Paul Krugman Podcast

Talking With Azeem Azhar

47 min · 13. juni 2026
episode Talking With Azeem Azhar cover

Beskrivelse

I last spoke with Azeem, the proprietor of Exponential View, 18 months ago — ancient history on this subject. So we revisited the state of AI. . TRANSCRIPT: Paul Krugman in Conversation with Azeem Azhar (recorded 6/12/26) Paul Krugman: Hi everyone. Paul Krugman back on my usual schedule of recording interviews. And today I’m talking with Azeem Azhar [https://substack.com/@exponentialview], who I spoke to in January 2025, basically centuries ago in AI time. And with AI on everybody’s mind, I thought it would be good to revisit. I should say Azeem is an independent researcher and founder of Exponential View [https://substack.com/@exponentialview], which is one of the top tech Substacks out there. So hi, welcome to another conversation. Azeem Azhar: Yeah, thank you, Paul. And it has been eighteen months, also known as one and a half centuries in AI time since we spoke. Krugman: Yeah. Let me ask sort of the dumbest question: what is this thing called AI? How does it do what it does? I mean, even skeptics have to admit that it’s really impressive how it’s sort of leapt over all of the previous barriers. How is this happening? Azhar: You know, I think we’re still figuring it out. I think of AI ultimately as a machine that does certain things, and it’s been built by passing first millions, then billions, then tens of billions, hundreds of billions of trillions of words of human output through a neural network to give it some sense of how humans have thought about the world. And because it operates at dimensions well beyond the form of space and time, it seems to be able to find relationships between quite complex concepts. And I think we’ve all had that experience, whether we’ve been using Chat GPT or Claude over the last two or three years, that it seems to be able to recognize things that are quite deeply related that don’t immediately spring to mind. And in the last year and a half or so, the labs have started to train the AI models not just on words in books, but actually on tasks, like, “what is the set of things that you do to write a piece of code that does something?” “What is a set of things you do to use a piece of software in an enterprise?” And they’ve tried to train those models on those particular tasks. Essentially it’s aping what we do, and they use various mathematical tools like reinforcement learning where the model notionally gets a reward. Of course it’s not a reward the way you and I think of it because it’s a machine. Paul Krugman: Right. Azhar: And so that’s what it is. It’s sort of reflecting back, but also I think discovering some really deep relationships in the world that we might not spot, you know, prima facie as humans. Paul Krugman: Brad Delong [https://braddelong.substack.com/p/note-to-self-my-views-this-afternoon] calls it “a vast stew of linear algebra,” which makes some sense to me because I think that Pagerank with Google was the last thing I actually understood. And that’s the eigenvector with the largest eigenvalue. Not that anybody needs to know that, but this is like a million times bigger, right? Azhar: That’s basically it. Yeah. Krugman: But it’s sort of not what artificial intelligence was supposed to be, right? Azhar: No, not at all. I mean, I sometimes go back and look at the TV series of the seventies that I grew up with as a child, and they’ll always have an AI in the spaceship. Space 1999 had an AI you could talk to. And it was very precise, it was very clipped, and it did things and got things right. And there was a sense that you could trust it. But you’d never think to say, as I sometimes do now, you know, “Find me five analogies to help make this point.” I use it as a brainstorming partner, or I give it tracts of my book, the book that I’m writing, and say, you know, “How would Paul Krugman criticize this argument?” And I get suggestions that I then work through by hand? I don’t think we really imagined it would look like that. Krugman: Yeah. In sci-fi it would talk in a monotone and would be relentlessly logical. And in fact these models are unpredictable, they’re sometimes temperamental, they’re not reliable. That’s probably one of the big problems. It’s not at all what we imagined. Azhar: It’s not at all and this point about reliability is so complex. A couple of months back, one of the versions of Anthropic’s Claude came out and I found it so sycophantic that it became unhelpful because I like these things to help me on hard problems and to challenge me. So I switched back to Chat GPT, which has always been a little bit less friendly. And what’s going on there, Paul, is that because we don’t really have a good theory about how to build these. They are developed almost like in a petri dish and nudged in particular directions so they take the shape that we expect them to take. And to use an economist term, they improve non-monotonically with every release. So you’ll see the latest release of an Anthropic model, and there are maybe twenty or thirty public benchmarks that they’re measured against, like how well they summarize text and how well they write software code. And the next version of the model won’t necessarily be better at everything than the previous version, because you lose something in order to get it. And that’s the complexity that the labs are wrestling with. Krugman: Wow. Okay. Second naive question. I don’t think I’m a Luddite. I’ve always been happy to adopt technologies, but maybe I’m incurious on some of these things. I tend to pick up things like mathematical techniques, as needed, because I see something that could be useful. Now, I’m using NotebookLM to extract tables from PDFs, that sort of thing. But what should I be doing? I have friends who are using Claude a lot, but I can’t quite figure out what particularly agentic AI should be doing for me. Azhar: You know, I’m really sympathetic to that because I have the same issue. These tools have been developed by software developers in a really particular part of the world, which is Silicon Valley, where the culture really revolves around the art of the programmer. And so if you have a programmer’s day and you think in coding terms and you have programming workflows, it becomes really obvious what you do with a really advanced AI tool. I do a lot of research, some of it qualitative, some of it quantitative, and in such a world, those workflows don’t match the way that I think through problems. And so the way that I get around this is that I do look at things on Twitter or X as it’s called because people are sharing tips. And I often just ask the models, you know, “What could I do with you given that I’m trying to do this thing? I’m trying to solve this problem.” And it will come back and give me a suggestion. And I have had some success with agents. So I have an agent called R. Mini Arnold. So R is a play Isaac Asimov’s robots. They’re all called R. Arnold is after the good Terminator in Terminator 2, played by Arnold Schwarzenegger, who protects humanity. And R. Mini Arnold is available on my WhatsApp and it’s available on email. Krugman: Okay. Azhar: And it has access to a whole set of resources. It can browse the web, it can access LinkedIn, it can access Twitter, it can look at my library of PDFs of research that I’ve downloaded. And I can throw tasks to it a little bit like I would say a pretty decent but slightly temperamental graduate student. So sometimes it just disappears for six or seven hours at a time. And one of the differences between using an agent like that and using Claude is that R. Mini Arnold has a lot of my life’s context. It knows the music I like, it knows the book I’m working on, it knows the investments I’m making, it knows the essays I’m doing, it’s got the calendar of speeches that I’m about to give. And so when it goes off and does a task, it tries to figure out what in my world is this going to be relevant to and where can I draw threads from? And when it works, it is really sublime and it does feel a little bit like science fiction. But I would say it’s incredibly brittle. I mean there’s breaks every four or five days. A specific example was, I was thinking about the Paul David’s research about why electrification took the time it took. And I wanted to understand what were the determinations of determinants of that thirty-five year lag from Pearl Street generation to, you know, productivity growth. What could the levers be? And so I threw that into R. Mini Arnold and it set up a team of sub agents which had personalities of key economists and was able to go off and do research the way the AIs do, but also research on all the academic papers that I have downloaded in the past. I have access to JSTOR, I’m allowed to download a hundred PDFs a month. It can look at all of those and start to compile an answer in a way that perhaps a Chat GPT can’t. And it knows the context of my book and it knows the context of the essay I wrote. So what then comes back is something a little bit more structured that I can then play with. It’s a marginal improvement on doing this on Chat GPT. I’m sure you could probably figure out how to do it. But it’s quick. I use it on my iPhone. I often do this when I’m walking through the airport and I want to solve this and have this result when I’m sitting on the plane. I’ll fire that query out and it goes back and goes out and sorts that out for me. Krugman: Okay, I guess I’m getting it. But obviously you and I are not typical. The people who are using AI the most are going to be middle managers, business people, etc. And I find myself thinking about what I think of as the homemade pasta problem. Azhar: Mm. Krugman: You’re probably too young for this, but there was a time when I when young and we were using stone axes for computing, and there was a big fad of making your own pasta. Little pasta machines were everywhere. And then at a certain point there was kind of a collective, “What the hell are we doing? You know, store bought pasta is actually better. The Italians don’t do this.” And I have to think that for most tasks, the range of agents can’t be that wide. But why wouldn’t they sell that kind of thing off-the-shelf, as it were? Azhar: Yeah, well I think it’s different for an independent person or a small business or a middle manager in a big company. I would imagine that you will start to see people selling specific agents that solve your marketing problem. If you have a barber’s shop and you’ve got four chairs and maybe 30 people a day coming through. Right now what you do is, you go to ChatGPT and you help it write your collateral for your website. That feels like it’s an interim step to somebody delivering the actual finished product. Why haven’t we seen it? I think we haven’t seen it yet because the terrain is still big enough. Beyond Anthropic and OpenAI, there’s a lot of other companies building agents that are these end-to-end workflows for businesses. They still believe that the prize for them is to build the generic platform that is the tool for all tools. Because if you get that right, you have a much, much bigger business than if you’re just a vertical application. And I think we’re only a year or two into these entrepreneurs building such businesses. I think as some of them succeed and some fail, the ones that are not able to succeed in the general space will start to verticalize, which is what we saw in the advent of the internet. We saw it in software as well. But I think within a big company it’s a different set of questions because you have far fewer degrees of freedom as a marketing manager in a large company than you do if you own your own barbershop. You have all these rules, you have all these other teams you have to interface with, you are held to the priorities and the plans of the company as a whole. And in that instance, I think, it’s much harder to see how you use AI to really change the way you work. Krugman: Yeah, I mean, again we’re talking about ancient history here, but you know, everybody still uses Excel, even though it has always been horrible. But the constraints of corporate life mean that everybody has to use Excel. So that means maybe we’ll see quite a lot less coding a few years down the pike because the people will just be able to purchase whatever it is they need. I don’t know. Azhar: I think there’s a balance. You hear people proselytizing heavily, saying, “I think this technology is going to be impressive and have a significant impact.” But when people pitch this, they forget that there are other actors in the market who might respond to what’s going on. Right now, if you’re a large company, you want to be building as much as you can because what you can buy isn’t right for the market. If you think about Henry Ford putting together the Highland Park plant, he couldn’t go to a supply chain and buy what he needed because nobody was thinking in those terms. I think we are slightly at that stage for large corporates now. Whether we’ll be there in five years, I don’t know. The question we have to consider is where the value will reside: between having your own capabilities to design software for your processes, or handing that over to another company designing software for a hundred businesses like yours. Historically, it has made more sense to hand it over to another company, but the cost curves may have changed sufficiently that you’d rather have the nuance and control to do whatever ‘vibe coding’ becomes in 2030. Krugman: I know with healthcare software, organizations like the VA that built their own have done much better than the ones who tried to buy it from Microsoft. So yeah, it might be a story that makes sense. And actually, since we’re talking about going for the models versus something much more specific, how do you think about the Chinese versus the big US AI firms? Azhar: I’ve just spent eight days in China and I was really fortunate. I got to speak to developers and engineers and management from about a dozen of the Chinese labs. In many cases they hosted us in their offices. The main thing the Chinese companies say about the US firms, is that Claude code is brilliant and Claude is the best model that is out there and they really couldn’t get enough of it. The term is, they’re Claude-pilled. They talk about the constraints on getting access to computational power but just in a way that’s a fact of life. I mean there’s no sort of commentary on it other than it’s hard. They have to figure out how to get around that and how to build a culture of efficiency when you don’t have as much [computational power] and I think they have built a culture of efficiency really, really well. I think it’s going to help them over the longer term. They don’t really talk about competition with US labs the way the US talks about competition with China. But they do see themselves competing with each other. And as you know, that’s what the Chinese economy is. It’s mayors in different cities who almost act as venture capitalists who compete tooth and nail with each other to become the electric vehicle hub or the solar hub or the AI hub of the nation. And what I would say is, the models are really, really capable. They’re very efficient, which is why they’re so cheap to run, which makes them very competitive for a whole range of tasks. But at the margin, it’s instructive to note that everyone was using Claude for coding as opposed to the cheaper Chinese version. Krugman: That’s interesting. So you can imagine a future where a lot of businesses are actually using these less comprehensive but much cheaper models. I think what I’m gathering from you and from other people is that a lot of entrepreneurs in the US are still dreaming of the uber-model that solves all problems but that probably is not going the way it all goes. That in the end we’re gonna end up with a lot of specialized models, but also the uber-models will still have a role. Azhar: Yeah, it never made sense to me that you’d have a single model that would do everything because if the single model is going to solve the Riemann hypothesis, it’s gonna require a lot of resources. And if all you need to do is get it to root a bill to the finance department, it seems a bit silly to ask Einstein to come and do that for you. We’ve had segmentation of markets for a long time and it’s like with airlines. There’s a reason why not every seat on an airline is first class. Some passengers don’t want it, don’t need it, won’t want to pay for it. So I do think that the ecology