30] Steve Leisman: Well, thanks for that intro. Tom, thanks for sitting down with me and thanks for having me down here in Charlotte. Yeah, it's always great to have you here in the district and get you out of New York. I was going to thank you for that. That's always a good thing to get out of New York. The trouble with getting out of New York is getting back into New York, [00:00:59] Tom Barkin: as you might know. We have a lot to do here in Charlotte, so stay the whole conference. [00:01:03] Steve Leisman: Okay, sounds good. The intro helped with thinking about where to start this. It's a CFO conference. You were a CFO and perhaps, I don't want to say unique, but among the people I know on the committee, you spend an awful lot of time talking to business. So I thought we might start there. You told me before that you had sat down yesterday with a bunch of executives. So tell us what you're hearing from business and how you go from that to a view of the economy right now. [00:01:32] Tom Barkin: Yeah, so I'll start by saying demand is solid. And there's a lot of things that have happened in the world that you might imagine would take you off the zone of demand is solid. But demand is solid. Now, that's not true everywhere. If you're in residential or nonprofits or a non-defense government contractor, you don't feel like it's solid. But overall, the economy is healthy. Despite that, you're not seeing a lot of growth in workforce. And that's because there's a lot of uncertainty in the world, not least of which is technology and how that's going to evolve. And so businesses are still in this cautious place on hiring. But they're not really yet tipping over to firing. And I can talk some more about that. And then you hear broadly that costs are high and are increasing. And in particular, over the last several months, you hear about gas prices, diesel prices. But, you know, it's also health care and health insurance. People are seeing that increase, those people who are exposed to tariffs, there's a bunch of people who got hit with cost increases six months ago and decided for various reasons, you know, not to pass them on, and are now saying, hey, I've got to pass them on. So there's cost pressure. And I think if you step back from all of it, the one thing from the business side that's keeping the economy together is the strength of earnings. And, you know, second quarter earnings were up over 30 percent average. The median was 15 percent. Third quarter looks like it's very much in the same range. And it's just a set of behaviors when earnings are strong that are pretty natural. You invest. You don't cut back on spending. You're looking at the forward and you grow. That's all good. You don't lay people off because why would you lay people off when you're making all this money? You may not hire and you still see the caution hiring. You don't lay people off. And those are the behaviors we're seeing. And I think that virtuous circle sort of continues. And that's part of why demand keeps, I'd say, surprising a lot of people to the upside. [00:03:32] Steve Leisman: All right. So there's only about 100 follow-ups from that whole opening line there. I want people to know that one of the things that Tom turned me on to economically early on in the pandemic was this notion of a culture of pricing among business. That's not in the theory, I don't think, which is, you know, they look at cost pressures on the front end and whether or not we have costs on the back, people pass them along or not. But an idea of a culture of businesses raising or not raising prices, which is, you know, you get to the CFO or you get to the guy who makes those decisions or the woman who makes those decisions and say, can I pass them along? Well, they're passing them along. I did it last month. It worked or it didn't work. So I want to come back to you, Tom, with your assessment right now of that culture. Is this a culture where businesses feel more comfortable passing them along? Or is it one where they're still reluctant to do so? [00:04:25] Tom Barkin: Well, so just an interesting fact I figured out after I took this job, you know, we all, people of my generation, you know, still revere Paul Volcker and the actions he took in the early 80s to bring inflation under control. At the end of all those actions, inflation in 1986 was all the way down to 4%. It wasn't 2%. It was 4%. And the grind to get it to 2% was another 10 or 15 years. And I think that's because inflation tomorrow has a lot to do with inflation today. And the story you're telling, which I appreciate your nice words, but, you know, I was a CFO of a $10 billion professional services firm. We really did sit down every October and talk about what should we do with our pricing next year. And the people who had to deliver activity didn't want to be very aggressive on pricing. And the people who want to deliver bottom line performance wanted to be more aggressive on pricing. And as you have that debate, and you guys are doing this every day, the topic of what inflation is, is a very important reference point in that debate. You win arguments more often as a CFO if inflation is higher to raise prices. You lose arguments if the rest of the world isn't raising price. And so I think it does take time to grind that out because you're always using today's lived experience as a reference point. And, you know, what I hear today, and I'll go back seven years, pre-COVID, most everyone didn't believe they had any pricing power. I mean, it's awfully hard to pass it on. Maybe if oil prices were up or tariffs were up or, you know, some big commodity that was an input was up, you could pass it on. But other than that, the big box retailers weren't going to take it, the customers weren't going to take it. Then, of course, we hit 2021 and 2022, and there were shortages everywhere, you had no option, you raised prices. And because prices were going up everywhere, you know, in your businesses, you raised prices too. I'd say if you flashback 18 months ago, I think that era was coming to an end. I think consumers were exhausted with higher prices, people were starting to push back. But there's no question in my mind that the set of cost pressures that we've been facing over the last year and a half, starting with tariffs, including oil and diesel, including healthcare, as I talked before, these are broad-based costs. They're very visible. Everyone's getting hit with them. Aluminum, I hear from a can manufacturer I talked to this week. People just feel like they have to pass them on. Now, they're having a lot of success B2B. So if you talk to B2B companies that get passed on, you have a generation of procurement professionals who've accepted price increases. It's not this foreign concept anymore. If you're pricing to the wealthy, it's getting passed on. I mean, don't look at what you're paying to get your kids home for Christmas this year. It's harder, I'd say, in the moderate-income consumer. The moderate-income consumer is tapped out. But I was looking at earnings last week in Whirlpool, just to pick on one company. They announced 7% price increases for the fourth quarter. Now, the segment that Whirlpool sells into is home sales. I mean, that's what really drives their business. That is one of the segments that's not very strong. And so if somebody's increasing prices 7% into a segment that's not very strong, they're sort of saying these cost pressures have hit. I've just got to do this. And if there's volume risk, there's volume risk. And so that's what I hear across the segments. Very straightforward in B2B, a bigger