Hottest Inflation Report In 3 Years Has One Big Problem

Ed Elson speaks with Justin Wolfers about how concerning the latest inflation data really is and what it could mean for the Federal Reserve. Then, Stacy Rasgon joins the show to explain the record setting rally in chip stocks. Finally, Ed breaks down reports that some Amazon employees are exaggerating their AI usage. Justin Wolfers is a Professor of Economics and Public Policy at the University of Michigan and founder of Platypus Economics. Stacy Rasgon is a Managing Director and Senior Analyst at Bernstein Research. Get your tickets to the Prof G Markets tour Subscribe to the Prof G Markets Youtube Channel Check out our latest Prof G Markets newsletter Follow Prof G Markets on Instagram Follow Ed on Instagram, X and Substack Follow Scott on Instagram Send us your questions or comments by emailing [email protected] Learn more about your ad choices. Visit podcastchoices.com/adchoices

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  Flowing ad budget on metrics that look great till the CFO sees them? That's bull spend. And marketers are calling it out in Dashboard Confessions.

• 0:09Speaker 01

  I remember telling my boss, "It'll be good for the brand," when leads were slow. Yeah, it wasn't.

• 0:17Speaker 00

  Cut the bull spend. LinkedIn lets you target by company, job title, and more. Advertise on LinkedIn. Spend $250 on your first campaign and get a $250 credit. Go to LinkedIn.com/campaign. Terms and conditions apply.

• 0:30Speaker 02

  Energía baja. Falta de concentración? Nebla mental? You might be dehydrated. Gatorade Zero Powders han sido diseñados científicamente para ayudarte a mejorar el balance de hidratación en el cuerpo. Hechos con la mezcla número uno de electrolitos, sin colores ni sabores artificiales. Shop now online or at retailers nationwide. Gatorade: Hydrates better than water. Is it in you? Signos de dehidratación leve y moderada. Consulta a tu médico si los síntomas persisten.

• 1:00Speaker 03

  Welcome to Sephora. I'm looking for a perfume that's not too perfumey? I got you.

• 1:06Speaker 04

  Serum moisturizer or moisturizer serum?

• 1:09Speaker 03

  Let's get into layering. My concealer is making me look worse? Sounds like the wrong shade. Let's get you matched.

• 1:18Speaker 05

  There's only one store that really gets what you're going for. Get beauty from people who get beauty. Only at Sephora.

• 1:24Speaker 03

  Hi, I, uh... Let's get you a basket.

• 1:29Ed Elson?

  Today's number? Thirty-nine. That's the percentage of Republicans who think Trump would beat them in a physical fight. That is according to a new poll from research firm YouGov. And in other news, celebrations at YouGov after winning the Pulitzer Prize for best survey question ever.

• 2:01Ed Elson?

  Welcome to Prof you Markets. I'm Ed Elson. It is May 13th. Let's check in on yesterday's market vitals. The S and P and the Nasdaq fell after the latest inflation reading came in higher than expected. We'll talk about that shortly. Oil prices continued to rise as tensions between Iran and the US remained elevated. Treasury yields also climbed. And finally, the Senate confirmed Kevin Walsh as Fed Governor, clearing a major hurdle toward becoming Fed Chair. Okay, what else is happening? Inflation soared in April, up 3.8% from a year earlier. That is the highest inflation reading in three years, and even hotter than economists

• 2:44Ed Elson?

  had expected. Energy accounted for 40% of the month to month increase and was up 18% from a year earlier. But price pressures are spreading beyond fuel. Core inflation, which excludes food and energy prices, climbed 2.8% from a year ago. And for the first time in three years, wages are no longer keeping up with prices, which essentially means that real wages are falling. So here to discuss these numbers and the state of the economy, we're speaking with our friend, Justin Wolfers, Professor of Economics and Public Policy at the University of Michigan and founder of Platypus Economics, his new YouTube channel,

• 3:24Ed Elson?

  Substack, and podcast. Justin, welcome back to the show. Let's get right into these numbers. Inflation up 3.8% year over year in April. Core inflation up 2.8%. This seems to me to be the worst reading we've seen in a long time. I guess the question is, how bad is it?