looks like a whole array of much, much cheaper models that are serving by volume lots of corporate needs, and then having more sophisticated, complex models for the more complex tasks. I think you’re already starting to see this. I don’t see it, by the way, as a shock to the industry. I just think this is what happens as an industry matures. You know, you start with one size fits all, then you start to segment your customer needs and you start to serve them in the most profitable way you possibly can. And that just feels to me like the way that the markets have matured. Krugman: Okay. Let’s move to more macro considerations. People have been worrying about a bubble. A lot of us still remember the nineties quite vividly and think about all of that. But you just aren’t seeing the bubble. You wanna talk about that? Azhar: I remember what it was like in the nineties. I lived through that one and also the housing bubble, which frankly was far, far worse and much more terrifying. I have a really simple mantra here, which is that honest customer revenues tend to be the engine that gets you through this, right? You know, what caused the problems with the US railroads in the 1870s and 1880s? It was that the revenues didn’t materialize because the tracks were being laid in places where there were no towns. That was a problem. The same was true in the dot-com era. My team and I realized last year that it’s very hard to get good quality data on how much was actually being spent by American businesses and consumers on AI. So we’ve spent several months building systems and gathering data to give a deduplicated view of what that number is. And just to give you a sneak preview, is $150 billion per annum, annualized at the end of May 2026, and about 90 billion dollars in the previous 12 months, from May ‘25 to May ‘26. So you can see it’s growing, and those are deduplicated numbers. So if you spend a dollar with OpenAI, and they have to pay Microsoft 60 cents to run the servers, we only count that as a dollar. We don’t count it as, you know, $1.60. It’s a much faster revenue growth rate than mobile or the internet. It’s also a small number because the US is a $32 trillion economy. And I think the thing is that at that level of spend, you are able to roughly cover the depreciation on the enormous capital expenditures that have gone into AI just this past year. But next year or the year after, you have to double your revenues again and again in order to cover these increasing commitments. The thing that often pricks a bubble is when financing starts to get a bit smelly. That was clearly the case in the global financial crisis, where synthetic collateralized obligations were magnifying the risk on subprime mortgages—it was all “smelly finance.” In the dot-com bubble, the dot-coms themselves didn’t really have much smell about them. There was a lot of disbelief, but the telecoms clearly had issues with their internal revenue generation. So the other thing that we look at is how bad, poor, or strong or robust is the funding quality. And that funding quality measure is definitely getting worse. It’s worse now than it was nine months ago. But it doesn’t seem from the numbers to be at the level that it has been historically when these things have imploded. Nor does it seem to be the type of exposure that is really systemic, which is what we saw in the global financial crisis. There are companies like Oracle and Coreweave whose debt looks very risky, and it’s harder and harder for them perhaps to raise money, although Oracle just did. But it doesn’t feel like it’s systemic. You know, when the the global financial crisis popped, no one knew who was in trouble, whereas now you’d be able to isolate it with a single company or a single firm. So at the moment we feel that this is still a demand-led boom, that funding quality has definitely gotten worse, but not so bad that I would say that there is an imminent problem on the horizon. Krugman: So at this point, you’re saying that roughly speaking, final demand for this is about half a percent of GDP. What share are AI-related stocks in market value? It has to be substantially larger than that. Azhar: They’re about forty percent of the S&P 500 right now. Krugman: That’s a huge mismatch. Revenues are not the same as profits, but you’re talking about what is still a relatively small business relative to this immense economy, yet it dominates the financial markets. That would be at least a possible source of alarm. Azhar: Let’s dig into that, because a stock price is a reflection of the expected future value aggregated across the market. Forty percent feels high, but if you look at the measure of earnings, these companies actually have a much higher proportion of earnings and earnings growth. If you look at the US stock market in 1900, after the railway calamities of the mid-to-late 19th century, railroad stocks were sixty percent of the capitalization of the US market. We had worked our way through the busts by that point. There’s a fantastic piece of academic work by an American finance professor named Bessenbinder. He looked at the stock returns of 23,000 US stocks from the 1900s through 2022. Those returns are highly concentrated. About two-thirds are concentrated in roughly 30 companies. Those companies are oil, electricity, or car companies—the general-purpose technologies at the start of the 20th century—or they are the IT companies like Apple and Nvidia. The only exceptions were Walmart, a couple of healthcare businesses like Pfizer, and JP Morgan. Historically, you get this concentration of a number of winners when you have a new general-purpose technology, and that is showing up today. I don’t feel we’re overly concentrated from the perspective of risk, and the price does not feel totally out of whack compared to where we were during the dot-com era. Krugman: One last devil’s advocate question. I keep thinking of the California gold rush. If you had looked at the revenue and spending on gold-rush-related businesses as a whole, it probably looked solid. But the trouble is it wasn’t the gold; it was the picks, shovels, blue jeans, women and whiskey that were the revenue streams. Is that a fair question to ask about AI right now? Azhar: It’s a great question to ask. The question is what determines that $150 billion annualized demand? We see that just under 30% of the S&P 500 have pointed to a generative AI project with a quantifiable result in their earnings calls. They are under pressure to say they do this, so maybe that’s what’s going on. But when I talk to executives, like 30 finance businesses in New York, they all plan to spend more next year, even though not a single one could point to even a 10 basis point improvement in their business from the investments made so far. Krugman: Right. Azhar: When we break out that $90 billion, $60 billion of it is in the US. That’s a lot of money for a single company, but spread across thousands of firms, it’s still at the experimental stage. We should consider whether these executives are learning by doing. The messages I get vary from those having success in the tens of millions who want to reach hundreds of millions, to those finding it harder but persisting. We’re slightly beyond pure picks and shovels, but in Paul David’s work, it took 50% of American companies getting electrified before the productivity rise. We’re a long way from that. Krugman: Headlines flashed about a KPMG study with case studies on the usefulness of AI [https://smallpdf.com/file#s=a35e268b-87cf-4380-b3f2-77b78c9e2274&r=read] that turned out to be AI hallucinations. It’s a wonderful thing. Azhar: It is brilliant. One thing that is quite challenging is that the market has talked a lot about bottlenecks. We saw this with railroads when the US couldn’t make enough steel. There are these bottlenecks, and there’s a lot of emphasis on power and getting electricity to the system. There’s more demand than supply capacity for AI right now, but there’s a question of whether there’s enough capital. We may see another few trillion dollars of intention from tech companies to build infrastructure to 2030, which starts to rival the new issuance of the US Treasury at $2 trillion a year. I’m wondering if this capital constraint is going to be an issue or if the market knows how to clear it. Krugman: Ordinarily, we’d expect to see that in prices. Real interest rates are well off their pre-COVID lows. They are higher now, but still substantially lower than at the peak of the nineties tech boom, when they were around four percent. They’re more like two now. It’s surprising, given the AI boom and massive budget deficits, that rates aren’t even higher. Whether this is an actual constraint, Nvidia is not the US Treasury. They need risk-tolerant capital. The possibility that these firms may not be able to raise enough money is something we need to think about. Azhar: Yeah. On that Nvidia point, I saw that credit default swaps on five-year Nvidia bonds—the cost of insurance against default—are currently lower than US Treasuries. Krugman: I saw that, and it strikes me as completely crazy. If you think the US government is not reliable, you shouldn’t be investing in chip stocks; you should be investing in canned goods for your bomb shelter. But anyway. Azhar: Are you telling me that markets aren’t perfectly rational, Paul? Krugman: Good heavens, I can’t say that; they’d take away my economist card. We’re recording this on SpaceX Day, and I’ve been wondering if there are limited pools of capital for cutting-edge investments. I wonder whether Elon Musk is diverting capital that AI might need. A whole lot of meme money is pouring into SpaceX right now. Is that something I should be thinking about? I mean, he’s got what everybody tells me is a crud AI product in Grok, and yet… Azeem: Musk showed his willingness to adapt; his AI product is now being subsidiarized using his capacity to serve customers like Anthropic. He has an incredible following, but people who have worked with him say his ability to relentlessly focus and optimize sets him apart. His first-principles thinking has brought down the cost of space launches faster than anyone in history. He pushes the rate of learning aggressively. For all the challenges and his mercurial behavior elsewhere, that’s generally a good thing because technology has brought down the cost of inputs significantly. We’re going to be much further ahead in space than we would have been if SpaceX had not been successful. It raises questions about how to govern what used to be a commons, but there is a definite benefit from coming down that learning curve so quickly. Krugman: That’s fair. The one time I looked at Musk’s activities and thought he was really onto something was when I realized he diagnosed that the cost of space launches is really the rocket, not the fuel, and recovering it makes all the difference. Being able to make it happen is a real productivity thing. This is all moving so fast that we don’t have time for the technical productivity issues we had in the past. It’s feeling like a Solow moment where people say, “I see the technology everywhere but in the productivity statistics.” Do you want to talk about that? Azhar: It comes up all the time. I wonder if we need things to happen more quickly than we used to. We aren’t seeing it in the numbers yet. Erik Brynjolfsson at Stanford says he thinks it is showing up in the aggregate numbers. How quickly should we expect a technology like this to show up? At $90 billion a year, that’s not much of US GDP. These are early stages where companies are learning. The first $100 million you might spend on AI is about learning, and we’re in that mistake-making phase. The model Paul David and William Devine talked about in electricity is helpful. In the first phases, you’re retrofitting your capital stock and processes with the new technology. It’s not until you depreciate existing capital and change processes—like Ford did at Highland Park—that you see productivity benefits. To put numbers to that, what would we expect to see in the Ford equivalent of Highland Park in terms of output? Krugman: Yeah. Azhar: I thought we might see what happens to revenues per employee in an AI-native firm. Across high-end companies like McKinsey, it’s about $400,000. For Meta or Google, it’s about two to two and a half million dollars. In AI-native firms like Mercor, that number is closer to seven million dollars per employee. For Anthropic, it’s close to ten million. You can measure the enormous commercial productivity of a single employee if a firm is AI-native. We’re talking about a handful of firms, but we can pick up the shape of what’s possible for the productivity of a single employee. It may be hard, it may take time, but it’s possible. Krugman: What would those numbers look like per dollar of invested capital? One worry is that this is an enormously capital-intensive business that replaces labor. The oil refineries of New Jersey have enormous revenue per employee because there are no workers, just monstrous capital installations. Is that a factor? Azhar: Anthropic has raised in the tens of billions rather than hundreds of billions and had a profitable quarter ahead of schedule. What we don’t know is how much of that capital goes into developing the next model versus monetizing previous generations. Their IPO in the next six to nine months will tell us. Chinese companies are using much less capital to build models that are nearly as good. So I think the harder part of your question is that if every model that OpenAI or Anthropic costs ten times as much to deploy and develop, but lasts only a couple of years before it’s defunct because of competition, what needs to be true for that to be sustainable for more than a year or two? To me, that is a really tricky question as well. Krugman: You’ve cited intermediate measures. Rather than revenue, we look at generated lines of code, which has exploded, versus actual usable applications, which hasn’t. Does that tell us anything? Azhar: Lines of code is an odd measure. We’ve made it much cheaper to write code, so less determined people are writing it now. It’s unsurprising the increase hasn’t been met by proportional productivity. Data suggests we’re getting more high-quality code, but also a lot of useless waste. This isn’t the first time a useful input in the economy generated waste. Think of a barrel of oil: we count the whole value in GDP, but two-thirds is thrown away as waste heat. Only one-third is useful energy. Sloppy lines of code are a similar form of waste we’ve been happy to tolerate in other sectors for a century. Krugman: A weird analogy is when widespread word processing came in. Books started getting longer. It was so easy for authors to turn out hundreds of pages. What might have been a two-volume series became five. Azhar: On that front, we’re at an enlightenment moment. In 18th-century France, the battle was over who gets to write and express their story. Men and women produced remarkable works with quill pens that encapsulated a world. Krugman: Right. Azhar: Is it worse that we allow for more expression? We are worse off when that connects to an algorithmic recommendation system that drives constant slop at us. But we aren’t inevitably worse off because we’re giving access to many more people. In reducing costs of access, we might find amazing people. In breaking down silos of knowledge, we might find connections—perhaps something in battery chemistry that is useful in cardiology. We don’t know because we’ve never been able to get those experts to talk. I look at each opportunity discreetly. Krugman: There is a potential book here: The Upside of Slop. This is an unrecognizable scene from eighteen months ago. Wow. Azhar: We could get ChatGPT to write it. Krugman: I started my career writing papers longhand on yellow legal pads. Amazing change. Azhar: I still write everything with a fountain pen. I’m writing my new book longhand and most of my research is too. The computer is turned off because AI does all the boring stuff like PowerPoint and emails, giving me time to apply my brain to things I want to think about. I’d be happy to continue this conversation in a few months. Thank you for inviting me. Krugman: Thanks so much. Take care. Get full access to Paul Krugman at paulkrugman.substack.com/subscribe [https://paulkrugman.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

Kommentarer

0

Vær den første til at kommentere

Tilmeld dig nu og bliv en del af Paul Krugman Podcast-fællesskabet!