call in B2C, but some people are making it. [00:07:53] Steve Leisman: One of the consequences of what you said, I always looked at, for example, cost pressures in the producer price index. But I'm spending a little more time with some of the surveys. And all the surveys I'm reading suggest that businesses are, if they could, they would. What does that mean to you for policy? Does it mean that there is potentially higher inflation coming down the road? [00:08:17] Tom Barkin: So inflation is elevated. Our target's 2%. Headline right now is 3.6. Core, I think, is 3.1. That's not 2%, and that means it's too high. I have to say, I'm not hearing it escalating. I'm not hearing that we're going from 3 to 4 to 5 to 6. But I also don't hear it diminishing. I don't hear that we're going from 3 to 2.5 to 2. And of course, the inflation is a basket of all of these goods and commodities. And some go up and some go down. But that's what I hear on inflation. I think the implications for policy are pretty straightforward. Demand is solid, as I said. Unemployment's near record lows. And inflation's too high. That means the path is pretty straightforward. The question, of course, is pace. And in the SEP in September, the signal was a gradual pace. Gradual makes sense if you think there's uncertainty in the world. And if you think that there's information you're going to get in the period. And I think that's still a pace that makes sense to me. And when you think about where [00:09:22] Steve Leisman: policy is right now, how do you describe it? Is it restrictive? Is it neutral, modestly restrictive? And where [00:09:28] Tom Barkin: does it need to go in your opinion? So for those who don't spend as much time on this, as Steve and I do, there's a theory in economics called the neutral rate. And you're supposed to think about policy and say, is the current rate of interest neutral or is it restrictive? And that it's above neutral, lower than restrictive. The problem with the neutral rate, it's like a unicorn in a forest. Everyone talks about it, but nobody's ever seen it. The modelers who look at the neutral rate would actually have a two thirds confidence interval of about 150 basis points. So it might be three, might be four and a half, might be one and a half. It's not that helpful, you know, for policy terms. It might not be a unicorn even. It might be something no one's ever seen. So, you know, we have good models in the Richmond Fed. The models we'd have, we'd say, policy right now is somewhere around neutral, not really that restrictive, not really that accommodative. Another thing to look at is just the level of activity. You know, one that I like to look at is nominal consumption. If you're restricting the economy, presumably you're restricting consumption, which is, you know, roughly two thirds of the economy. And that's relatively elevated. So that would make the case that we're a little bit accommodative somewhere, somewhere in that play. When you look at what's happening, [00:10:46] Steve Leisman: this is a two part question, but I'll ask the first part first, which is when you look at what's happening with the AI build out, I'm going to ask both parts together. Do you see any restrictiveness in policy? And then the second part is, do you see any ability to have policy [00:11:04] Tom Barkin: restrictions? Well, so there's no question that, uh, in this demand, uh, story that I was telling earlier, AI is a big part of it. And, um, you know, I think in a week in February, we saw $700 billion of announcements. Um, business investment grew, I think 12%, 13% year over year. I mean, you're seeing enormous investment. That's absolutely fueling, uh, parts of the economy. Um, uh, you know, I don't think the number one thing that would constrain or reduce a investment would be rates. You've got investors with a lot of cash on their balance sheet. Um, you've got a transformation that may be historic and people want to participate in it, but I absolutely wouldn't say that rates don't do anything to do that. And you can certainly see what's happened to the financing environment, you know, even over the last month or two, since we started raising rates and say, um, you know, it has some implications to some of these, uh, I think they're called neoclouds that have had some issues with their, um, access to capital. And so, uh, you know, when you raise rates, uh, it has an impact. All economists believe you can't say it has zero or a hundred percent impact. There's marginal impact. It certainly would have more impact. And I apologize to those of you who are home builders or auto dealers, or, you know, even banks. It has more impact on them than it does on [00:12:21] Steve Leisman: AI, but I wouldn't say it has no impact on AI. So I asked that question in context of the other question, which is how far do you think the fed needs to go here and how will you know that, uh, [00:12:32] Tom Barkin: you've gotten there? Well, we'll sort it out as we go. I don't think there's a fixed destination in my, in my mind. I think you're trying to evaluate, uh, consistently what's happening on both sides of the dual mandate. Uh, and inflation, as I said, is the most important risk right now. And so we should monitor what's happening to inflation over time. And I hope that our move in, uh, September and whatever else you might choose to do in the future will have some impact, uh, on inflation. You also have to, at the same time, be watching the employment side, uh, of the house. And there are certainly risks, uh, risks there. And, you know, how, how this thing plays out, I think you just have to see. Um, there's an election this week. [00:13:12] Steve Leisman: When was that? I wasn't, I've been working. Tuesday. This is Thursday. So it was two days ago. Holy cow. Yeah. Yeah. What happened? It was on the news. Yeah. It was on the news. Um, a funny thing happened. Um, it's fallen. The States changed colors. Yeah. Um, I might steal that line for next week. That's a good one. I stole that from somebody else too. Um, that's what we do. Right. Um, so I know this is like the, your favorite topic to talk about what's happened in politics. I'm going to try to frame this in a way you might answer it, which is this, um, uh, do you believe that fiscal policy over the past several years has been expansionary? And what is your expectation for fiscal policy over the next couple of years? And does the election change that at all? I think that's a fair one, right? Yeah. Um, [00:14:09] Tom Barkin: um, uh, well, so thank you for mentioning politics. I really appreciate those questions. [00:14:13] Steve Leisman: Um, you didn't commit for me all the way down here for the easy ones. [00:14:16] Tom Barkin: So this is the simplest, uh, the simplest way to think about whether fiscal policy is expansionary is to look at the budget deficit and the budget deficit is sizable, uh, and historic levels in the middle of an upturn. And so that would make a strong case that it's been expansionary. And I think that's how I see it. I think the fiscal policy in total over the last two, five or 10 years has been a tailwind, uh, to the economy. Now there's then a whole bunch of other things that, uh, happened to the economy that you could put in fiscal, um, everything from, you know, what's happened geopolitically to, you know, what are happening to tariffs to, uh, what's happening to regulation to, you know, there's a, uh, you know, what's happening to energy policy and those things have, you know, some, uh, tailwind, some headwinds, you know, it's sort of hard to put those together. What, what we