• 3:45Justin Wolfers?

  Ed, just hearing you go through those numbers caused the worry lines to itch a little deeper right now. And I was in a good mood before I started talking to you. Two things are true. This is a high inflation read. The cost of living is rising. It's rising sharply. People are finding it hard to keep up. The affordability crisis is real. The second thing's true is none of this is yet.

• 4:14Justin Wolfers?

  Deeply problematic. So let me back up and say and point that out. There's two inflation measures you just pointed to, Ed. One was headline inflation. That answers a specific question. If I care about the cost of living, what happened to the cost of living? That's it. It rose. And so if you feel like you're not keeping up, the truth is, when you're paying four bucks a gallon, you're not. Second question is, what's going on with the future of inflation? And that's where economists tend to look at core inflation. We exclude food and energy. Food goes up and down with droughts and avian flu and stuff like that. Energy goes up and down with strife in the Middle East. And so neither really capture the underlying inflationary momentum or the

• 4:57Justin Wolfers?

  inflationary psychology. And that at 2.8% is lightly uncomfortable, but not miserable. And a question, and perhaps the question over the next few months is, how much will we be able to keep this as an energy crisis, Iran temporary story? Remember, there's still some tariff business going on, versus how much does it gain its own momentum, start to feed its way into inflation expectations, and become something substantially more to worry about? Because getting inflation down from 3% to 2% would require slowing the economy quite substantially.

• 5:36Ed Elson?

  So this gets to the heart of the problem, which seems to be that it's really, I mean, when you look at the numbers, it's mostly a fuel problem, it's mostly an energy problem, or at least 40% of the problem was was the cost of of energy, which is because of the fact that the Strait of Hormuz is closed, because we're at war with Iran, which leads to the simple question of, when will that be over? And this was asked in so many words to the president yesterday. He was asked whether the financial pressure facing Americans was going to influence his decision making in Iran. And his response was striking. I just want to play you this clip and get your reaction.

• 6:18Speaker 08

  Not even a little bit. The only thing that matters when I'm talking about Iran, they can't have a nuclear weapon. I don't think about American financial situation. I don't think about anybody. I think about one thing. We cannot let Iran have a nuclear weapon. That's all. That's the only thing that matters.

• 6:37Justin Wolfers?

  Your reactions to that clip. Actually, I thought it was going to be much worse. So, what a relief.

• 6:44Justin Wolfers?

  Look, let me start by agreeing with the president.

• 6:48Justin Wolfers?

  I think it's probably immoral to go to war in order to deliver low gas prices. I don't want him to walk around doing that. I think the stakes of war are so much larger, they're human.

• 7:00Justin Wolfers?

  Gas prices is narrow, short term, small and selfish. What I noted, though, is the mission changed, which is originally we're going in there to free the Iranians from the yoke of oppression. That all we wanted was regime change. And we did, in fact, get rid of Ayatollah Khomeini, which is why we now have instead Ayatollah Khomeini. I understand there's a vast difference between the two. And so, scoring this war is so damn difficult because the president refuses to say why we're there. Having said that it's okay for him not to be laser focused on what this does to the cost of living for Americans, let me come back and give the other side of that.

• 7:44Justin Wolfers?

  You know, partly because I do think the stakes, if we're talking about the freedom of 90 million people, or the possibility a crazy regime may bomb us to smithereens, those stakes are just enormous. And I'm happy to pay a little more at the pump if those are what we achieve. I want to be clear, so far we've achieved nothing, right? So, we, at the start of the war, the Strait of Amos was open, we had Ayatollah Khomeini and the Iranians had nuclear material. The Strait of Amos is closed, they have Ayatollah Khomeini, and they still have nuclear material. So, whatever the, I don't feel any sense of why my life's better here.

• 8:28Justin Wolfers?