Kom i gang

1 måned kun 9 kr.

Derefter 99 kr. / måned · Opsig når som helst.

  • Podcasts kun på Podimo
  • 20 lydbogstimer pr. måned
  • Gratis podcasts

Alle episoder

73 episoder

episode Lisa Graves on the Supreme Court cover

Lisa Graves on the Supreme Court

For all my interviews and more, subscribe on YouTube [https://www.youtube.com/@PKrugman]. Lisa Graves is the author of Without Precedent [https://www.hachettebookgroup.com/titles/lisa-graves/without-precedent/9781645030676/], a history and analysis of the Roberts Court and the expert on how the Federalist Society has been working to undermine democracy. Yesterday this happened: So I managed to arrange a conversation with Lisa about what is happening: . . . TRANSCRIPT: Paul Krugman in Conversation with Lisa Graves (recorded 6/29/26) Paul Krugman: Hi everyone. This is a bit of an emergency recording and podcast because today, which is Monday as we’re recording this, the Supreme Court has just handed down several decisions, of which one was really alarming. And I had talked earlier with Lisa Graves [https://substack.com/@thelisagraves], head of True North Research [https://truenorthresearch.org/lisa-graves/], and Court Accountability who had been warning about all this stuff when I talked with her previously and on other occasions. And I wanted to get somebody who’s actually following this to weigh in. So, hi Lisa. Lisa Graves: Paul, thank you so much for inviting me back on. I really appreciate it. Krugman: Yeah. So, there were obviously several decisions that came down, but Humphrey’s Executor… This is absolutely mind-boggling. So why don’t you talk about Humphrey’s Executor and then we’ll talk about the background and what this says about the Roberts Court? Graves: Yes. Today’s [https://www.scotusblog.com/2026/06/court-allows-trump-to-fire-ftc-commissioner-and-overturns-major-restraint-on-presidential-power/]Slaughter [https://www.scotusblog.com/2026/06/court-allows-trump-to-fire-ftc-commissioner-and-overturns-major-restraint-on-presidential-power/] case [https://www.scotusblog.com/2026/06/court-allows-trump-to-fire-ftc-commissioner-and-overturns-major-restraint-on-presidential-power/] involving the FTC, is a case where the Roberts Court has overruled nearly a century of legal precedent which prohibited presidents from firing commissioners, specifically on the Federal Trade Commission, the case that you mentioned, Paul. Humphrey’s Executor was specifically about the FTC. It was about this provision that barred FDR, Franklin Delano Roosevelt, from firing someone whom Hoover appointed to that commission. When Congress created the FTC, it set a standard that required that you would have to have cause to fire someone outside of their term. The way the FTC is set up is that it has five commissioners by statute. Three of those commissioners are appointed by the president, two from the president’s party, and two are not from the president’s party. What’s happened over the past year is since Donald Trump became president again, he fired the Democratic commissioners on that commission. And so for the last 15 months, that commission has had only two commissioners, two Republican appointees, which has meant there hasn’t been a Democrat even in the room to consider these cases that are brought to the FTC, which relate to the power of huge corporations to merge with one another or not. And so, in essence, the FTC investigates proposed mergers and other things that might be considered a restraint on trade or affect the ability of consumers to get a fair price on things, for example. And what Trump has done is that he’s basically dominated that board in defiance of the plain language of the statute. So any legal action by Donald Trump that has been countenanced by this Supreme Court, which allowed that firing to stand, in essence, did not allow the lower courts to reinstall Rebecca Slaughter to that post while the case was pending, and then just today ruled that, in fact, under their new approach to the Constitution, Donald Trump has the power to fire anyone at any independent agency—other than the Fed, it seems—no matter what Congress has said. And this is the result of an invention of a theory from the Reagan revolutionaries back in the day to try to aggrandize presidential power through what they describe as “the unitary executive theory.” So the bottom line is that this ruling by the Roberts Court will allow more of the corruption that we’ve seen going on by this administration in terms of people who may be donors to Trump seemingly getting out of investigations or having their mergers go through, and we will not have a commission at the FTC that has any independence—just loyalists for Donald Trump for the foreseeable future. Krugman: And it applies obviously not just to the FTC but to any agency except not the Federal Reserve. But the FDA—if there’s an FDA official who is viewed as being too hostile to RFK Jr’s vaccine doctrine or something, then Trump can fire him. And the Supreme Court has said, “Well, yes, the president has that power. Congress cannot set any ground rules that can’t be overruled by presidential edict.” Is that a correct interpretation? Graves: Well, that’s the broad strokes of it, but just to add a little bit of gloss to that, which is that Roberts has been pursuing this agenda for years now. And he accomplished part of it through a case called the Seila Law LLC case, where he allowed Trump to remove the head of the Consumer Financial Protection Bureau. And the CFPB also had restrictions on whether that person could be fired without cause, and Roberts already moved the ball forward on that. And that has had a cascading effect along with the terrible, unprecedented immunity decision, where the Roberts Court gave Donald Trump immunity from criminal prosecution for any of his so-called official acts, which included directing agencies like the Justice Department to do his bidding. And so, up until now, what was left was this notion that if Congress created a board that specifically had limits on how a person could be removed from that board because of its regulatory function to implement congressional will to act in a legislative way, boards like the Federal Trade Commission or the National Labor Relations Board could not be swept of their members. But before this decision, through the machinations of John Roberts, in essence, Donald J. Trump was already exerting a power to fire anyone within the executive branch, whether they were on an independent board or not. Krugman: The Consumer Financial Protection Bureau felt like it was a little bit different. It was basically Elizabeth Warren’s creation. It was a relatively new agency. But this now generalizes it to everything, and it goes back to the FTC, which is a very, very old institution and is where this whole Humphrey’s Executor comes from. So now that’s everything. Graves: Yes, except for the Fed. That is correct. And I think, as you were describing it at the outset, Paul, this has enormous implications for the American people and American consumers, because, in essence, consolidating power in this way in a president is not required by the Constitution. Although Roberts is somehow claiming it is, it’s not. This is part of this invention of this so-called “structural Constitution” under this very rigid notion of separation of powers, which basically guts some of the core powers of Congress and vests those powers, in the views of this court, in the hands of the president and the president alone. And this ignores the reality of what’s happening, which is that you have a president who’s not actually executing the laws passed by Congress; who’s thwarting those laws. When you look at Article One and Article Two of the Constitution, what you see is that the president’s primary job is “to faithfully execute the laws of the United States.” And what’s happening here is that the president is circumventing those laws, thwarting those laws, and doing so in ways that raise serious questions of corruption or potential corruption or undue influence. And the area of trade and mergers is one of the most, you know, potentially profitable areas for people trying to curry favor with Donald Trump. And what we’ve seen over the past fifteen months is that the FTC under the control of two Trump loyalists has dismissed more than thirty-three investigations into mergers that were begun before Donald Trump took over. Krugman: This is basically a Federalist Society thing, and when they began pushing this, they probably had in mind that we would be pursuing an ideological agenda, that this would be something that would allow a right wing president to essentially just overrule Congress when it was doing things that weren’t sufficiently right-wing, or weren’t sufficiently “Reaganesque,” since this goes way back to the 1980s. But now it’s very much personalist. We’re talking about this as what one guy gets to do, which might not even be ideological. As you say, it might just be corruption. Graves: Yeah. I think it’s both, right? So there is this throughline from the Federalist Society that helped get these judges on the court. These six members were all either active members or noted to be members of the Federalist Society. They were part of the pipeline to power that the Federalist Society was created to accomplish. Roberts and Alito and the three Trump appointees, they were all appointed in the aftermath of the “No More Souter” campaign by Federalist Society activists who said they didn’t want judges who were going to follow the precedent. They wanted people who were going to basically be ringers and change the law to reverse the progress of the 20th century. And this attack on what they call the administrative state is really an attack on expertise. It’s an attack on the ability of agencies to do the job Congress has entrusted them to do, which is to faithfully implement those laws in defense and in pursuit of the public interests of the American people. So you now have this convergence between a president who is so determined to take cash, in many ways, out of other people’s pockets, including the pockets of the taxpayers, in order to advance himself, to aggrandize himself. And this has come to a head at a time when the court has the majority. So this captured court has the majority it needs to accomplish its long-term ideological agenda. And what a lot of people don’t realize is that when you look at the Supreme Court, this nine-member court, and the six members in the Republican-appointed majority, five of those six members were executive branch attorneys. They were people who cut their teeth for years in advancing presidential power, in trying to expand presidential power. And they’ve acted, I think, with a real arrogance toward congressional power, hostility toward congressional power, and with a bias toward their own long-standing agenda as lawyers, as Republican lawyers in this cause. And they’re moving forward even at a time when we have a president at the helm who is abusing his power almost on a daily basis in terms of asserting extraordinary king-like powers to do almost anything he wants. Krugman: I would actually disagree with that. I don’t think it’s almost on a daily basis. I think it’s at least several times a day. Graves: It’s an hourly basis, right? Yes. I’ll take that correction. Krugman: But yeah. You say that these cases were not brought to try and stop corporations from doing stuff, but the ability to exercise the function of the FTC as a guardian of the public interest. But it’s also a negative power. You can imagine that a merger that the FTC would normally have blocked is allowed, but also one that it would normally have approved could be blocked if, you know, the corporation in question has not cut Jared Kushner in on the deal, basically. Graves: Well, right. You know, I looked into those instances where you can see the list of the various mergers that the FTC has stepped away from investigating. And one of the things I saw was that before Trump’s appointees took office, there was a Joe Biden appointee—her name is Lina Khan. She’d been a vocal critic of Google and Amazon’s market powers. And what happened in one of these cases was that the FTC terminated the examination of one of Google’s acquisitions. You had Google’s CEO at the inauguration last year, you had Google Alphabet giving like a million dollars to the inauguration committee. You have a tie-in on underwriting the Trump ballroom, and then you have an FTC that is not pursuing a further examination of that acquisition. Maybe it would have been approved with the full commission, maybe it wouldn’t have—we don’t know. But what we do know is that there are other ongoing investigations, perhaps around Facebook, for example. Maybe in their view or in someone’s view, Facebook hasn’t ponied up enough cash to get that dismissed. It creates this real environment of coercion and shakedowns, the perception that if you’re not playing ball with Trump, you’re going to be treated unfavorably. And in fact, he’s routinely threatened companies that he thinks aren’t sufficiently loyal to him. Krugman: Yeah, I mean, it’s so raw and out in the open now. And by the way, Lina Khan is impressive as hell. I had a dialogue with her at the Graduate Center a few months ago, and she’s now advising Mamdani in New York. It’s just worth saying that a lot of the players in this have been around for a while and keep on showing up in different venues. But the power to do favors is also the power to withhold favors. So this is, as you say, an enormous source of potential corruption. You also have to wonder, if you’re a business, do you even know what the ground rules are? That’s what I’m wondering about a lot now. Graves: Well, you know, I think that that is a really good question. And it reminds me of this historical episode from that robber baron era when Teddy Roosevelt was president and the big companies—Standard Oil, for example, the mega-millionaires who would be billionaires today—were exerting such power over Congress that the smaller companies, the median-sized and smaller companies, were being shut out of the ability to really influence legislation. And that resulted in the Tillman Act, which is still on the books, that bars direct corporate contributions to a candidate. They get around that through PAC donations that are allowed, or through giving now to these C4 nonprofit groups. But you know, since about 110 or 115 years ago, it has been banned for there to be direct corporate contributions because, at that time, other businesses who weren’t the super-bigs were feeling like they were getting the raw end of the deal. And so I think that’s probably repeating now, where there are smaller companies that can’t afford to make million-dollar donations to the inaugural committee or make tens of millions of dollars of secret donations to this ballroom boondoggle, who are going to be disadvantaged because they can’t potentially buy their way in to favorable treatment. Krugman: Yeah. I mean, even big corporations who happen, for whatever reason, to be not sort of buddies. You know, it definitely looks like this administration has it in for Anthropic. I’m sure that they’re not angels, but this is still pretty amazing that this is one of the best AI out there, but they are not friends with the president and so they’re… Graves: Right. And look what happened at TikTok. For quite a while there we were hearing all of these attacks on TikTok, concerns about security or security access through that app. And with the visit of one billionaire, Jeff Yass of Pennsylvania, who’s one of the fifteen richest billionaires in the world—who made part of his fortune on super-rapid trading, but another part of his fortune on an early investment in ByteDance, the owner of TikTok—with one visit to Mar-a-Lago by Jeff Yass, suddenly Donald Trump was singing a different tune about TikTok, and then ultimately intervened in a way that ended up giving some of his allies ownership in TikTok. And so you have this real... I would say the most generous thing I could say would be it’s unseemly. I mean, it’s outrageous, actually, to have a president