try to do is not to forecast policy. I couldn't come up with a point of view on what's going to happen in the future. Um, but we are trying to, we have to react to what happens in policy. And so if there's a lot of stimulus as happened in 2021, you know, that's expansionary. And that's probably something, if you go back to 2021, we should have done more to acknowledge and do something about, um, if, uh, a set of actions happen that are contractionary, okay, that, that tells you something. And so you react to whatever's happened and you try to stabilize the economy from an employment and inflation standpoint is as best you can. Uh, but I don't, I'm not in the anticipation world. And honestly, I don't even know how to anticipate where policy is going. Is it something when it is expansionary that you have to lean against? Uh, if it's going to have an impact on inflation, yes. I mean, we know, expansionary policy in and of itself is not a problem. More employment is not a problem, but if inflation starts headed in the wrong direction, that is the problem we've been tasked to solve. [00:16:03] Steve Leisman: To date, we got a huge productivity number. It was three, six. I want to say we were looking for two something, I mean, two, nine, I guess. Um, and that's two months, two quarters in a row of good productivity. Do you have a theory as to where the productivity is coming from now and what your outlook is for productivity and how that will influence your view of policy? [00:16:28] Tom Barkin: Yeah. So I want to start by saying how happy I was that the unemployment claim number was low and the productivity number was high. By the way, this is what I do to the chairman all the time. I Exactly. Three, three, three questions and he hates it. I, I, I come to, I, I, you know, it's much better than coming and saying, well, the claims just spiked. What are you going to do about employment? Right. So, uh, a bunch of good numbers this morning. Um, I definitely think productivity has expanded over the last, uh, five or six years. And I think, uh, what's happening, uh, right now is consistent with that. A lot of people want to credit that to AI. Um, and I, I don't really think, uh, that's where I am at this point. Uh, I think in 2022, a lot of people got caught short workers. And if you get caught short workers, you know, what do you do? You, um, uh, you automate, uh, you come up with new staffing models, you come up with new operating processes, you invest in lean. And in the factories that I've been to, and I've been to a zillion of them, that's the conversations we're having about this is the stuff we implemented two years ago, three years ago, four years ago. And here's how that's paying off in productivity today. And I think the biggest part of the productivity has been those investments. Now, I do think AI is also helping people who deploy it. Um, and in particular, you hear that in coding, you hear that in call centers, you hear that in, I'll call it pretty repetitive compliance or documentation processes. But there, you know, you don't see the classic productivity. I'm going to lay a bunch of people off. You do see, um, and this would be where AI is helping productivity, a whole bunch of companies saying, why can't we just figure out a way to do this technologically as opposed to hiring people to do it? And so I think the, the incoming spigot has been shut down in part because of the potential of AI. But I don't think the reality of it is what's driving the productivity. I think the productivity is really being driven by the combination of, um, all the investments people made two, three, four years ago and, uh, expanded continued growth in the top line while people are being conservative on the employee side. Do you, do you think that the initial impulse from [00:18:29] Steve Leisman: AI right now is an inflationary one given what it's doing to memory prices and, uh, the cost of the [00:18:35] Tom Barkin: capital goods that are being imported? We definitely see the inflation on that. And if you, I mean, don't try to hire electrician, uh, the, the data center build out and hire one before. Yeah. The, uh, well, uh, you'd be very impressed with what people are paying for, uh, uh, for electricians these days, materials that are going into the data center build out. Absolutely. Memory chips, you know, you've seen what Apple's done with their pricing. So it's absolutely had an inflationary impact, uh, in the near term. And, um, you know, I think there's a lot of, uh, hope and expectation that as we invest in AI, it'll create a lot more productivity. It'll bring prices down and it may well in time. But when you talk to businesses, they don't start with, I'm going to invest in AI and therefore bring prices down. It's they're going to invest in AI and bring profits up. And then eventually competition will have its way and prices will meter. So I think one comes before the other. It's a cool thing that [00:19:30] Steve Leisman: Virginia's in your district, right? And Virginia is like ground zero for data centers. Yeah. Tell us how the, what the economic impact of that is in your, in your district. Well, so, um, uh, you know, [00:19:43] Tom Barkin: it's, uh, when you build a data center, there's of course, a bunch of construction activity, and that's very good for the construction people, some of whom are in the district and some of whom travel to build those, uh, data centers. Um, you know, part of the challenge, and we have the same issues in Virginia as many other places have, I heard in South North Carolina, Salisbury has actually passed a ban on data centers, which is a suburb of, not a suburb, they would be insulted, which is an hour outside of Charlotte. That was, almost screwed that one up. Um, but, uh, uh, but the, the challenge that data centers have is there's this initial burst of construction activity, and then there's an ongoing tax base that gets played, but there aren't that many employees. And I kind of think if you build a manufacturing plant in a small town, uh, and you go to a little league game and you sit in the stands, you'll talk to, uh, the parents, some people who work at the manufacturing plant about their kids. And if you go to a little league game in Reston, you won't find anyone who works at a data center. And so there's no natural support constituency, um, beyond the economic developers who like the tax benefits. And so in a world, and I get to see this out on the road, in a world where, um, there's lots of concern about AI and what it's going to do, and also concerns about environmental impact that have been amplified. You know, people are really anti data centers, but no one actually hears the positive. I sometimes joke they ought to start naming the elementary schools, you know, after Neo clouds or the Google elementary or whatever. Cause you know, again, when I grew up and you played little league, you wore the name of the local business on your Jersey. That's not how it works in a data center. [00:21:22] Steve Leisman: It's just saying that there's a joke, right? One of the three or four good economics jokes I know, which is that the factory of the future will have a man and a dog in it. And the man is there to feed the dog and the dog is there to make sure the man doesn't touch anything. [00:21:38] Tom Barkin: I like economic forecasting was a better weather forecasting look good. That's my favorite economics. [00:21:46] Steve Leisman: So let's follow that through. First of all, in, in, in the regions where you have the data centers, do you see higher electricity inflation, high utility inflation? Is