  I think the reason people have a right to resent this is the president never tried to make the case to the American people, here is an important and noble goal from which I'm going to ask sacrifice. He seems constitutionally incapable of asking for sacrifice. Sacrifice is a thing that populations do when they're considering war. That's an adult, grown up, mature conversation, just like saying, I'm going to start a tariff war and prices aren't going to rise, is an immature lie. You might think there's a serious argument to be had about defending American manufacturing or some argument, it's not my argument. And there'll be some costs and there'll be some benefits, but I think this leads to a better world. That's a mature conversation. Um, not even having the conversation before

• 9:12Justin Wolfers?

  putting Americans' prosperity, livelihoods, and in some case, children at risk. I, it's not my wife doing things, man.

• 9:24Ed Elson?

  When we think about what's happening with inflation here, and you mentioned the difference between core inflation versus overall inflation. We strip out food, we strip out energy because it's volatile. You know, it can skew the data in a lot of ways. And, and Trump has kind of made this argument where he, he says that the, the increase in oil prices and fuel prices is fake inflation. Like it's, it's, it's a different thing. And on the one hand, you can kind of make that argument, but on the other hand, it's like, we're still at war here and there's no indication that this is going to end anytime soon. And in fact, the, the more that they keep on telling us a deal is about to happen, we're about to get somewhere, there's going to be a ceasefire, and then suddenly it doesn't

• 10:09Ed Elson?

  happen. It happens over and over and over again. It starts to become kind of clear, at least to me, that this is going to last for a very, very long time. And it starts to look a lot like Iraq. It starts to look a lot like Afghanistan, where they said it was going to be a couple of weeks, it was going to be a couple of months, and then it ended up being eight years, or it ended up being two decades. And so it seems as though a very similar situation could play out here, where we're telling ourselves that it's transitory. Like we said a few years ago, we're telling ourselves, oh, it's, it's just for now, it's fake inflation. It's all, it's just, it's just for now. It's just the fuel, because the straight up moves is just blocked up for the time being. I, I just, I'm not

• 10:50Ed Elson?

  sure I believe that anymore. And it starts to seem as though the 3.8 number could be something that lasts for a very long time. And I wonder if that would have a structural effect on the economy and potentially on markets as well.

• 11:04Justin Wolfers?

  And I have a memory that at the very start of the war, I came on your show and I said, everything that economic theory and history tells me is that there's a very large risk. This takes a lot longer than four to five weeks and almost no chance it takes less. And that there's a long history of excess optimism at the beginning of wars. And that a more realistic assessment should be that it was likely to be far longer with a far greater economic cost. Um, I take no great pleasure in being right. I just want to remind you I was. Um, the food and energy, let me say two things about it. Um, one, you know, we take it out for the reasons I described earlier, which is if you want to, we leave

• 11:49Justin Wolfers?

  it in for the reasons I described. If you want to describe the cost of living, it's gone up. Americans are sacrificing. If you want to know what's happening to inflationary psychology, we take it out. But there's a very real risk. It spills over and becomes part of the process. And that's very much what's on the Fed's mind right now. That in our models, people shouldn't think about inflation, should think about inflation coming from energy differently from other things. From reality, they may not, particularly after several years in a row of inflation being above the Fed's target. And finally, this wasn't the question you asked, but I do want to put it in there. Um, food and energy is something we look through as part of a statistical exercise. I want to predict inflation, what are the most persistent aspects, blah, blah, blah. But food and energy are

• 12:34Justin Wolfers?

  also a very large part of the consumption basket of the world's poor. And so the World Bank and the IMF have pointed out, if you wanted to think about who's really going to be hurt by high food and energy prices, it's actually the most vulnerable in the world. And again, if we could have a mature, responsible conversation about whether or not to go to war, part of that, one hopes the American conscience cares somewhat about hunger and poverty beyond our shores. And the consequences there are very large. And you and I so far have been talking about the United States economy, and it's worth remembering that the US economy is actually much less affected than many of our allies,

• 13:18Justin Wolfers?

  and that Europe and Asia are feeling much, much, much more pain than we are. And, um, those folks, even more than the American people, wouldn't have asked their feelings about this either.