engaging in sort of corporate manipulation in this way to reward his friends and punish his enemies, as we’ve seen with Trump going after law firms, Trump going after universities, Trump assailing different corporations whom he dislikes or whom he considers not to have donated to him or advanced him. This is extraordinary in American history. I think it’s unprecedented, actually. Even with the corruption that was unfolding during that robber baron era, I think we’ve never seen anything like this kind of grift and graft. And this corruption is inherently destabilizing to American society, to American business, to investments in the United States. If the U.S. becomes, as it is becoming, a society in which basically you have to engage in these sorts of legalized bribes—although I’m not sure how legal some of them may ultimately be—that changes the U.S. as a stable economic superpower. Regardless of what a particular policy preference may or may not be at a given time, this is an extraordinary devolution of Americans’ role in both the U.S. economy and our role in the world—to have a situation where companies and countries have to pay up to a president or cut a deal with the president’s son-in-law or Howard Lutnick on minerals or what have you, or where someone can make a call to the Pentagon to get a special contract approved for Donald Trump’s sons. As you said, this is not just corruption on a daily basis. Whether it’s legally actionable—some of them may be, some of them may not be—but on a moral level, it’s corrupt on an hourly basis. Krugman: Yeah, on average every hour now. So, you mentioned devaluing expertise. The role of experts in a lot of this stuff—I think part of the issue is whether there are sort of procedures for consulting experts on things that will now be by the chopping block. Is that right? Graves: Yeah, I mean, that is part of a broader trend. It transcends the Slaughter case. But what we’ve seen is a real war on expertise. Certainly part of that was through the DOGE efforts of Musk, but those efforts, those firings of so many people, so many experts across the board in all these agencies, that has really decimated, not just the baseline of our skilled workforce in the federal government, but also demoralized the people who remain. And that’s across the board. That’s in areas of vaccines, it’s in public health, it’s in climate science, it’s in earth science, it’s in trade. It’s in all areas where we’ve probably cumulatively lost I don’t know how many thousands of years of expertise that the people actually invested in through paying these civil servants who were hired on a non-political basis, who were hired for their expertise to serve the American public. Whether it’s through the National Institutes of Health, or the FDA, or the U.S. Department of Agriculture, we’ve lost an enormous amount of expertise. And we’ve also seen the ways in which this Roberts Court has not intervened to protect against those firings. You’ve had people who should never have been fired, and months and months later, some of them are reinstated. Meanwhile, they may have moved on to other jobs, they may not want to come back into the government. This loss of expertise is a disadvantage for we, the people of the United States, in terms of having people who are looking out for our interests based on years of work, as well as scientists who’ve been reliant at the universities on these grants and more. This also is a real disadvantage for our national security, because it’s not just in U.S. civil society that we’ve seen these firings. We’ve seen people fired from the Pentagon who were, leading lights, people who had records of impeccable service to our country. We’ve seen that happening in the national security arena, in the intelligence community, where people who have deep expertise have been fired. We’ve seen that at the Justice Department where people who had expertise in anti-corruption, in enforcing DOJ’s rules to make sure that prosecutions weren’t politicized. We’ve seen the FBI firing people for just the act of doing their job to protect and investigate those who committed violent acts against our Capital Police officers. So this war on expertise, this war on civil servants, is deep and wide, and it is going to take a lot to repair. Krugman: Yeah, my experience in dealing with higher-level civil servants has always been that we had far better people in those positions, in a sense, than we deserved. You had all of these people at Treasury or at agencies that I would deal with, who could have made two or three times as much money in the private sector, but they did what they did because they thought they were doing something meaningful. They felt that it was a better use of their lives. And now, even if you haven’t fired them, if you’re disrespecting them, we’re going to lose that. Graves: Yeah, I mean, I have a bias on this because I was a career hire for the Justice Department and ultimately became Deputy Assistant Attorney General in the Office of Policy Development, the Office of Legal Policy. And I was there with other attorneys who could have been making three, four, five, six times as much in the private sector, but we were honored to be able to serve the American people. And I felt, every time I walked into the Justice Department back then—this was in the Clinton administration with Ms. Reno as the attorney general—there was this engraving that said, “The place of justice is a hallowed place.” And I thought about it as a place where people set aside their partisan political views or their personal religion to do the work of the American people. And almost in every instance where I had the chance to work with someone across that agency and other agencies, I was so impressed with the devotion and intellect and wisdom of the people who had chosen public service as their career. Krugman: Yeah. I mean, this was the Clinton administration, but you know, I had my year in government in the Reagan administration at sub-political level. But you know, the kind of deputy assistant secretary and office director level were astonishingly good people. Once in a while, you would get to see one of them lay down the law to the political appointee above them and say, “No, that’s not how it works.” So, this is a long-term project, as you said—the unitary executive, the stripping away. Presumably, when John Roberts and his friends began this, they didn’t imagine that “the unitary executive” to whom they would give all this power would be—to use the technical term—a f**k-up like Donald Trump. I mean, I’m a little surprised that didn’t at least give them some pause here. Graves: Yeah, it’s interesting. You’d think that they could slow things down for a moment given how reckless and destructive Donald Trump is, but they haven’t, right? This is not necessarily something they’re compelled to do. So, for example, Humphrey’s Executor has been on the books since the 1930s. They could easily have just said, “We affirm the lower court in Slaughter’s case and we affirm Humphrey’s Executor as still good law,” and just been done with the case. She would have been back on the commission months and months ago, maybe last summer. For Lisa Cook, you know, they sort of held that for a moment, held out this notion that they were going to allow her to remain, that they weren’t going to intervene on the Fed, but everything else was fair game. This court is taking cases where it could easily just either affirm the ruling below or reverse it based on a citation of long-standing legal precedents, but it’s not. And what most people don’t know is that this Roberts Court is only hearing 60 to 70 cases a year. A couple of years back, the Supreme Court was hearing up to a hundred or more cases. In the 1980s, there were changes to the court’s jurisdiction to basically leave most of its jurisdiction discretionary. The court only has to take cases basically where there’s like a fight between Arizona and Colorado over the Colorado River—when there’s a fight between two states. Otherwise, every case they take is discretionary. They are choosing this docket. They’re choosing, in Donald Trump’s term, to take up cases on the shadow docket where they ruled for him almost all the time—more than 90% of the time—and on the main docket, the docket that we’re seeing the decisions coming from now. Those are all cases where the Roberts Court has decided to have an oral argument and issue a decision, even where the long-standing precedent is against Trump’s actions. And so I wish that they had some heartburn over it. They don’t seem to, though, because they could easily, in the Slaughter case, have said, “No, she has to go back on that commission under the long-standing precedent of Humphrey’s Executor.” And they chose not to do so. And they chose not to do so even as the Musk operation, the DOGE operation and more were decimating our agencies; as they were decimating—not the FTC in that particular sense because it was hitting all the agencies,—but they did so knowing who Trump is and knowing what he’s doing. Krugman: And now they have to know him even more, right? In some sense, we are all in the reflecting pool, and yet they are giving him unbridled power. Just amazing. Suppose that we actually do manage to have a genuine election in 2028 and the next president is a Democrat. What do you think this court does then? Graves: Well, if the court were principled, and I don’t think it is, then it would, in essence, allow a Democratic Senate to confirm only Democratic appointees to those agencies and let them revise the rules and restore the statutory actions, the regulations that were stripped away by Trump. The reason I’m reluctant to believe that they will allow this to happen is how this court behaved toward Joe Biden and toward Barack Obama. During the Biden administration, there was a very modest student debt relief proposal. It was, I think, about $10,000 for people who made just about the average income in the United States, and that was based on a law that allowed emergency debt relief. It was expressly allowed for emergency debt relief; we were in an emergency under COVID. People were losing their jobs or weren’t working as the economy was cascading, and that was the basis, the statutory basis, for Biden doing so. But this Roberts Court asserted that Biden couldn’t do that, that this was a so-called “major questions doctrine,” and things like that had to go through Congress—that a president couldn’t just implement this through a regulation. And yet Donald Trump has done something far broader and deeper than that small, modest debt relief on a daily, if not weekly basis, and the court has barely breathed the words “major questions doctrine.” And that so-called doctrine, which is really a theory, was invented to gut our power to mitigate climate change through the EPA. That was in a case that Charles Koch and his billionaire-funded groups brought—or basically aided—in the West Virginia v. EPA case, where the Roberts Court invented the notion that the EPA could not regulate carbon without specific congressional approval under the so-called major questions doctrine. And so, in that instance, what you had was again what I consider to be a modest effort to mitigate climate change by way of requiring utilities to invest more in solar and wind energy to help address the climate changes underway—not a radical cutting off of fossil fuels or anything, but just a transition. Krugman: Right. Graves: And that was blocked by the Roberts Court. So when it comes to a Democratic president, they seem very eager to block actions that are ameliorating or compromises. But when it comes to a Republican president, they seem very eager to do the opposite—to allow even some of the most radical actions to take hold and proceed while litigation goes forward, or even to authorize and give a blessing to those actions. But there’s hope. Should I say there’s hope? There is hope. Can I tell you why I think that? Krugman: Well, sure. What is the hope, actually? What does “Supreme Court Project 2029” look like? What do we do? Graves: Look, this is a situation where we have not just executive branch aggrandizement, but we have judicial supremacy happening. We have a Roberts Court that is deciding that it is the decider—basically being the kingmaker for Donald Trump—but also the decider on almost any issue that they want to take up and issue a decree on. And that is disempowering to the American people, to representative government, to our democracy. And so what we need to have is a really robust Congress. In my personal view, we need to elect people to Congress who are going to clean house, who are going to deal with this corruption in the near term, who are going to investigate this corruption and engage in oversight in ways that make it crystal clear to as many people as possible how this is not a tenable way to run a democracy, and then build on that to hopefully sweep into power real reformers in 2029. Like what happened with Franklin Delano Roosevelt, where you have a real surge in demand by the American people to clean things up and create policies that actually help the American people. If we can move forward despite the threats this president is making with the aid of this court in terms of intervening in this midterm election with the map drawing, if the will of the people can prevail as you see it in the polls, then we have a real chance not just at a rebuke of this corruption, but actually of creating new ways to protect our interests and having an even more vigorous federal government that can help serve our needs. But that’s going to also require court reform to be part of it, because this court will strike down the same thing if it’s passed again unless we reform this court. Krugman: Okay, but what does reform look like? Is it court packing? Is it something like that? Graves: I think it’s all of the above. I think we have to have a real conversation about how we contain this out-of-control court. It could be expanding the court; there’s no number set in the Constitution. Nine isn’t the magic number. The Republicans were willing to have eight as long as it served them, you know, when they were blocking Barack Obama’s nominee. I think we have to look at jurisdictional reform. Article III of the Constitution specifically allows Congress to set the jurisdiction of the Supreme Court—it’s in plain language there. We certainly need ethics reform because the court has also been enveloped in a cloud of scandal over trips and kissing up to billionaires by Clarence Thomas and other members of that court. And we also need to have a real sense of our power in the 21st century, the American people’s power, to have a people’s Congress that actually represents our interests and not just the interests of the richest few. Because of this Roberts Court in the Citizens United ruling, because of the concentration of wealth as Ronald Reagan sought to take down the progressive taxation of billionaires, we now have these billionaires who have so much money they can invest millions, hundreds of millions, in elections without missing a beat. I did a calculation once, by the way, on Jeff Yass in terms of his spending in the Pennsylvania Supreme Court race. His spending, which was in the millions, was the equivalent of an ordinary Pennsylvanian buying a coffee and a bagel once a week. And that’s not just because they’re lucky and they’ve got money to invest in risk; it’s because the tax rate is so out of whack and has to be fixed as well. Krugman: Okay, well, let’s hope for the best. But my god, that’s among the reasons not to celebrate the 250th anniversary as wholeheartedly as one might have liked. Thanks for talking to me, and especially on such short notice. Graves: Gosh. Well, I will just say, you know, there still is a lot to celebrate, and we can set the course of the next 250 years if we don’t give up and we don’t give in. Krugman: Okay. On that note, let’s press on with the fight we are in, as Lincoln would have said. Get full access to Paul Krugman at paulkrugman.substack.com/subscribe [https://paulkrugman.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