that something that's affecting [00:21:57] Tom Barkin: local people? Do you see that? Um, I haven't seen them as connected. Um, I'm sure there are people who are closer to the, um, mechanics of it. Um, but it is true that data centers are going up in large parts of the country. And it is true that electricity costs are going up in large parts of the country. Um, that doesn't have to be connected, right? I mean, there's been a lot of issues going on in terms of cost, cost of gas and cost of oil, and depending on what fuel you're using. And so, um, people have absolutely put those things together and I, it's not obvious to me that they are. I mean, I think the, um, um, uh, uh, uh, the, uh, rate makers in most of the states that I'm in are highly attuned to the question of whether net net data centers affect, um, individual electricity rates positively or negatively. And the rate makers that I've talked to are very strongly of the opinion that it's not, but I'm not close enough to the data to know. What I really want to do though is pivot from [00:22:54] Steve Leisman: electricity rates in Virginia on a micro level to the macro quandary that I see AI providing to Fed policy. And let me see if I can lay this out. On the one hand, it creates this need perhaps for higher rates in the sense of the inflation we talked about on the front end of the build out as well as the demand for capital, but what happens and how does Fed policy respond if the result of AI [00:23:20] Tom Barkin: is large layoffs and unemployment? Um, do we have a positive scenario we could talk about here? [00:23:28] Steve Leisman: I mean, there's a positive scenario embedded in that. No, I'm serious. I mean, in that you could have people making tons of money, uh, the stock market could be way up and, um, uh, you could have tremendous gains in productivity, but that could come with large numbers of people losing their jobs. [00:23:46] Tom Barkin: Well, you're always in this situation where, um, the economy is this big, wonderful, confusing, complicated, uh, algorithm, and you're trying to sort out what the best policy is in a very complicated world. So, um, you know, just to take your scenario, you know, let's imagine for a second that, um, uh, AI rolls out and it makes everybody more productive and nobody loses their job and inflation comes down. I mean, that's also a scenario. In that case, I will go on CNBC and talk about what a great job we've done in monetary policy. Another world is where, um, you know, AI is inflationary, um, but it's not having that big of an impact on jobs. You could argue that's where we are today, in which case policy sort of says, lean against inflation. Um, there are two versions of the world you just described. One is where, um, AI is significantly disinflationary, but it's leading to lots of job losses, right? And that disinflationary could be either, um, direct because everything's cheaper and people are competing it down, or it could be indirect because people lose their jobs and they pull back on their spending and that is disinflationary. In which case we should be leaning to the part of the, uh, mandate that is employment. And then there's another scenario you might've been talking about, which is it continues to be inflationary and it's leading to lots of job losses. And that's of course, the, the box that's very hard for us, but I'm going to root for one of the first three boxes, three, three chances to make [00:25:09] Steve Leisman: it. Um, I want to pivot to something that is very, very arcane, I think, but it has a potential for having, um, uh, an effect, which is the chairman told us that you're going to go to fewer, fewer meetings in 27 from eight to seven rate setting meetings, and then from eight to six or seven to six in 28. Um, did you approve of that idea? Do you have a feeling about it? Is it a good thing or a bad [00:25:39] Tom Barkin: thing? And then I want to ask follow up on that. Yes. I'm not, I don't think it's much of a big deal whether we meet every six weeks or meet every eight weeks. That's really what the conversation is. Um, I don't think the world moves at a pace that we have to move, you know, if you don't like, uh, moving two weeks up, should we move two weeks? Should we meet every month or every three weeks or every two weeks? It's history has said six weeks, let's try eight weeks. We'll see. Uh, we'll see what happens. There is a, so I don't think it's that big of a deal. There's one very attractive feature of it to me though, which is I really don't like having FOMC meetings and then walking out and getting a jobs report on Friday. I feel like if you did it every two months, then you could in fact have the full suite of data twice in a row and make decisions based on two cycles of data, not one or one and a half, or in some cycles one and some cycles two. And maybe even you could schedule it in a week where we don't have a jobs report the day after the meeting and we don't, or we don't have an inflation report, which has sometimes happens the Wednesday of the meeting. So I think if you could pick the right week and actually separate it from the data cycle and give yourself the time to really stop and reflect, I think that's a very interesting feature. And I don't know if we're going to pick the right weeks or not, but that's certainly what I'm hopeful we'll do. So my, my follow-up though is, [00:26:53] Steve Leisman: does it create the possibility for more volatility in the sense that there's longer between meetings [00:26:58] Tom Barkin: and more things can happen in between? In theory, I mean, and of course, again, you could meet every four weeks and that would have less and every three weeks and two weeks. I don't think there's a right or wrong answer. I think there is this challenge where, or maybe it's an opportunity, is that we've all gotten kind of grooved, right? And I'll just, you know, grooved in how communications works, grooved in how the meetings work, grooved in how the reaction function tends to be based on a certain set of data with this sort of frequency. And anytime you make changes, you get a little bit out of the groove. That's not a bad thing, but it does take some time to adjust. So I could imagine, you know, if you start meeting less frequently, you'll have times where we say, rather than increasing 25, we'll increase 50 or reduce 50. Because, you know, there's fewer meetings to make a choice or maybe we're less likely to do a gradual, every other meeting pace. And we do every meeting pace because, and so some of those things might not be expected because we haven't had the same. So sure, things could change. But again, and I think the chairman's been relatively articulate on this, their cost of being too grooved to, you know, expectations get built up and then it's hard to move. And so, again, I don't find it to be, you know, there's a hundred things we're worried about. I'm worried about it. It's not on my top 100 list. But others, I think, may have different views. [00:28:26] Steve Leisman: Broadly, when it comes to communications, do you think the Fed needs to rethink that? [00:28:32] Tom Barkin: We have a new chair. A new chair has a different view on communications. And so, for sure, we need to rethink it. We need to land in a place that he's comfortable and the committee, you know, buys into and we move forward. I think, I think he's made a really good comment, which is that, well, I'll say, I'll say two things. There's a, there's a historic view in the world of economics. One of the recent Nobel Prize winners was a prominent proponent of it, which is the better