• 13:30Ed Elson?

  Yeah, it seems as though this has to come back to bite us in a, in a very real way at some point. And so far, the markets have been extremely resilient.

• 13:40Justin Wolfers?

  I disagree with that. So let's pick that up.

• 13:43Ed Elson?

  Please, yeah.

• 13:44Justin Wolfers?

  Um, so look, here's the glib answer. The glib answer is the stock market's at an all time high, therefore the market thinks the war is fine. That's glib because markets reflect lots of things. So let's just think about it in a more focused way. Let's say that the market's partly responding to the war, partly the AI boom and partly other stuff.

• 14:04Justin Wolfers?

  And if what you want to do is take the market's response as a vote of confidence in the war, then what you need to do is isolate the part that's due to the war. How might we do that? Well, one answer is we can look every single time that the president has been more belligerent and pushed into the war. Markets have fallen, literally as he speaks. Every single time he steps back and looks like he might taco, markets rise. So that's markets saying they believe, they're acting as if they believe that Americans, large companies will be more profitable over the longer run if the president were to step back from the conflict rather than stepping into the conflict. You could actually go even a step further and try and quantify this.

• 14:48Justin Wolfers?

  So if only we had some index of trouble being caused by the Middle East, then what we do is we'd see how that index was correlated with the stock market. I don't have the perfect index, but I've got a pretty good one, which is oil prices. All the movements up and down in oil prices are about, is there more trouble, more strife? Are we moving towards war or not?

• 15:11Justin Wolfers?

  And it turns out on a day in which oil prices rise 10%, the S and P 500 falls by 0.9% on average. You can just draw a scatter plot and you'll find that. You can get fancy and call it a regression coefficient. And so that tells you there's very strong statistically significant evidence that whatever it is that's driving oil prices up and down, and we know what it is, has big negative effects on US stocks. We also know that the oil price is up about 60 or 70% over the course of the war. And so if you want to extrapolate my point, negative 0.9 coefficient, that says you take the 70% rise in oil prices multiplied by minus 0.9 and you can say that US stocks are about 6% lower than they would otherwise be, 6.3%.

• 15:56Justin Wolfers?

  As a result of this war, and then I'm just going to be conservative and round that down to 5%. So I think this analysis, very back of the envelope, but we're in the middle of a war, no one's giving you strong answers, you know, I'm not going to defend it till the day I die, but I think it's informative. This says that US stocks are five percentage points lower as a result of the president's aggression so far.

• 16:19Ed Elson?

  Because we have all of these other things that are that are buoying the economy and that are we're all very excited about because earnings have been very strong. We do have the AI build out occurring as planned. We'll see what happens. But we got a lot of a lot of other other stuff going on, which is great, which is, I think, a lot of the resiliency is reflected in that. But it just does seem that, I mean, we, at the same time, two thirds of the economy is consumer spending. The consumer spending has thus far been somewhat resilient. And I guess we're just playing this waiting game at this point of hoping this thing will end next week or next month. And that's essentially what we are what we are betting on and what we're hoping for. And I don't know, I just,

• 17:03Ed Elson?

  I mean, you hear it in my voice, this, this, this report makes me quite anxious. I mean, just as we wrap up here, the Senate confirmed Kevin Walsh as the Fed governor yesterday. This, obviously, all of this inflation data is going to be very important for his decision as to what we do in terms of interest rates. Will we cut them as the president has advocated for? Doesn't seem like a great idea now. Just as we close here, what are your thoughts on on Walsh's appointment? And what do you think the path is going forward for the Federal Reserve?

• 17:36Justin Wolfers?

  So the most important question is, who is Kevin Walsh?

• 17:39Ed Elson?

  Yeah.

• 17:41Justin Wolfers?

  So from 2006 to 2011, Kevin Walsh was a Fed governor. And so we know a lot about him. That was the period in which we had the global financial crisis. The global economy created, the US economy created, unemployment soared, inflation was low, and we risked deflation.

• 17:59Justin Wolfers?