I går36 min
episode The Court Sides With Dictatorship — and Chaos cover

The Court Sides With Dictatorship — and Chaos

For all my interviews and more, subscribe on YouTube [https://www.youtube.com/@PKrugman]. Note: It’s Lisa Cook, not Lisa Graves. Talking to Lisa Graves shortly. Earlier today, the Supreme Court declared war on U.S. democracy. It also declared war, basically, on modern society, on everything it takes to function in the 21st century. And I’m not sure that people understand that yet. Hi, Paul Krugman with a quick video update. I’ll have more on this tomorrow. Really shocking decisions handed down by the Supreme Court. There were a couple that were not awful. Lisa Cook gets to stay at the Federal Reserve, although that in itself is a huge contradiction to the important stuff that the court did. I mean, Lisa is important and the Fed is important, but much more important is Humphrey’s Executor, which is the generations-long precedent that says that when Congress creates an independent agency, it is independent. It’s able to make decisions. Of course, the president has some role. Typically, the president can choose the agency’s head subject to congressional approval, but the president can’t just go and fire officials that he doesn’t like for whatever reason or for no reason, because the agencies that operate the U.S. government and basically run our society are supposed to be professional. They’re supposed to be following their legal mandate. They’re not supposed to be personal tools of a dictator in the White House. Well, the court just scrapped that. Now, lawyers, people who are legal experts, can do a better job of explaining just what went down. But what I think is important to understand is not only does this give essentially dictatorial powers to the occupant of the White House, but it also makes it extremely difficult for the economy to function. It makes it extremely difficult for society to function. We live in a complicated world, a world of technology, where there are all kinds of spillovers, all kinds of ways in which it’s important that there be well-established ground rules. If you’re a business, take the example of medicines and foodstuffs, where we have an FDA, Federal Drug Administration, that is charged with ensuring that products that people consume are safe. We do that for very good reason. We know that not just that that there have been examples, historically, of products that were foods, medicines that were not at all safe, but also that people want some assurance. The fact that something has been FDA approved is a bit of a warranty, that it might turn out to be very harmful, but probably not. Businesses that want to invest in developing stuff need to know that there are some ground rules that determine what they can and cannot sell. Now imagine that all these decisions are made by political appointees who are loyalists to the president, who basically do whatever the president wants, whatever the people around the president want. Do you want to invest in something where you have absolutely no idea what the ground rules will be, whether it will be approved or not? Do you want to invest in a whole business line when, for all you know, the White House will abruptly decide that your product isn’t safe and that a competitor’s product is, based on spurious grounds? And what would cause those decisions to happen? Well, how about the fact that some businesses are better at the business of bribing the president and his family than others. And if you think that this is outlandish — you know, a few years ago you might have said this was outlandish, things like that wouldn’t really happen — well, as we speak, these things are happening all the time. So you are setting up a situation in which, you know, it’s a little bit like traffic laws. Traffic laws, yeah, they can be annoying, but aren’t we all kind of glad that there are in fact rules about when you can turn and when you can go through an intersection? In order to function, in order to drive your car around you need to have a set of stable traffic rules, not a situation in which a police officer can decide you broke the law and the other guy did not because I say what the law is. And especially not where the police officer does that based upon who’s been paying him off or who he expects to be paid off. The real world is far more complex than traffic rules but we need those rules and we need some stability and those rules cannot be specified with every letter, every punctuation mark set by Congress. The world is too complicated and changes too much. You need to have standing ethos, standing doctrine at the agencies that make modern life possible. Now all of that is gone. Now, it just adds to it that all of this is being done to empower a president who is the worst possible person for this job. This is not somebody you want supervising anything, everything that Trump touches turns to crud because he doesn’t care and he doesn’t actually understand or recognize that there’s such a thing as expertise as knowing what you’re doing. So this would be terrible even if we had a temporarily competent administration. But now you’re doing all of this, the Supreme Court is doing all of this to empower the guy who brought you the Reflecting Pool, who brought you the Iran war. Utter nightmare. Now, what will happen, hopefully, we emerge at the other end having fended off dictatorship. Then, I mean, as everybody knows, this Supreme Court is not actually empowering the presidency. It is empowering this president. And as soon as there’s a Democrat in the White House, suddenly there will be all kinds of restrictions on what that person can do. Well, this cannot go on. This is a clear argument that says we have to one way or another disempower the Supreme Court. I don’t know enough to tell you what is the best route to do that but court packing or something else is going to have to happen. Because this has been the clearest signal yet that we have six people (there are three who are not part of it, but we have six people) who are fundamentally hostile to democracy, fundamentally hostile to the modern world and determined to put the catastrophically bad leader that we currently have sitting in the White House in charge of everything, which is a nightmare scenario on every level. Take care, I guess. Get full access to Paul Krugman at paulkrugman.substack.com/subscribe [https://paulkrugman.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

29. juni 20267 min
episode Corruption for Make Benefit Glorious Family of Trump cover

Corruption for Make Benefit Glorious Family of Trump

For all my interviews and more, subscribe on YouTube [https://www.youtube.com/@PKrugman]. Transcript It’s kind of hard to believe, but the original Borat movie was 20 years ago. It’s time for a second sequel. And I already have the title. It would be Corruption for Make Benefit Glorious Family of Trump. I hope that some of my listeners are young enough to not remember the original Borat movie. But it was a mockumentary, a satire, in which Sacha Baron Cohen pretended to be a journalist from Kazakhstan investigating and interviewing Americans about American mores. It was not about Kazakhstan, although he did insult the country along the way. The reason I think about it is that today’s New York Times has a piece that reports, investigative reporting, on an immense mining deal in Kazakhstan, which, what do you know, turns out to be a big profit center for the Trump sons and also the sons of Howard Lutnick, the Commerce Secretary. Check out the investigative reporting for the details, but basically here’s another one, another big one. It’s part of an immense series of corrupt deals, often with petrostates — which Kazakhstan is — that financially benefit Donald Trump and his family and some of his cronies and cabinet members as well and their families. It’s all on a truly epic scale. This is a message I have been trying to get across. I don’t think many people even now understand just how much of a departure what’s happening now is from past US history. I still see people saying we might be, could be heading for another Gilded Age. But we have a level of concentration of wealth in the hands of a few people that is something like three times what it was at the peak of the Gilded Age. We’re in a super duper Gilded Age. And I sometimes hear people say, well, could we be returning to old kinds of corruption? Might we have another Teapot Dome scandal? Well, my God. Teapot Dome was a scandal actually involving mineral rights and bribes during the Harding administration, although not bribes to the president’s family, which is, again, something entirely new. The scale of the bribes was about $500,000: adjusting for inflation, that’s something like $9 million today. So how much has Trump enriched himself since returning to the White House about 500 days ago? The answer is certainly more than four billion dollars, almost certainly more than four and a half, maybe five billion dollars. Divide that by 500 and we basically have a Teapot Dome sized corruption scandal on an average day under Trump. So it’s basically day after day of scandals as big or bigger than Teapot Dome. Our corrupt grandfathers, great-grandfathers were pikers compared with this, just as the Gilded Age robber barons were pikers compared with the modern-day tech bros. This is obviously not good. It’s actually quite horrifying. How did we so quickly descend into becoming a truly massively corrupt country on a level that we used to think of as being associated only with tinpot dictators in the third world? And yet here we are. This ought to be a political issue and it ought to be a legal issue as soon as the government is back in the hands of people who actually take the rule of law seriously. Again, without going into the details of the deal, it’s surely illegal. I mean, it’s illegal under the Emoluments Clause. Probably since there are definitely Kazakhs on the take as well, it’s illegal under the Foreign Corrupt Practices Act. This is just, it’s illegal up the wazoo. Of course, it will not be prosecuted as long as Trump is in the White House. But forget any Democrat who isn’t promising to go after this massive corruption when they regain power. If they don’t, then none of this matters, but that should be a core part of anybody’s platform. I’m not a political expert — sometimes I think nobody is — but my God, again, this corruption is so blatant. And it does resonate with people. It’s really clear that corruption at the top and the sense that ordinary people are paying the price while people with power enrich themselves is an effective popular issue. That is actually the issue that brought Viktor Orban down in Hungary, which is one of the hopeful signs for what may happen to America going down the pike. So here we are, just to remind you that this scandal, it’s a huge thing. It’s page one in the New York Times, but in a way it’s actually kind of ordinary, since even this size of scandal is happening every few weeks these days. Do not make the mistake of treating what’s going on as in any sense normal. This is hugely abnormal, and I believe that the American people will understand that it’s abnormal even if pundits get bored of talking about the corruption. So drive it home, maybe for make benefit American people instead of the Trump family. Get full access to Paul Krugman at paulkrugman.substack.com/subscribe [https://paulkrugman.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