you communicate, the more markets can help you do the work. You set expectations, markets move. When we first moved in March of 2022, 25 basis points, markets moved a lot more than 25 basis points, because they were sort of anticipating this is a hiking cycle and we know what happens. And that, I think, is a very articulate philosophy, which I believe in. And then the chairs made another point, which I also believe in, which is that if you bind yourself too closely to any given policy, then sometimes you show up in a meeting having committed to a policy that's not actually the optimal policy for that meeting. And I would argue in the fall of 2021, when I said earlier, I wish we'd moved a little bit sooner with 2020 hindsight, part of why we didn't move sooner was we felt a little bound by the forward guidance we've given. So I agree with that too. How you land those two things so that you're given the optimal amount of guidance, but not binding yourself. That's, that's what I think is a good topic for conversation. [00:29:54] Steve Leisman: I'm going to pivot again. I think this is the third one here. There were a lot of bankers in the audience and I want to ask you about risk when it comes to financing. We talked about AI and the amount of financing and build out there. We didn't talk about private credit. When you look at the lending landscape, how much, where are the points of concern you have and how much concern do you have? [00:30:19] Tom Barkin: So, you know, the lending landscape has the banking system and the non-banking system. And we have a lot of transparency and visibility to the banking system and much less visibility to the non-banking system. So I'll say the part I have a lot of visibility into in the banking system, what I'm seeing and what I'm hearing is is quite reassuring. You know, consumer credit, you hear all the story of the K-shaped economy. You're not seeing significant increases that would worry you on the consumer side. Same on the commercial side. Books are actually relatively clean. Today is not a promise for tomorrow. Things can change. But I think there's a whole set of folks who lived through the Great Recession and don't really want to do that again. And people made different underwriting decisions as a result. And so that's the banking side. That doesn't mean bad things can't happen. But I think so far so good. You know, the non-banking side, you have a lot less transparency. But you do hear things that don't feel like things you should hear. And one of those things you hear is, you know, we underwrote better than others in the Great Recession. And so you can be confident we'll underwrite better now that maybe that's like things are always right till they're not. And the second is don't worry if we if we have losses that can be held by, you know, equity holders, you know, not depositors. And therefore the run risk is much less. And, you know, we've seen non-banks have run risk before. So again, sure, but I'm not sure I believe it. Now, I couldn't point you to a particular, you know, asset class, though it is certainly is worth watching all the lending that's going on in the AI sector. But I do, you know, I do feel nervous about that as a concept without any specifics on with that, you know, without the transparency to be able to point you to an asset class, have you sat down and tried to unpack [00:32:14] Steve Leisman: to unpack how some of these deals are structured in the way that they are kind of looped around and circular? [00:32:22] Tom Barkin: Yeah, not in the detail that I should or could, I guess. But, you know, again, it's always it's pretty hard to convince yourself that some of these big hyperscalers who are generating so much cash in their core businesses, you know, aren't good credits. And so, you know, and they have a lot of cash, you know, and so it's more who are the intermediaries in this world, who are taking on risks that you don't know about. And again, we saw a couple guys sort of stumble a few weeks ago, and you do wonder where in the intermediary sector is the biggest risk. [00:32:59] Steve Leisman: I saw exactly what you're talking about the other day, this private credit presentation, and they show the graph and it's this is the big slice of equity. And then there's this layer on top, which is the lending. And you're like, oh, that seems fine. You lose all your money and the credit gets paid back. But you look at it like, what am I missing here? And I don't know what it is. But [00:33:18] Tom Barkin: Well, the, you know, you had the issue with the the redemptions on these private equity development companies public. And of course, that seems to have passed. But when whenever you're redeeming past the cap and people say, oh, no, no, no. Just to make sure things are sound, I'll do more, you sort of worry about why they have to do that. You know, long term capital would be a good example of something that only had equity. But when it hit in 98, somebody had to step in. And so, you know, this issue of runs is not, doesn't really yield itself to analytics. It's emotional by its nature. And so analytically, you can convince yourself there's no reason to have a run and you still have one. And that's what you worry about. [00:34:01] Steve Leisman: Do you, are you concerned that there are deregulatory movements when it comes to Dodd-Frank and the levels [00:34:08] Tom Barkin: of capital that create risk? I think you always have to figure out how to balance, you know, the the risk of having a banking system that I'll call you can't afford to make certain types of loans and puts them into another vehicle and the risk of having a banking system that makes the wrong loans. And that's a, that's a delicate calibration. And, uh, you know, we've had various, uh, advice chairs who've tried to work out exactly where that sits. I don't have a better, better answer. I will just say it's a hard calibration. [00:34:40] Steve Leisman: Um, we're going to open it up to questions in just a couple of minutes. I got one more. There's five task forces. I can't name them all. Maybe you can. Do you have a favorite where you're most interested in what the, uh, uh, conclusions will be in places where you're most interested in what the, uh, recommendations are? [00:34:57] Tom Barkin: So five task forces, uh, one's on, one's on communication. We've already talked about that. Uh, one's on the balance sheet. That'll be an interesting, uh, conversation. Uh, one's on data and what are alternative sources of data. And, you know, the Richmond Fed and the Atlanta Fed are both the pioneers in the system in terms of alternative data that we can use. Um, one's on AI and productivity and what's it likely to do on productivity. Uh, and then one's on inflation and what's really driving inflation. And, uh, I should say, I love all my task forces equally, just like I love all my children. Um, but of course the one I'm most interested in is inflation. I mean, I, you know, we'll do, what do we do on the, but I think all that is, is tactical in its own way, but the fundamental of what we do is make policy to keep inflation under check. We've got a Nobel prize winner and a couple other incredibly well thought of economists, uh, working on this. They're going to bring, uh, I'm sure orthogonal views and challenge some of the views that we have. And I welcome that. And I don't know what they'll come out with, but, um, I don't think, uh, that any organization who's charged with keeping inflation in check at a time when inflation is not fully in check can afford to not pay deep attention to the best thinking you can