  And throughout that entire recovery period, Walsh kept saying, I'm worried inflation's going to rise. I think we should have higher interest rates. So two things you could learn from that. One, that was a spectacularly bad judgment. It's almost certainly the case that the US economy would have been better with a more aggressive response to that particular crisis. And the second is that that Kevin Walsh, the 2006 to 2011 Kevin Walsh was an inflation hawk. He saw inflation around every rock through that period, was terrified of it and wanted a forceful Fed response to it, even though we were risking deflation more than inflation.

• 18:40Ed Elson?

  Yeah, the opposite. Yeah.

• 18:41Justin Wolfers?

  There's a different Kevin Walsh. That Kevin Walsh was at a Senate confirmation hearing and said, I believe in Fed independence and was asked by Fed independence, of course, means I'm willing to serve the interests of the American people, even if it hurts the president's feelings. This Kevin Walsh was asked by Elizabeth Warren whether the president lost the 2020 election and thought it's so important not to hurt the president's feelings that he evaded the question altogether. That Kevin Walsh sounds like a sock puppet and a sock puppet who'd follow the president's desires, which would be to cut interest rates dramatically at a moment that the economy doesn't call for it. Hey, man, there's two Kevin washes. One of them wants really high interest rates. The other one wants really

• 19:22Justin Wolfers?

  low interest rates. I'm glad I'm not trying to, living inside his head. And we're going to find out in the next few weeks and months who Kevin Walsh is.

• 19:31Ed Elson?

  Justin Wolf is a professor of economics and public policy at the University of Michigan and founder of Plastibus Economics, his new YouTube channel, which I encourage you to go check out. It's also a Substack and a podcast. Justin, always appreciate it. Thank you.

• 19:45Justin Wolfers?

  Great pleasure, Ed.

• 19:47Ed Elson?

  After the break, why chip stocks are on a tear. And by the way, we are heading out on tour at the end of the month. So for more info and to get tickets to a show near you, head to profitymarketstour.com.

• 20:04Speaker 00

  Flowing ad budget on metrics that look great till the CFO sees them? That's bull spend and marketers are calling it out in Dashboard Confessions.

• 20:14Speaker 01

  I remember telling my boss, "It'll be good for the brand." When leads were slow. Yeah, it wasn't.

• 20:21Speaker 00

  Cut the bull spend. LinkedIn lets you target by company, job title, and more. Advertise on LinkedIn. Spend $250 on your first campaign and get a $250 credit. Go to LinkedIn.com/campaign. Terms and conditions apply.

• 20:35Speaker 02

  Energía baja, falta de concentración, nebla mental? You might be dehydrated. Gatorade Zero Powders han sido diseñados científicamente para ayudarte a mejorar el balance de hidratación en el cuerpo. Hechos con la mezcla número uno de electrolitos, sin colores ni sabores artificiales. Shop now online or at retailers nationwide. Gatorade: Hydrates Better Than Water. Is it in you? Signos de dehidratación leve y moderada. Consulta a tu médico si los síndomas persisten.

• 21:04Speaker 03

  Welcome to Sephora. I'm looking for a perfume that's not too perfumey? I got you.

• 21:10Speaker 04

  Serum moisturizer or moisturizer serum?

• 21:14Speaker 03

  Let's get into layering. My concealer is making me look worse? Sounds like the wrong shade. Let's get you matched.

• 21:22Speaker 05

  There's only one store that really gets what you're going for. Get beauty from people who get beauty. Only at Sephora.

• 21:29Speaker 03

  Hi, I, uh... Let's get you a basket.

• 21:38Ed Elson?