28. juni 20266 min
episode On Holding Elon Musk Accountable cover

On Holding Elon Musk Accountable

For all my interviews and more, subscribe on YouTube [https://www.youtube.com/@PKrugman]. Transcript For most of last year, Elon Musk was the second most powerful man in America. He was running a large part of the government’s budget. And during that time, he established a track record of evil incompetence. I mean, really evil and really incompetent on enormous scales. And why aren’t people talking about it more? Hi, I’m Paul Krugman, doing a brief follow-on to my discussion that was posted earlier today with Ro Khanna, the Congressman from Silicon Valley, who’s a very interesting guy in many ways. One of the things that has made him especially interesting in the last few days is that he said something entirely reasonable, which is that if Democrats retake Congress, they should hold investigations into the role of Elon Musk as head of DOGE, the sort of not exactly but effectively government agency, in destroying USAID, the agency that was the principal channel for aid to the most desperate, poorest people in the world. That’s entirely reasonable, and Khanna went on to say that there are credible estimates that the cancellation, the destruction of Doge has led to millions of unnecessary deaths, including millions of children — which is exactly true. There are studies that say that there is both in the field evidence of widespread death as a result of the cancellation and, of course reasonable health models. Because what do you think happens when you cut away tens of billions of dollars of aid to people who are living right on the edge? So of course it’s a reasonable thing to say. Musk, of course, responded not by saying, no, it’s not true or something like that. He did say that not a single person has died because of those cuts, which is utterly implausible. But he also went on to say that he was going to sue Khanna, though he hasn’t actually so far, and that Khanna should be in prison for saying — not even saying that Musk killed people, but that there are studies that say that he killed people. It’s quite evil and so much for free speech. Musk is very much like Trump, somebody who can dish it out but can’t take it, can’t even handle the kind of criticism that any public figure should expect to receive. Honestly, you shouldn’t be at all in the public domain unless you’re prepared to deal with a lot of insults and accusations. When you have the kind of role that Musk did that would come with the package even if he had done a decent or non-catastrophic job. But of course he didn’t. And so let’s talk first about the evil. It’s not just that Musk more or less personally set out to destroy this aid agency set out to cut off healthcare, nutritional assistance, just basic necessities of life for millions and millions of extremely desperate people. But he did so callously, carelessly, he even actually tweeted out, oh, “I just fed USAID to the wood chipper and I could have gone to some great parties instead.” What can you say? This is an extraordinarily evil act. It came in the context of somebody who made enormous promises about what he was going to do. People have kind of forgotten that Musk came into DOGE promising to find trillions of dollars in waste, which he would eliminate, none of which happened. Overall, it’s pretty clear that DOGE actually worsened the budget deficit at least a little bit. He also made specific claims along the way, most notably his claim that there were something like 20 million dead people receiving Social Security benefits. That was because the 19-year-olds that he put in positions of great influence, the Muskrats, whatever you want to call them, didn’t understand government databases. You know, you get parachuted into an agency with access to the computer system but absolutely no knowledge of what the agency does or how it does it and then couple that with a kind of arrogance — believing that these people must all be stupid and I can just sit down for a day or two with their data and find vast waste and fraud. Well, nobody in a position of responsibility should believe that kind of thing. It’s possible that Big Balls and his other hench people actually believed that they knew what they were doing. But my god, if you’re put in charge of a hugely important government function, you don’t assume that everybody there is an idiot and that your neophyte attaches have somehow stumbled on things that nobody else noticed. And of course, Social Security is so pervasive, such a large part of everybody’s life, that the idea that there could be tens of millions of dead beneficiaries and nobody has noticed it, that’s completely crazy. You even wonder, did Musk really believe that? Does he even have a notion that some things are true and some things are not? But in any case, there you are. And so it was a total disaster. He left the government not, clearly not because Trump thought that he was too extreme, too bad a guy, but because it was so clear that he did not know what he was doing. And the reports of alleged savings from DOGE: it was starting to get embarrassing because it was so easy for news organizations to find out that the claims were utterly false, that none of what they claimed was happening was actually happening. So he left. and then he goes back to his companies and becomes at least temporarily a trillionaire with an enormous public offering. Why didn’t people think that his record with enormous public responsibility was somehow relevant to his financial future? I mean, if a guy who can convince himself that there are 20 million dead Social Security recipients, who can convince himself that you can massively slash foreign aid and it’s all waste and fraud and nobody will be hurt — why would you trust that person to run a company? And furthermore, the character flaws that are revealed here — flaws is what too weak a word, but anyway — when you have somebody who refuses to acknowledge uncomfortable reality, refuses to acknowledge error, who responds to any perfectly truthful statement that reflects badly on him by saying, I want that guy put in jail. — those are not the character traits that make for an effective manager. If you can’t accept that you are ever wrong, how are you ever going to get things right? Because things will go wrong, and you will make mistakes. We all do. So all of this seems terribly relevant, and yet it says something, I guess, about America that people piled in to SpaceX stock, although some of that has come off now. It really was clearly an early frenzy, a fear of missing out frenzy. There are now reports that SpaceX also sold bonds, which itself is a little troubling. Why should they be needing to go into debt right away? What is that about? And those bonds have already lost some of their value, which is much more serious than the stock coming down. When bonds lose value, that’s because people think that there is now a risk that this company might default, might not be able to honor its promises. So seeing those bonds start to trade at a discount almost immediately is a pretty bad sign for the company. But again, why did anybody believe any of this? Musk is a horrible, terrible person and has the blood of millions of children on his hands. Let’s be clear. Yes, it’s not something that has been proven, but it’s close to. It’s so overwhelmingly likely that it clearly has to be true. And he’s also a weak personality — very much like Trump again — he can’t take criticism, he can’t admit error. So what does it say ultimately about our society that so many people are willing to throw money at this guy and that they’re so willing to forgive the incredible failures that he carried out, the incredible disaster of his time in a position of public responsibility. And I don’t really know the answer to that. There’s a real question about how it is we got at our current age of irresponsible oligarchs and with so little public backlash. And it’s starting to develop. But still, the fact that Elon Musk is still in business, let alone the world’s richest man, is in some sense an indictment of all of us. On that happy note, take care. Get full access to Paul Krugman at paulkrugman.substack.com/subscribe [https://paulkrugman.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