get on the topic of inflation. [00:36:09] Steve Leisman: I used to hate asking this question because the answer is always everything, but I'm going to ask it anyway. Is, and the reason I'm asking it now is because there's so much out there. Is there a different way you personally, when you look at the inflation numbers, the fed has said core PCE, and I know you don't want to come off of that. Are there better ways to be measuring inflation in the economy that tell us more of what we want to know to guide policy? [00:36:39] Tom Barkin: PCE So, um, so I was in business for 30 years. Uh, I came here and the one thing I was convinced of when I came here was that the inflation numbers were fluey, um, and were fluey. They were not good. You know, I know what inflation looks like. It doesn't look like those numbers. And so I tried to dig into how they come up with inflation. It's actually pretty sophisticated. You know, there's a basket, there's 170 elements in the basket. Um, they do something called a hedonic adjustment, which is if you take, if you shrink the size, this candy bar, they actually adjust for the smaller candy bar. Um, uh, there's a study that you've read, I'm sure that Boskin did back in 2007, that sort of, you know, put it under the microscope and said, how can we do it better? It's, of course, not perfect, but that metrics pretty good. And I sort of feel, um, that when it comes to employment, I get advantaged information. If I can talk to a hundred of you and figure out whether you're in the process of planning layoffs or what your hiring plan is for the next year, when it comes to demand, I get advantaged information. If I can talk about, you know, what do you see last week or, you know, what's your forecast for the year? I think it's very hard to get better information on inflation. So the numbers we get, PCE is better than CPI because it, um, uh, adjusts dynamically as, you know, people's move their volume shift. That's good. Um, of course, I look at the distribution, you know, percent over 3% year over year, percent over 2%, um, to try to get a sense of how broad it is, but I haven't come up with a better way of looking at it. And that's a, maybe our data task force will become my favorite task force if it comes up with a better way to look [00:38:16] Steve Leisman: at it. Can I just supplement your answer? The hedonic adjustment is much cooler than the size [00:38:20] Tom Barkin: of a candy bar. Yeah. Okay. Hedonic adjustment is... I apologize for not making hedonic adjustments sound cooler. I think if I had said like, uh, if I had pronounced it like hedonism, maybe that would [00:38:31] Steve Leisman: have been... Right, hedonic. What it does, just so people know, is it says like, okay, we've added something to this product, but the keeping the product price the same, so should we reduce the price of the product because there's more to it? And it began the 1960s with automobiles. That's what they originally started doing when you added, um, automatic transmission, but kept the car price the same or lowered it. Um, and they now have a bigger basket of stuff that they hedonically adjust to try to figure out. But when they lower the price of the candy bar, it's hedonic adjustment the other way. It's [00:39:07] Tom Barkin: where you actually have inflation. And so part of the challenge is I go talk to new car dealers. I say, okay, new model year, what's your price increase? And they'll say it's 4% or 5%. And so you'd like to think that's 4 or 5% inflation, but it's not actually, because if they've added a new rear view camera feature or AI into the thing, there is an adjustment. It's a, it's a very thoughtful, though, of course, inevitably inaccurate kind of judgment on what the impact is. [00:39:32] Steve Leisman: It underscores how hard it is to figure out what's going on with pricing in the economy, because there's also prices and market prices and implicit prices. Let's, um, go from, from that, uh, rabbit hole to [00:39:42] Speaker 4: opening up to questions to the audience. We have microphones and if you can wait to, um, one come to you with a question. So please raise your hand. Okay, got one over here. [00:40:02] Speaker 5: Okay. Thank you. Hi. Uh, my question is actually on employment. My daughter graduated from college a year ago. Half her class is still unemployed. So it's actually more than a year. It's like 15, 18 months. Uh, younger people in this country definitely feel like there's an employment problem. What's your thoughts? [00:40:34] Tom Barkin: Um, and your daughter, is she employed? Yes. Well done. Um, if she wasn't, I, you know, but, uh, uh, so there's no question that, you know, big picture, what you're saying is absolutely, uh, true that you've got a generation of people graduate from college a year or two out of college, even out of some of the business schools, um, who are having trouble finding jobs compared to what they normally would have done. And this is usually a segment that's actually been, you know, quite employable. Um, you know, the, the technical attack, the sort of big picture reason why that's happening is that in the context of external uncertainty and AI uncertainty, the folks who generally hire those people have decided to cut back hiring, right? And it's the consulting firms and the finance firms and the, um, uh, accounting firms and the, the classic employers have decided to cut back. The same thing that's happening is that the place where that cutback has been most pronounced is in technology, right? Where technology employment is absolutely shrunk. And, and I said earlier that one of the big use cases that seems to create productivity is coding. Um, a shocking number to me is 20% of current college graduates are computer science majors. And when I was in college, which was a while ago, that number would have been one or two. And so that's a whole generation of people who trained up for this and people aren't hiring for this because, you know, okay, we're not going to fire coders, but we're just going to do more with what we've got. Now I'm told that sophomores in college today, uh, computer science is down under 10% as a major and people are moving toward engineering and other places that seem to have better employment prospects. And I think big picture, if you believe as I believe that, um, we're all going to have to become more technologically capable to survive in tomorrow's world, uh, I would bet on the 22 year olds over the 62 year olds, right? Um, in terms of getting that technologically competent. So I think there is an adjustment factor that's going on. Um, the companies aren't going to not hire forever. That will come back. In some cases, it already is. Um, and, uh, I think the world is going to move toward, we need people who are AI natives, technologically native to manage this new technology. And that will be those people, not, [00:42:53] Steve Leisman: not us, but that it's an adjustment process. I want to follow up and get to another question quickly. Is there a, a government response that makes sense to this? It strikes me that even the private sector is not good at predicting what the next jobs are going to be. Then you tell me government's going to retrain people. So they'll retrain them as coders. Well, guess what? You know, um, I mean, I told my son not to go into journalism and he's incredibly successful at it, having pieced three different jobs together. And he's in journalism. He's in journalism. I said, stay the hell away from this job. Um, because everything I knew was crumbling. Yeah. You know, and he now has a podcast, a book deal and