  We're back with Prof G Markets. Chip stocks just posted their biggest run on record. Over the past six weeks, S&P 500 semiconductor companies have added nearly $4 trillion in market cap. Chip stocks now account for almost a fifth of the entire S&P. And three companies, TSMC, Samsung, and SK Hynix, account for more than a quarter of the entire MSCI Emerging Markets Index. Meanwhile, an index of semiconductor companies hit a record high on Monday and is up 65%. Year to date, however, the index did fall 3% yesterday as traders took profits following all time highs. So here to help us break down what's happening in the chips

• 22:20Ed Elson?

  market and what is driving this rally, we're speaking with Stacy Rasgon, Managing Director and Senior Analyst at Bernstein Research. Stacy, thank you for joining us. Just some unbelievable numbers here. SanDisk up 50% in the past month, Micron up 78%, Intel 80%, AMD 80%. I mean, all of these chip stocks are just tearing right now. And I guess the question is, why is it happening right now as opposed to say a year ago when Nvidia was also tearing?

• 22:57Speaker 09

  Yeah, and to be fair, we've had a number of names over the last year or two or three, you know, move like this, but we haven't seen the whole group broadly move like this. And I think it's a few things. First, there's been a real appetite to play other names, I think, in the AI supply chain. And frankly, AI is now getting so big that it's dragging everything along with it, right? You know, it started with the GPUs and the accelerators, you know, a year or two years ago, whatever. And then, one at a time, different other end markets that are more peripheral to it, but important for it, have come into focus. I mean, we went from the GPUs and accelerators, and then we went to memory, and then semi-cap, and then optical, and then power semiconductors, and more recently CPUs. And so, one at a

• 23:42Speaker 09

  time, as demand has grown and grown and grown, it sucked a lot of these other things into it and driven unexpected upside. These were not areas that investors broadly were looking at for upside from the AI trade. So I think that's one. Secondly, and maybe explain some of the recent rampant, to your point, I mean, I think the SOX index, which is a broad index of semiconductor stocks, is up, at least as of Friday, I'm not sure how it ends up today, but as of Friday, was up 66% year to date, and was up almost 50% since the beginning of March, when earnings season kicks off, kicked off. Earnings estimates have actually exploded to the upside and if you look at

• 24:24Speaker 09

  the interplay between earnings and multiples year to date, multiples, valuations in the industry year to date are actually down slightly. All of the growth in the stocks has been explainable via growth in earnings estimates. We've actually had a just a huge cycle of positive revisions in earnings across the board, probably most pronounced in memory. But we've seen earnings go up like pretty much across the board and I think finally there's been a real scramble. Because not all investors maybe have been involved in. Frankly, if you haven't been there, given the magnitude of the move, like you're starting to get with into the realm of potential career risk. Right?

• 25:08Speaker 09

  If you haven't been invested in these things and so I think there is a lot more interest and attention being placed on the space now and that may be driving some of this. And maybe a fourth thing which at the margin probably drives volatility, I don't know how much of the upside it drives, but I mean you have a lot more retail involvement as well. And so that probably does drive a little more volatility and maybe more willingness to buy some of the narratives over like some of the other fundamentals that are there. So I think it's a combination of all those four things that have taken the stocks up, but they've really been on a tear.

• 25:41Ed Elson?

  Yeah, on the one hand, it almost looks like it's just expectation, speculation, momentum. I mean, people talked about GPUs and then suddenly CPUs are hot and now memory chips are hot. And suddenly everyone decides, oh, we're very excited. But to your point, I mean, if the fundamentals are there, if the earnings growth is there and if on an earnings multiple, it's actually gone down for a lot of these names, then maybe it's not too much of a concern and maybe that means that there actually is still more room to run despite the fact that we've seen these ridiculous charts.

• 26:15Speaker 09

  Yes, I mean, like there's a relevant debate on like how sustainable are these earnings? Right, so that's, let me take memory just an example because that's where I think we're seeing the more extreme behavior and I don't cover the memory space, it's a colleague of mine, so I will refrain from discussing investment conclusions on these. But just on a factual basis, take a company like a SanDisk. They just guided to an earnings next quarter that is higher for that quarter than the entire stock price was, you know, 18 months ago when they went public. I'm gonna make up the numbers, but it was, you know, they guided, I can't, it was something like $32 a share and it was, it went public at like 30 bucks or whatever it was 18 months ago, so. On that basis, 18 months, it was trading at like, you know, .25 like price to forward earnings.