27. juni 202610 min
episode Catching Up With Ro Khanna cover

Catching Up With Ro Khanna

For all my interviews and more, subscribe on YouTube [https://www.youtube.com/@PKrugman]. Ro Khanna represents a large part of Silicon Valley, and not surprisingly is a very smart guy. Perhaps more surprisingly, he’s also a very interesting progressive, who has drawn considerable ire from the tech lords, with Elon Musk most recently calling for his imprisonment. I caught up with him Friday: TRANSCRIPT: Paul Krugman in Conversation with Congressman Ro Khanna (recorded 6/26/26) Paul Krugman: I’m talking again to Rep. Ro Khanna [https://khanna.house.gov/], the Representative for Cupertino, as it were, representing the heart of Silicon Valley in ways that don’t always please the tech oligarchs. I had planned to ask about AI, but there’s so much going on and Ro is right in the middle. So, welcome to this interview. Ro Khanna: Well, I’m honored to be back on. I usually just read you to learn, but I’m glad we’re going to get to have another conversation. Krugman: As it happens, tech-related politics is really central now. And you seem to be in the middle of at least three big issues: Elon Musk, AI generally, and the California Wealth Tax Initiative. I want to talk about all of those, but maybe let’s start with Elon Musk, who has called you evil, which is a great honor. Do you want to talk about that controversy and where you came in here? Because I think it’s very interesting. Khanna: Well, he’s called me evil, he’s threatened to sue me, and he’s threatened to jail me. I have this quaint idea, Paul, that in a democracy, Elon Musk should have one vote. He doesn’t seem to think that, and the reason he has been so triggered is that I not only cited a Lancet study [https://www.thelancet.com/article/S0140-6736(25)01186-9/fulltext]—which said that his USAID cuts could potentially lead to the deaths of 4.5 million children and over 10 million adults—but I also cited an Atul Gawande/ Boston study showing that some of these deaths have already taken place [https://www.foreign.senate.gov/press/dem/release/the-dangerous-consequences-of-funding-cuts-to-us-global-health-programs]. This triggered him, not just because I cited these studies, but because I said he’s going to have to come before the House Oversight Committee when we take back the majority; we’re going to have the power to subpoena him. And, of course, defying a congressional subpoena could lead to contempt of Congress and penalties. And so he’s been spending the last few days obsessively tweeting about me. You would think if you had $1 trillion, you’d have better things to do, but this is what’s occupying him. Krugman: Yeah, let’s back up a bit. One of the things that I found really kind of astonishing in the whole discussion—and obviously, SpaceX went public and there was amazingly little discussion of Musk’s role at DOGE where he was a quasi-government official. I guess it was kind of weird what the legal basis for all of that was—but this had immense impacts. And as you say, one of them was that he just, more or less by personal fiat, eliminated USAID, which is our premier aid agency. Do you have any thoughts just generally about what Musk did at DOGE? I think it’s a hell of a story, so let’s start with that. Khanna: The keywords you used were “by personal fiat.” I mean, he literally went there, didn’t consult Congress, didn’t report to Congress, and just started cutting programs that Congress had explicitly authorized. And no one stopped him. We voted to subpoena him, but he defied coming in and explaining anything to Congress. By some accounts, he cut 83% of the programs that were at USAID. And some of these programs are to feed some of the poorest people in the world; some are to provide medicines to some of the poorest people in the world. So, you literally had the world’s richest person hurting the world’s poorest people. And he was doing it with no accountability, in defiance of Congress. Congress then fortunately restored some of these programs, so he was not able to end all of them. But the USAID programs are a shadow of their former self. They now are scattered in administration, and many were so disrupted that these academic studies have shown that it potentially could—or in some cases already has—led to the deaths of some of the poorest people and children in the world. Krugman: Yeah, the important thing is what he did. But the attitude also at the time... I think he said something like, “Oh, I just fed USAID to the wood chipper, and I could have gone to some great parties instead,” as if it was, you know, annoying that he had to go out there and cut off medical aid for millions of children. Khanna: Yeah, it was total arrogance. He said it was all fraud, but of course, he then didn’t have the guts to come before Congress or the American people and explain where he found fraud. He didn’t consult any of these programs. It’s not like he was on a plane to Africa or a plane to other parts of the world where these programs were being administered. And for someone who was going to go after cuts to the federal budget, instead of starting at the Department of Defense, which is 65% or so of the discretionary federal budget, he decides to start with an administration with less than 1% of the federal budget. It was a purely ideological agenda that, turns out, has real-world consequences—especially with this Ebola outbreak. I mean, one of the things he cut was the oversight and testing in places like the Republic of the Congo, and now we’re seeing the consequences. Krugman: And his reaction has been really quite over the top, considering, you know, if you’re any kind of public figure, you expect to be facing criticism. Khanna: It’s just denial, right? I think he said that not a single person has died because of his cuts, which is totally implausible. He said, “there’s not a single documented case.” And he said that everything he cut was simply a fraud, and that these academic studies are totally fraudulent. Granted, the Lancet study is a model of what could happen, but the Atul Gawande study is actually a documentation of actual deaths that have taken place. And there are a lot of anecdotal statements which Nicholas Kristof and a lot of people have reported on. Krugman: And now he’s threatening to sue you. Presumably, I don’t think we’re that far gone that there’s any chance that such a suit would prevail, but how much of that is an actual burden on you? Khanna: Well, I put it into Grok to see how strong a case Elon has, and Grok doesn’t think he has a very strong case. So there’s that. Krugman: In case anybody doesn’t know, Grok is Elon Musk’s or xAI’s LLM. It’s a competitor to ChatGPT and Claude, except it’s not really a competitor because it’s awful, right? But yeah. Khanna: I would have a better case of defamation given what he has said about me. But, of course, I believe in free speech, Paul. I thought he did, too, and I would never think of suing someone for calling me a robber or calling me names. That’s the First Amendment. But I’ll tell you what it does: it creates a doubt in other people who are on the Oversight Committee—you know, “Is this really worth the bother? Should we really criticize Musk?” So that’s one thing. Obviously, Musk has more than one vote. He’s got millions of dollars that he can spend on candidates, but now it turns out it’s not enough for him to just have the ability to support Super PACs; he also wants to be able to intimidate any public official who dares to go against him. It’s not just that he would spend money against them, but that he could actually sue them. And so if you’re a member of Congress, you’re thinking, “Well, do I really want this fight? Or maybe we could just focus on the hundred other issues.” So, it’s less about the headache for me and more about the signal he’s sending to other elected officials. Krugman: So you aren’t trying to do a GoFundMe for a legal defense or anything like that? Because I know people who have faced other spurious lawsuits and it’s actually cost them money, even though there’s no chance of it prevailing. They feel that they do need to hire people, but you’re not in a position where you are personally feeling liable? Or are you just well-positioned to sort of weather this? Khanna: Well, he hasn’t sued yet. If he does sue, it will be a drain on resources and we would have to raise funds. We would. But I don’t want to have people do something before there is an actual lawsuit. We’ll see what he does. But that’s exactly what his strategy is, whether it’s against someone like me or just a message to others that he has unlimited resources and he can make your life very, very difficult. Krugman: Okay. And he’s also called for you... I guess there was nothing specific about why, but for you to be put in jail, which is even more amazing. Khanna: Yes. And ordinarily you would kind of laugh it off because he’s a private person with no power. But of course, in this administration, him calling for that and the way the Justice Department works—there are political motives to how they’ve been operating. I mean, they have the governor of California and his wife that they’re threatening along with Adam Schiff... the list is long how they have operated. Krugman: Yeah. I’m a friend of Lisa Cook at the Federal Reserve and for her, this has been much more. She was, in fact, targeted and all of that by essentially the same gang. So yeah, it’s quite something. Just the last bit: Musk has also then gone out with this claim that USAID somehow is responsible for COVID, and also went full-in on the conspiracy theories about COVID. Do you have any comment on that, just since it came up in this context? Khanna: It’s so nonsensical. I don’t even understand what he’s talking about. I think what he is trying to argue is that the lab that some people believe was the testing ground for the virus somehow is connected to USAID, but he just puts these things out there with no evidence and for ideological reasons. The reality is the large part of USAID was to help poor people with food and medicine, and it had support from everyone from George W. Bush on. Krugman: Yeah. I’m pretty sure that USAID doesn’t actually spend money creating labs in China. Khanna: I’m 99% sure that’s true. I mean, Paul, you and I usually check things before we say something definitively. Elon doesn’t have any of those filters, so he’s just throwing these things out there. Krugman: Yeah. It’s pretty terrifying. The world’s richest man with a very strong political in with the U.S. government and... just, wow. Well, that was in the news so I thought I’d ask. Khanna: He does have a huge platform, right? I mean, he has 240 million followers. So him saying, “Okay, I’m going to sue Khanna. Khanna is a horrible, evil human being,” you know, has more reach by far than when I go on Meet the Press or ABC News. And he’s putting out basic falsehoods, so it’s a real danger. Krugman: Yeah. Okay, let’s move on. So the technology of the moment is AI. Last time we talked, which is a while back now—I mean, a while back in tech time, anyway—we were talking about crypto, and there’s a little bit of the distracted boyfriend thing where people are looking now at AI instead of crypto. But AI does look really much more substantive. You can actually almost start to see its impact on productivity, maybe on layoffs. So it looks like a serious technology. And you’ve been staking out a position which is calling for a lot more intervention and regulation. Do you want to talk about what you think is happening and what needs to be done? Khanna: Well, so far, AI has been enriching tech lords and tech billionaires, but it’s caused deep anxiety with ordinary Americans. And I would argue that there are four things we need to do. We need to first care about jobs. Now, here’s the good news, Professor Krugman: it used to be that the people who cared about a jobs program were folks in de-industrialized communities—blue-collar, or people who had lost factory jobs. Now you have kids at Brown, kids at Yale who are worried about whether they’re going to have a job. So I think there’s an opportunity for a broad coalition to have the most ambitious jobs agenda in a generation. And what does that look like? I would say first, it means taxing agentic AI more than we tax human workers. This is not my idea; it’s Daron Acemoglu’s idea [https://news.mit.edu/2020/3-questions-daron-acemoglu-us-tax-system-automation-1015], which is basically that the tax code is biased towards capital. If you have to hire someone, you’ve got to pay their health insurance and you’ve got to pay a payroll tax. If you want to have an agentic AI worker or a robot, you don’t have to pay that. So, neutralizing that tax code. Second, we should—and I’ve argued this—have a “Work for America” program, a federal jobs program for young people out of school, out of trade school, or out of college to rebuild communities, maybe to come to the federal government. Maybe they go to a community that they didn’t grow up in. This can be akin to military service and can help rebuild not just the physical infrastructure of America, but the social infrastructure. Third, bargaining power for employees. So, not just go retrain them, but give them an actual say in the company if there are going to be layoffs. What role will they have? If there’s going to be displacement, what jobs would they have? Are they going to get a share of the profits from the increase in AI productivity? Are they going to get time off with the increased productivity? And finally, a sense of ensuring that jobs are there, that there’s intervention in having humans in the loop in various jobs—whether that’s the four million truck drivers and thinking about their role, or whether it’s jobs making decisions about people’s healthcare or making decisions about their finances. Krugman: So, yeah, I mean, a few things to unpack here. One is, obviously, nobody really knows what this is going to be, but we are starting to see, or we think we’re seeing, real job impacts and income impacts from AI. Probably. If you had to say, where would we be seeing these things first? It would be kind of in your district, right? So, what do we actually see? What are you hearing from your own constituents? Khanna: First, it’s much harder to get hired into these tech jobs. There’s a lot of anxiety from 21-, 22-, and 23-year-olds and their parents. The job market used to be, even at a place like Stanford, “Okay, I’m going to get 10 or 15 offers before I’m done with my senior year.” Now, they’re lucky to get a job, or it’s much harder to get a good job. Second, there have been a fair amount of tech layoffs. Now, some people are arguing that was because they overhired in the pandemic and they’re correcting for that, but it’s hard to imagine that AI is not at least part of the factor in that, and that it’s not just a correction for overhiring. And then third, just the sense of what the new jobs in these tech companies are going to look like in terms of being able to implement AI or use AI, and what computer science is going to look like in different schools. Krugman: Yeah, it is interesting what you just said, which is that we have a better chance of getting action because the jobs at stake here are sort of high-education rather than blue-collar work. In a way, that’s an indictment of our politics—that in some sense, we think Stanford graduates feeling aggrieved carry more weight than ten blue-collar workers in Ohio. But on the other hand, it is really striking, right? Obviously, you’re hearing from people who are just seeing that entry-level jobs are not there. To what extent is this actually manageable? Can we channel this, or is this technology just going to sweep away efforts at, particularly, job retention? Khanna: I do think it’s manageable in that there are a lot of human tasks, in my view, that can’t simply be automated: goal setting, team building, and the origination of customized new ideas for settings. And there’s a lot of work, public work, that can be done—whether it’s opening new parks, whether it’s helping represent people who are underserved, whether it’s making government services better, whether it’s providing counseling, whether it’s providing teaching, or whether it’s providing childcare. So, in my view, there is a role for a robust federal jobs program, and it could help in de-industrialized areas and for factory jobs. And we should keep in mind, we wanted to do this years ago when we saw the devastating effects of globalization, but our politics, for whatever reason, didn’t allow it. And now you have a much broader set of people with anxiety. Of course, it’s not the Great Depression when FDR had 20 to 30% unemployment and a total collapse in demand, but it is one where you meet an average person who’s concerned about it. And I think there is polling showing 30 to 40% of Americans are anxious about jobs. That seems to me to provide a moment where a politician coming with a jobs agenda or intervention in the free market would have a reception which, in a lot of the last 30 or 40 years, has been very hard to get. People just say you’re interfering in the markets. Krugman: Just to say—I mean, you kind of implicitly said this—but in effect, you’re calling for something like a WPA (Works Progress Administration) or CCC, ‘30s-style, but at least in part for tech workers, not just for people with shovels, but people doing skilled—I hate that word—but high-education-content work. Have you put any kind of numbers to this, or is it just a general outline at this point? Khanna: I wrote an op-ed called “Work for America” in The Wall Street Journal, and it was about $50 billion a year, which I said you could fund through an AI token tax. And it would be hiring anyone out of high school or out of college for jobs to open a park, to help with their local community, to teach, or to come to the federal government to do something. What I was particularly excited about is that kids growing up in Fremont, in my district, could go to Middletown, Ohio, to do something there so that you’re building things. And it would help for folks who may be displaced. And I explicitly said it was inspired by FDR’s Works Progress Administration, which hired 8 million people. Of course, that’s where the “boondoggle” idea came from, because some people back then said some of those jobs weren’t real—they were criticizing it. But my understanding, and you’re a better student of history, is that it did work in creating meaningful employment and many meaningful projects, and certainly helped the social infrastructure of the country. Krugman: I want to come back to jobs in a second, but you’re basically at least accepting as a strong possibility that this technology is biased towards capital and away from labor. Are you seeing that? Is that really what’s happening? Khanna: We’re certainly seeing it in terms of the explosion of wealth in my district and with billionaires. And we’ll get to the idea of a billionaire tax, but I mean, they have reaped massive amounts of benefit from the AI revolution, and we haven’t been seeing that for the average worker or even the average tech worker. They’re not reaping the rewards in the way that a few people have. And you’re seeing this also in terms of, certainly, the difficulty in entry-level jobs. I mean, when I was at Suffolk University and giving one of the commencement speeches, the line that got the most applause was when I said we need to tax agentic AI more than human workers. Young people are concerned about AI, and I don’t think their fears are totally irrational; I think they’re finding the job market to be harder. And I have a lot of cases in my district of people at these tech companies who are being laid off or told that they need to be let go. Now, you talk to the tech leaders and they’ll say there are other factors too—they’ll cite overhiring in the pandemic, they’ll say they’re just adjusting. So, I don’t know if there are academic studies that show it’s correlated completely to AI, but I certainly think it is one of the variables. And I think there was one study at Stanford showing that for young people in automatable jobs, AI had contributed. Krugman: Okay, you gave a commencement talk and got a positive response, unlike Eric Schmidt and, there have been multiple instances, but I guess Eric Schmidt is