a movie deal. Cool. Um, but, uh, um, I could have told him to go into coding and that would have been brilliant dad again. [00:43:46] Tom Barkin: Um, so I really do think there's this, there's a great book, uh, uh, called the race between education and technology, which has always spoken to me. And it makes the point that we had a farm economy. And part of how we, you know, handled the employment mismatch of moving from a farm economy to a manufacturing economy was the rise of K to 12 education, which really didn't exist at any scale, uh, in the farm economy. And then I said, we went to a information economy and the way we got from manufacturing economy to the information economy was the rise of, um, colleges, the GI bill and all that. And that's how we got that workforce and generationally retrained to another economy. And, you know, we can go all apocalyptic on this, but if you, if you argue, we're headed toward some third economy out there, I do think it's a really interesting question, whether the current course and speed of our, I'll call it work preparedness programs matches the skills we're going to need. So for example, um, we're all getting older. I mean, some of your exceptions, but, um, you know, in-home healthcare or hospital, that's going to be a huge need. Um, maybe we'll have half as many, uh, coders as we had last time, but if that's the case, can you really imagine that whole population deciding to be home healthcare workers? I don't know about that. I mean, and so, you know, college got us into the information age, but if it's a different set of things, is the answer more education? Is it different education? Is it, I think we, we haven't, you know, come across that. There's some interesting work people are trying to do on it. Um, but it is not at all obvious to me that if the economy we have in the future has a lot more, you know, use of your hands and a lot more, uh, use of your care and relationship skills, if the answer to that's more education, different education, or some [00:45:37] Steve Leisman: other lever. That's the thing with the data centers, right? You, you, you go over there and you build the data center and then the data center is built. Yeah. Is there a follow-up there? I don't know. Maybe they go on to the next data center, but that could be in another town, another state. Yeah. Let's go to another question. Did somebody else have one? [00:45:51] Speaker 6: Thank you so much. This is fascinating. So following up on your hedonic, uh, model discussion, [00:45:56] Steve Leisman: as well as sort of the automotive- I spent three hours in a hedonic adjustment seminar, so I wanted to get some use out of it, if you don't mind. And you, and you wonder why he wanted his son to get out of journalism. Exactly. I want to, I want to double click on that, [00:46:07] Speaker 6: actually. Um, so Gordon Moore, Moore's law, you know, number of chips and microprocessors doubles every, you know, couple of years. We've seen this for 70 years. Recently, we've seen AI models doubling every, call it four to five months. And we're seeing this really as an exponential curve. A couple of years ago, McKinsey estimated job churn roughly, you know, 10% over 10 years, so about 1% a year of jobs structurally changing. World Economic Forum a year or two ago, their future jobs report was estimating four or 5%. When you think about this, what I'll call skills inflation. So if all of this technology is advancing at a faster and faster pace, and you're seeing these capabilities increase, um, you, you talked about, you know, economic inflation, but I'm curious how you factor in from a hedonic standpoint, the need to keep pace with skill progression. Does, does that make [00:46:56] Steve Leisman: sense? I don't quite get it, but that was one of the most informative questions I've ever heard. That was full of facts. I really liked it. Why don't you restate it again? Uh, just the question part of it. [00:47:08] Speaker 6: So, so I would suggest that if the skill change is accelerating faster than it has before. So the, the skills that you need to be successful in the workplace, uh, there's an inflation happening there that just to keep pace with what you need to be, call it a competitive worker, is actually accelerating. So if people aren't keeping pace with that, then they're essentially losing value in the workplace. And so it kind of goes back to the education question. If higher education can't keep up with the changing demands of the workforce, call it AI skills, digital agility skills, things like that. I'm curious if you're thinking about that when you track inflation in terms of where, where this all settles out in terms of, uh, the different skills required to be competitive in the workplace. [00:47:54] Tom Barkin: I'll try that. So we don't generally think of inflation as being, um, workers. We think of inflation as being prices. And so to your technical question, we don't really, you know, uh, track it like that. I will say we do track wages and we track unemployment and the issue you're talking about feels to me actually more of a wages issue, more of an unemployment issue than a wages issue. I, it is easy for me to imagine, I'll just put it that way, that we will end up with this big mismatch of employers out there needing these highest level skills and not finding enough people. That would then lead to wage increases for that, for the people who are available in that sector. But at the same time, a bunch of workers out there without skills who can't find jobs, that would lead to wage decreases or more likely unemployment in that sector. And I think that is a very conceivable path for the, you know, economy. And that would be, uh, obviously, you know, obviously a challenge. Um, I'll just say one more thing, which is, um, you know, most of us, you know, grew up in business during the nineties and the nineties was sort of a special time, uh, as you think about the labor market, because, uh, you had the baby boomers fully working and fully productive, uh, you had better health outcomes. So people live longer and work longer. Uh, you had women finally, fully participating in the workplace. And that led to a lot more, uh, supply workers. You had the availability of labor, even offshore. Um, and, and I think a bunch of industries that used to own their labor market decided they no longer had to own it. I think at the banks, when I graduated from college, one of the greatest jobs you could get was working at one of the big banks. They all had two or three year training programs in the nineties. They all blew them up because there's so many extra bankers. Why do I need to spend two or three years on someone who's going to leave? So, and I think we got as a economy used to the notion that, um, the world would produce our workers, whether that was the colleges or the community colleges or the, um, or, or our competitors who were laying people off. And if you believe, as I believe that, um, the combination of demographics, fertility, immigration is leading to a much smaller workforce in the coming years. And you've got this mismatch, Rich, that you're talking about. I think your employer out there who's faced with inflation, wage inflation, because you can't find enough people with the skills is actually going to be faced with the challenge of investing enough to develop their own. And of course, keeping your own is an important part of developing your own, that it's going to change how people think about the workforce. I think that's, that's a highly likely, you know, path, at least. I mean, I've seen this over the years [00:50:31] Steve Leisman: that the, the businesses get used to the government essentially training your workers. And you do that through the, through the public school system, you do that through the community college system. And that's, and I don't have to do that. So you give me people who are trained, and I don't have to take responsibility for it. Maybe now they do. I'm worried about another aspect of it, which is a world where one of my favorite economists is Gregory Clark, who's an economic historian. He spent a lot of time in the basements of, um, textile mills and industrial revolution in England, um, finding out how much workers were paid and he wanted to see what happened when there were technological advancements. And he found that, and I think, you know, this, this is out there that over time, a dollar productivity goes 65 cents to labor and 35 cents to capital. But what about a world where capital doesn't need labor anymore? Capital keeps taking a bigger, bigger share of that chunk of productivity increase. That's a different world where, um, as you say, we don't need the coders. We need somebody who's gonna, I don't know, feed the dog. [00:51:34] Tom Barkin: Well, so, you know, economists would say, if you can't find enough people to feed the dog, raise the wages and you'll find enough people to feed the dog. And so one of the, there's no specialty in feeding the dog. No, but, but I mean, supply demand. So one of the interesting things we're going to play into, I think, is this question of if, if we do have this mismatch and we don't have enough people, we have tons of college grads who can't find jobs. Maybe they're comp adjusts over time. And if we can't find enough people, you know, to take care of our parents or ourselves, maybe those, that comp will increase over time. I don't know. But, you know, the, the answer to, you know, how we solve these things, the market's going to play some role in it. [00:52:11] Steve Leisman: I'm not a doomer on AI, but every other day I seem to wake up and worry about it more. Are there other additional questions out there? Yeah, right here, sir. [00:52:26] Speaker 7: Oh, you're next. So thank you for holding this. My question is a little bit of an adjunct of what Steve talked about with fiscal policy. So when you're sitting there thinking about monetary policy and you have these exogenous events that occur, the war, you know, you have fiscal policy that maybe is inflationary when you're facing challenges, combating it. When you sit around, think about monetary policy in the South, we say dadgummit sometimes, because you can't, you feel like your hands are tied or you're fighting something that, pardon me, you can't control. So when you, I'm particularly thinking about oil prices right now, because clearly you can raise rates and reduce demand, but it's not going to impact prices. And so then, you know, one of the things we talk about when we sit around our table and my companies, we talk about what if stagflation comes forward, because that's sort of the recipe we've seen historically happen. So talk a little bit about your views on that, that would be... [00:53:23] Tom Barkin: You want to talk about your, I'm just teasing. No, no. I mean, the, so, so we don't get to determine what happens in the world. We just get to deal with what comes out of it. And there will be times where oil prices are up and times where oil prices are down and times where you've got big inflationary forces and times where you have disinflation forces. I kind of think about it like sailing. It's a lot easier to sail when the wind's behind you and you put out the sheet, but sometimes you're sailing into the wind and you got to tighten up. And that's sort of how I think about it. I think there is, I don't think it's true that if you, I think it is probably true that if you raise rates, you won't immediately and directly affect oil prices. But I do think you'll affect demand. And in time, demand will affect prices of oil and other things. And that's sort of the theory of the case. If you've got these exogenous variables, you're always trying to tune to how much is ongoing and how much is, forgive me for using the word, I won't use the word temporary, the word that must not be named. And oil prices are a great example. They went up in April. It was a pretty rational thought, I think, even now looking back to say, this looked like a three week or six week or eight week kind of thing. You get to June or July, and you say, well, it's still sort of hanging out here. But by the way, oil prices came down a lot in June and early July. So by the time, at least for me, when I got to September, I'm sort of like, okay, this has been six months. That's inflationary. You sort of have to start leaning against it. Those are the sort of thoughts that come in. And the more the external uncertainty, things could go away for things that don't control, maybe make you a little humbler about trying to oversteer. But you can't abdicate steering. You just have to maybe be a little humbler. [00:55:12] Speaker 8: One more question. The prediction markets were pretty good during the midterms. You mentioned alternative data. Is that an area that economists are increasingly looking towards? People definitely are looking at prediction markets. I guess I have to repeat that, you know, [00:55:34] Tom Barkin: economic forecasting was invented to make weather forecasting look good. I mean, it's just, it's, you know, you have to be a little careful about just, by the way, financial markets are already out there as prediction markets. And so, you know, we absolutely, and I absolutely do pay attention to those things that maybe I'll learn something, but it's all kind of early. [00:55:54] Steve Leisman: I worry about it, Tom, when it comes to these prediction markets, the concept, I think, underlying your question is, is there a wisdom of crowds? Yeah. But the wisdom of crowds idea, I think, comes from people who have something on the line with the outcome. But if I'm just a punter, as the Brits would say, and I put 10 bucks on the outcome of Texas, it doesn't matter to me which way it happens, right? Other than my bet. So I just wonder how much information is in those prediction markets. [00:56:22] Tom Barkin: Well, and again, we have financial markets, so you can go check and see what the odds are that we're going to have a rate increase at the next meeting or the one after that or the one after that. There's a lot of people with their own money on the line who are making those bets. I don't know if they're right or wrong, but that's probably a better prediction market than, you know, a bunch of, would you say punters? Punters. Than a bunch of punters. But I mean, we'll figure these things out over time. We'll figure it out. [00:56:45] Steve Leisman: It was a hundred million bet on the Texas Senate raise. I was just, it was amazing. [00:56:49] Speaker ?: Really? [00:56:49] Steve Leisman: Yeah, it was an enormous amounts of money. Wow. Please join me in thanking Tom for... [00:56:53] Tom Barkin: And thank you, Steve. My pleasure. [00:57:00] Speaker 9: So since I ended up with a microphone, I'm just going to stay right here. Thank you so much to Steve and Tom. I think they set us up brilliantly for the conversation over the next day and a half. There is one logistical item that I want to point to before we end, we, we moved to a break. We are using CVEN events app for this event. There are QR codes all over the place. There's one there. They're on your printed thing. They're on the tables. This is important because with the excitement,