• 27:00Speaker 09

  I mean, it's just been amazing how much, and memory is its own special case, right? But, but you have that. You've also had multiples in memories come down, actually probably the most of everything. The multiples are probably down 20% year to date on an average basis. Some of them are more, while earnings estimates have, I don't know, quadrupled or quintupled, something crazy. And the reason is, investors, you get worried when, when numbers go up that much, you worry that are we close to peak? Right? And if you're worried close, that you're close, you're going to pay less for that earnings. And so that's some of the dynamic, I think, not just in memory, but I think broadly we've been seeing. And I think that is where the debate is, is, are these earnings sustainable?

• 27:41Speaker 09

  The level of AI capex that we've seen, you know, from hyperscalers, we're, you know, we're getting not that far off of a trillion dollars this, this year, are those levels of sustainable? How much more can that grow? These are the kinds of conversations I think that investors are having as they're trying to evaluate whether or not that the stock still makes, makes sense to be in or, or not. But, but for now, the fundamentals have actually been, been fine, like as earnings have been growing and at least from a demand standpoint, at least within the AI space, it doesn't look like we're seeing any signs of slowing at this point. We can talk about some of the other end markets, but I mean, AI is, is really the driving force behind everything, right? And by the, in and out of semis, you know, if AI demand ever does roll over, like we're all screwed.

• 28:25Speaker 09

  In semis and out of semis, it almost feels like even in the broader economy, it's the data center spending and AI spending that really is supporting everything right now.

• 28:35Ed Elson?

  The sustainability of earnings seems to be such an essential point here because we are seeing a lot of people pointing to the earnings saying, okay, everything's fine. But at the same time, I mean, we're witnessing what looks to be a little like a little bit of a gold rush, but for data centers. And that is the reason that all these stocks and all these or their products are in demand is because everyone's trying to build a data center right now. But I guess the question is like, are we going to be building data center, this many data centers perpetually into the future? The answer seems to be logically speaking, just no. I mean, once you've built the data center, it's built. But what do you think?

• 29:16Speaker 09

  Well, I mean, it depends on how much compute you need. And so that this gets now to the usage and and the return on these assets. And again, it's still early days, right? And and the amount of revenue that is getting generated from these assets is still probably relatively small. Most of the spending we've seen so far has been used, they built these massive compute clusters to train these models. I don't know how much your listeners know about AI models, but you can sort of think about like a massive amount of data and parameters that have to be set to create a model that can perform a task. And so most of the compute spending so far has been to train these models. So what we need to see is is these models actually get used for for something.

• 30:00Ed Elson?

  Yes.

• 30:01Speaker 09

  This is actually called inference, right? And it's still early days, but we're already seeing evidence. You can take a company like an Anthropic, for example, who discloses like their their annualized like run rate of revenue. And Anthropic makes makes something that's called Claude that's used for, you know, for agentic coding, for example. This is where you can actually use the AI model to help you code. And I mean, their revenue has gone vertical. I mean, as of the most recent announcement, which was, I think, a week ago, maybe, there's something like forty four, forty five billion dollars of revenue annualized. In April, I think they, I can't even remember, they were at thirty billion or something. And in like January, they were at fourteen billion and in December, they were at nine billion. And a year ago, they were, I don't know, one billion, maybe even less.

• 30:46Speaker 09

  So, I mean, and, and this is probably a, a good example of an application where customers are absolutely willing to pay for it and their revenues have gone vertical. Right? So the question of whether or not we need to keep building all this data center or either not as much or much more, comes down to how much computer are we all going to use? Like, how many tokens are, are, are, are people going to use? And I, I don't know the answer yet, but right now, it's, it's, it's, it's going literally vertical. So like that, that is, is something I guess we'll have to see as we go forward. But as of right now, uh, the, the only thing we're hearing from any of these folks is that they don't have enough compute.

• 31:23Ed Elson?

  Just to wrap up here, some investment conclusions. I mean, it seems to me that you, you shouldn't necessarily bet the farm on, on semis, but it doesn't seem like a good idea to not be invested in semis. I mean, whatever happens in terms of AI, I mean, what are the, the investment conclusions uh, at these prices?