the famous one, the former CEO of Google, giving a commencement address in which he started to talk about AI and immediately got massive boos from the students. Khanna: I took the opposite tack. I said AI is not doing your generation a good service, and it’s something that we need to be tackling—not preaching all the benefits of AI. And it was a surprise to me because that was not the place where I expected to get applause. It was not the central part of my address, but the two places that got the most applause were calling for a billionaire tax and calling for a jobs program and taxing AI, which I was almost going to keep out of it because I thought, “Is that too political?” But the students, actually, that’s what resonated with them. Krugman: And so, at least conceptually, there are two separate issues. There’s a wealth tax, which I want to get to in a bit, but you’re talking about essentially—you call it a token tax—a tax basically on the use of AI. Are we able to implement that? Do you think it can be done reasonably well? Khanna: I do. It seems to me that’s the easiest thing because right now there’s a cost, of course, to the use of AI. And there’s a large debate, by the way, about what that cost is because it’s fairly expensive. It turns out it’s fairly expensive in terms of the energy consumption of AI; it’s expensive in terms of the capital expenditure for data centers, which is a whole separate conversation. And so, the question of labor displacement, I think, also depends upon how much AI costs actually come down or don’t come down. But right now, companies are paying a lot for the use of these tokens, which is basically the output of AI when you type something into ChatGPT. And so, if you just put a tax on that, that would both disincentivize automation and would raise revenue. Krugman: I’m not aware of an earlier parallel where there was something—sort of an output of machinery at some level—that could be compared in a way with labor. And of course, aside from income taxes, the FICA on every paycheck shows that we tax labor. And you’re just saying that we should do something for the stuff that’s coming from AI capital, right? Khanna: Yeah. That’s exactly right. And simply put, the idea right now is that it’s not just that people have a higher degree of variability because you could get sick, you need to be with your kids, or you have to pay health insurance. We’re not taxing what these tech people are saying is labor-replacing, and so we should tax that. Krugman: Okay. Now, people’s immediate reaction is, “Oh, but we’re in a competitive race with China.” What’s your answer to people who say, “Oh, you know, if we start to tax this stuff, we will forfeit the lead to other countries. It’s a great international race.” Khanna: Well, first of all, even China is changing its policies. I read recently that some of the court decisions in China are saying you can’t lay off people based on AI, and they have almost 18% youth unemployment. When I went there, a lot of the young folks didn’t want to work in the factories, and they’re concerned about losing jobs. So, I think China itself is realizing that having just unregulated AI is not healthy for society. The second thing is we want to compete with excellence. That’s always been the American aspiration—that we want to have products that have the highest standards. We want to have high safety standards, the highest set of standards in terms of privacy. So, if we’re producing AI that is safe, where agentic AI isn’t going to go do crazy things and isn’t going to engage in surveillance, then that should be something that we can export and be a model for the world. I don’t think we have to have a race to the bottom in the type of AI we produce. Krugman: Okay. And AI that’s safe, which, of course, is one of the big concerns. Any thoughts on the runaway models? Grok, which you mentioned, apparently was used for targeting in Iran with not-very-good results. Are you hearing anything, or is there any movement on intervention—basically congressional action to try and avoid some of these dangers from AI? Khanna: There hasn’t been, because this administration has basically said, “Let the tech billionaires do whatever they want.” The only time they’ve shown any interest in regulation is with Mythos, Anthropic’s latest model, which could detect cyber vulnerabilities. And it’s unclear whether their concern is simply motivated by the unsafety of Anthropic’s model or is retribution because Dario Amodei got into a fight with Pete Hegseth. But other than that issue, the administration has basically said, “Do whatever you want.” And it’s really scary because usually, even by these tech leaders’ own worries, they say, “This is transformational. This is going to change the world. This is the most important technology since fire.” Well, if that’s really the case, we have a federal agency for electricity, we have a federal agency for nuclear weapons and nuclear power—why wouldn’t we have a federal agency for AI, on your own terms? And yet there’s been no effort to do that. Krugman: Okay, for listeners, by the way, Amodei is the CEO of Anthropic. The two big models out there are OpenAI’s ChatGPT and Claude, which is Anthropic. Most of the buzz that I’m hearing about usability involves Claude, but Anthropic is politically not that aligned with the administration and has particularly said that it will not allow its AI to be used for autonomous weapons, and that has made them on the outs. And it’s really very hard, right? When the administration lays down rules or policies on AI, you can never tell whether they’re really concerned or whether they’re just trying to punish a company that isn’t on their side. That’s what you’re saying about Mythos, right? Khanna: Exactly. And I mean, given the administration’s history in general on retribution across so many places, but also in this explicit retribution against Anthropic... there, Amodei basically said that he didn’t think technology should be used in a way that would violate privacy. He didn’t think AI should be used to make decisions about what to strike without human judgment. Hegseth didn’t like that; they had a whole fight. And so now that they have Mythos, it may be that there really should be regulations and export controls because this technology is explosive and could cause cyber vulnerabilities. The problem is we don’t know, because the administration also has a motive for retribution, and they’ve lost the credibility of any independence. Krugman: Yeah, that makes it especially hard now. All right. It’s actually amazing how much impact Anthropic’s products are having. I’ve been talking with senior financial types on stuff, and it’s amazing how often I hear, “Well, I was thinking about that, so I asked Claude.” It really is shocking how—you know, we’re not talking about saying, “I had my staff go and look it up,” it’s, “I went and asked Claude myself.” So, like it or not, this is the world we’re in now. Okay, it seems to me that your whole vision is a step beyond. I mean, if you go back to actually quite early on when they were still making apocalyptic warnings and Sam Altman was saying, “Oh, well, given AI, we’re going to have to have something like—” I don’t think he exactly used these words, but something like, “We’re going to have to have taxes on capital to pay for universal basic income.” And that’s kind of the Silicon Valley vision. But your idea is more that we should have taxes on the wealth that’s been created to help provide for job programs. So, it’s not just that we’re going to give people money so they can sit at home and let the machines do stuff, but we’re going to subsidize ways that give people work. Khanna: Absolutely. And that work could be childcare, it could be home care, it could be new types of industry, it could be helping provide better government services, or it could be doing something meaningful in the community. I believe we have the need for productive work, and that the federal government should play that role. And by the way, the hypocrisy of some of these tech folk saying, “Just tax me so we can have universal income”—well, they’re not willing to pay the tax. I mean, when you look at Sam Altman’s proposal on universal basic income, which is, “Take a 2% tax on my company every year in terms of equity shares,” if you just did the math on that, after five years, maybe every year, each American would get about a $1,000. That’s not exactly universal income. So, I’m not for just taxing and giving everyone a check and saying work doesn’t matter. I don’t think that’s a healthy society, but they’re not even willing to do the first part of that, which is pay the tax. It’s just empty rhetoric. Krugman: Yeah. I always had a problem that these proposals for UBI—even if they raised enough money—the amounts are not enough to live on, and also just collecting what we used to call welfare is not a substitute for actually having meaningful work. So, let’s talk first about the California proposal for a one-time wealth tax, which you are supporting, but is amazingly controversial even within the Democratic Party. Tell me about the proposal and some of the criticisms. Khanna: There are three million people in California who risk losing their healthcare because of the big, ugly bill that Trump passed, which everyone acknowledges cuts Medicaid and cuts the subsidies in the Affordable Care Act. So, that’s a fact that everyone acknowledges—that these folks are going to lose their healthcare. The second fact that people acknowledge is that there are about 200,000 healthcare workers—nurses, aides, hospital workers—who are going to lose their jobs. And what this program, this ballot initiative, says is: let’s have a one-time 5% tax on billionaires. There are about 250 of these billionaires. Their worth, as Gabriel Zucman’s work shows, is about $2 trillion. That is the equivalent—and I’m not saying it’s the same thing—but it’s the equivalent of about half of California’s GDP. There are 250 people who are worth that. And if you tax them one time at 5%, you could literally raise about $100 billion and make sure that we cover all of these Californians, and that we don’t lose 200,000 jobs. The ballot designers went and said, “Okay, let’s just do 2%,” and that proposal was rejected. That would have raised $40 billion and staved off the crisis for two years. And so now we’re going to the ballot on this. These 250 billionaires, by the way, have made about 150% over the last three years. Their wealth has increased 150% largely because of AI, and yet they’re not willing to pay a 5% one-time tax to make sure that Californians don’t lose healthcare. Krugman: It always astonishes me how small the number of people that we’re talking about is, right? It still annoys me when people talk about the 1%, because we’re talking about a tiny, tiny fraction of 1%—just 250 people in California. But it’s a quite significant amount of money that could be raised by such a tax, right? So, the first question people ask is: won’t they all just decamp, leave? What would be the possibilities for avoidance—not evasion, since evasion is illegal, but avoidance is not—so that everyone won’t just pull up stakes? Khanna: So, first of all, we have actual data on this. We know that in Q1 of 2026, 85% of venture capital in America went to California—the highest ever. And this is months after the state ballot initiative was announced, and when you’re seeing reports of Sergey Brin and others leaving. So, in terms of capital investment into California, it has only increased since the announcement of the ballot initiative. And that’s obvious; no one thinks that the AI revolution is happening in Miami or happening in Austin. It’s happening in Silicon Valley. It’s happening in my district and the surrounding areas. So, you may be losing some individuals, but you’re not losing the capital into Silicon Valley, and that’s just what the data shows. The second thing is, okay, maybe you lose some of these individuals, but as Zucman’s work has shown, these billionaires are only paying about 2.5% of the total general fund in California. And the reason they’re paying so little is because they basically weren’t being taxed—I mean, they don’t have income. And so, if you lose a few people, it’s not some devastating blow to the tax revenue of the state, and you’re not losing the capital investment. And the final point is, if you haven’t moved already, you’re subject to the tax the way it was designed: it’s one-time, and it’s based on whether you were in the state by the end of last year or not. Krugman: Okay, that’s really important. It’s a retroactive tax, in a way. It is a levy, but that’s kind of okay, so there would be no possibility of people avoiding it. But I guess one criticism has been that while they can’t avoid this tax, they won’t pay income tax in the future. But you’re saying that they basically weren’t paying income tax before. Khanna: And the irony of it is, what’s the point of making money? Part of it is you get to do things that you want to do. One of the most basic things that people want to do is live where they want to live. And the idea that you would be a billionaire and then not want to live where your family is, or where you like, or where you grew up, or where you find it most fulfilling simply because of tax considerations seems to be quite ironic. And the truth is that there are a lot of billionaires who will grumble and say all of that, but aren’t going to be leaving California. Krugman: One of my favorite lines was about the attempts to turn Miami into the new Wall Street. There was some Wall Street guy who told Bloomberg, “The trouble with moving to Florida is that you have to live in Florida.” There’s a California version of that. So, this would be a one-time California thing. Do you have a vision for what an attempt to kind of make AI and just general technology less of something for a few hundred people would look like? What would America 2035 look like if we could have a Ro Khanna vision of policy? Khanna: We would have a new social contract. We would be taxing these billionaires and trillionaires, and that would raise about $4 trillion if you did it at 5% a year. You would have other basic taxes—have an actual effective corporate tax rate that is at least 28%; right now, they’re not even paying 21%. You would have capital taxed the same as ordinary income. You would have a step-up in basis. You’d raise that revenue rate— Krugman: We should mention “a step-up in basis.” Why don’t you explain what it means? Khanna: Well, that’s when these people die and their kids get their estate. But if they had huge stock appreciation in their lives, their kids don’t have to pay taxes on that stock appreciation. Krugman: Yeah. We’ve got a system in which a large part of capital income is basically never taxed. So you’re talking about eliminating that. Khanna: Why would we have a system that’s already capital-biased, where basically, if you have this capital, you’re making money in your sleep and you’re paying less taxes than someone who’s a doctor or nurse or a factory worker who pays ordinary income tax? That should be leveled. And when people say, “Do billionaires deserve what they make?” I don’t deny that they have built something often of value, and that they’re hard-working, and that they’re entrepreneurial. I’m just saying that the system—because of the way we tax capital less, because of the way that corporate taxes aren’t really collected, because of the fact that we don’t have a wealth tax, because of the way we have allowed the estate tax to operate—has allowed for the accumulation of extraordinary wealth beyond what a system with a rational tax code would allow. And so if we had a rational tax code, we’d have all of this revenue, and then you could do things like having universal childcare at $10 a day, having a thousand new trade schools, having free public college (which we had in California in 1960, and in many places as well), having a jobs program, making sure that we had a livable wage and union bargaining power, and expanding healthcare. I mean, I’m ultimately for a single-payer, Medicare-for-All system, but at least expanding it, doing things like dental, vision, hearing, and making sure that we had drug negotiation. All of this is to say something very simple: when I go around the country and I say Elon Musk has become a trillionaire, I’m met with huge boos. And when I talk about these tech billionaires, huge boos. That was not always the case in America where people would just boo successful business leaders. It should be a wake-up call that most people don’t think that their lives are improving, even though we’re generating more wealth than ever before. And my view is: why can’t we have a society, if we’re generating all this wealth, where most Americans feel like they have more economic security? And how do we do this? And the last point I would say is, I’m the nice guy. I’m 49 years old, about to turn 50. You know, the folks in their 30s, the folks who are winning in New York, they’re not as nice as me saying, “Okay, let’s just have a new social contract.” They want to rip the total system down. They’ve had it. They want a total revolution. And so, either we’re going to have this transformation, or we’re going to have a far more radical new generation that is totally upset at society. Krugman: So, you’re basically saying you can do these reforms, you can do something that will spread the benefits, create societal sharing, or the pitchforks and torches will be coming for you. Is that a good way to summarize it? Khanna: [Laughs] That’s my message. I’ll say pay it as an anti-revolution tax. But you know what? Even in my district, Paul, when I have town halls and I say, “What do you think of a billionaire tax?”—and I remember in one of the most affluent districts in the world—90% of folks will raise their hand: “Yes, it’s a good idea.” To your point, this is not talking about the 1%. I can’t do the math, but the 0.0001%. Everyone wants them to pay taxes. The doctors do, the investment bankers do. And then there will be people who say, “No, Ro, I disagree.” I say, “Why is that?” And they’ll say, “Well, why is it just 5%? I want 20%.” I mean, they’re not thinking of the wealth tax necessarily and what consequences that would have. But this is the sentiment, not just in Pennsylvania, Michigan, or Ohio; this is the sentiment in my district. And I think a lot of people are oblivious to the anger and the anxiety young people have. They can’t buy a house, they have huge debt, they don’t think their lives are going to resemble their parents’, and don’t understand why that’s the case in a nation that’s producing so much wealth. I mean, you’ve done a lot of work on this, and I’d ask you, in development economics and often in the developing world, there’s a trade-off between economic development and economic fairness, right? But it seems to me what’s so ironic in our case is that trade-offs don’t need to exist. We’re producing all this wealth; it’s simply a matter of values that we’re not allowing most Americans to have economic security. Krugman: That might be a good coda here. I mean, it is an extraordinary thing that we don’t seem to be facing a trade-off. It really is the case that in almost every respect except the wealth of a few hundred people, this kind of fairness agenda looks positive. So, how are you feeling about the politics of it? Do you think you’re getting traction? Khanna: I do. You know, they poured in $1 million-plus against me with my primary opponent [a Democrat who opposed the wealth tax]. And California’s a weird system: Democrats, Republicans, we all run together. I got 62%, my challenger got 6%, and the Republicans got the rest. So, I think that was a bit of a wake-up call for some of these folks that, you know, democracy still works. And I’m very, very optimistic heading into the midterms that this central idea of fairness is one that’s resonating with many people. And I am confident we’re going to take back the House. Krugman: Okay. The congressman from Silicon Valley says democracy may still work. I think that’s a really optimistic punchline. Thanks so much for talking with me. Get full access to Paul Krugman at paulkrugman.substack.com/subscribe [https://paulkrugman.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

27. juni 202648 min