• 31:44Speaker 09

  Like, like I said, semis and, and AI spending in general are kind of driving almost everything in the industry and kind of everything in the economy right, right now. And, um, and, and it's understandable because right now you're building out the infrastructure and these are key components of, of the infrastructure. Um, and you, you know, and, and I, I think you can look at the returns we've seen so far. And again, the, the question is, is it sustainable? If it's sustainable, I don't, I don't think that the performance that we've seen at this point has been, has been crazy. And you can decide where, where, where you want to play if you're in, in semis. Again, we, we've liked, uh, you know, we've, we've liked the, you know, the, the accelerator names, the Nvidia's and the broad comes of the world. I'd like semi cap, uh, the, the, the guys that actually make the equipment that make semiconductor. If all of this demand is really going to be true, you know, we're, we're, we need more, more chips

• 32:28Speaker 09

  and more wafers and that requires more equipment. So these are the kinds of things that, that I, that I've liked. And again, what, what is always in a gold rush? You know, it's, it's always the, the guys that are making the picks and the shovels that are, that are making the money. So all these guys are making the picks and the shovels. Um, and then over the long term, we'll see what the usage looks like.

• 32:44Ed Elson?

  Stacy Rasgon, MD and Senior Analyst at Bernstein Research. Stacy, really appreciate your time. Thank you.

• 32:50Speaker 09

  Yeah, you bet.

• 32:55Ed Elson?

  The Financial Times revealed something striking yesterday about the state of AI. And that is that Amazon employees are faking their usage of AI in order to not get in trouble at work. And that might sound a little ridiculous, so let me explain. Earlier this year, Amazon made it its mission to adopt AI in the workplace as much as possible. Their target was for 80% of developers to be using AI every week, and they even created an AI leaderboard, which tracked who was using AI the most at the company. Now, the result was that employees did start using AI more, but, as the FT reported, they weren't using it to get actual work done.

• 33:39Ed Elson?

  Instead, they were using it for meaningless tasks with the ultimate goal of showing their managers that they are indeed adopting AI. One employee said that workers at the company are simply using AI to, quote, "maximize their token usage." And another said that it has created, quote, "pervers incentives" and that people are, quote, "very competitive" about it. By the way, a similar dynamic has cropped up at Meta, which is also using AI leaderboards, and employees are now engaging in what is known as "token maxing," which essentially means using as many AI tokens as possible to rise up the ranks of the AI leaderboard and impress your boss.

• 34:23Ed Elson?

  There is an economic term for this known as Goodhart's Law. It's named after British economist Charles Goodhart, who said that, quote, "When a measure becomes a target, it ceases to be a good measure." And in this case, that measure is AI adoption in the workplace. But unlike most measures, this one has become the underpinning of the entire trajectory of the stock market. And so what these anecdotes at Meta and at Amazon are telling us is that this measure is probably no good. Yes, we are buying chips, and yes, we are using AI. But in this case, that usage isn't actually real. And if it isn't real, well, then the question becomes,

• 35:06Ed Elson?

  is the demand for AI actually sustainable? To which the answer is maybe, probably, no.

• 35:14Ed Elson?

  Now, to be clear, does this mean that all of the AI usage is fake or forced or contrived? No, there are still hundreds of millions of people who choose to use AI every day because they do find it genuinely useful. But the question is, how much of it is fake or contrived? We don't yet know the answer to that question. But when we do, rest assured, it will either make this market or break this market. Because as we've said before, the economy and the stock market are now a giant bet on AI.

• 35:51Ed Elson?

  Okay, that's it for today. This episode was produced by Claire Miller and Alison Weiss and engineered by Benjamin Spencer. Our video editor is Brad Williams. Our research team is Dan Shalan, Isabella Kinsel, Chris Nodinehue, and Mia Silverio. And our social producer is Jake MacPherson. Thank you for listening to Prof G Markets from Prof G Media. If you liked what you heard, give us a follow. I'm Ed Elson. I will see you tomorrow.