Free with a sign-in.
The best 6 min of 51
The Economic Risks Keeping Paul Krugman Up at Night | Prof G Markets
Paul Krugman says the AI boom, weak underlying demand and political uncertainty leave the economy exposed, while climate action remains urgent.
Prof G Markets · Scott Galloway & Ed Elson
- Paul Krugman is a Nobel Prize-winning economist and distinguished professor of economics at CUNY.
The best minutes
Also worth watching
- 11:14Relative poverty better captures exclusion than a single income cutoff5 min
- 29:51Wall Street sees political risk in Washington more clearly than tech5 min
Skip the eight-minute opener of sponsor reads and host banter, the long mid-episode sponsor break, and the closing discussion of media models.
Full analysis · 6 min read
Real Alpha · Nov 8
Paul Krugman puts data center spending near one to two percent of GDP
- Krugman puts data center investment at something like one to two percent of GDP.
- That scale matters, but he says it cannot alone produce a crash on 2008’s scale.
- Teams should test whether AI demand and financing can stand without a public rescue.
Free with an account
Read the rest of the analysis
6 sections, the 3 other voices it tests Krugman against, and what it means for you. Every episode we cut gets one.
Check your inbox
We sent a link to your address. Click it to finish signing up. Not there in a minute? Look in spam.
or
Signed up before? The same buttons sign you in. The episode, its moments and the transcript stay free.
2025 put tariff upheaval and AI investment in collision
Krugman calls 2025 a weird year because unusually large forces are arriving together. He says 90 years of U.S. trade policy were abruptly discarded for very high tariffs.
At the same time, he calls the AI boom the biggest thing since the housing bubble. He says history rarely offers a clean experiment, and this year’s forces interact.
That makes it harder to assign economic weakness to any single cause. A business plan built around prior trade rules may now face different costs and demand assumptions.
For example, the hosts sketch Nvidia falling 60% or 70%, followed by steep market losses. That is Scott’s scenario, not Krugman’s forecast or probability.
Krugman’s narrower warning is about concentration in economic decisions. He says data center spending depends heavily on capital spending choices by a handful of people.
Our read
The episode’s strongest lens is exposure, not a confident recession countdown. When policy shocks and private investment overlap, forecasts based on one driver can mislead.
Real income rose while access to housing and work tightened
Krugman says typical real income is a little above its pre-pandemic level. That challenges the simple story that higher prices have made everyone poorer.
But he also points to higher interest costs and the cost of buying a first home. The squeeze can persist even when an average income measure improves.
He describes a frozen job market. People who already have jobs may be stable, while new entrants and workers who lose jobs struggle to find another.
The host raised Michael Green’s proposed $140,000 poverty threshold against the official measure of a little over $30,000. Krugman said he had not read Green’s article and thought changing the cutoff missed the point.
Krugman distinguishes affording essentials from being included in broader society. He says measuring poverty at less than half the median income shows poverty rising as inequality increased.
Our read
Averages cannot show who is shut out of housing, work or middle-class life. Leaders should track those pressures alongside real income, not treat one measure as the whole economy.
AI spending props up demand, but limits direct crash scale
Krugman puts data center investment at something like one to two percent of GDP. He thinks the United States would probably be in a recession now without that spending.
That is large enough to influence demand, but he says it is not enough to produce a crash on the scale of 2008. He compares the possible damage more closely to 2001.
He warns that falling stock prices and sentiment are harder to measure. The hosts’ claim that 10 companies represent 40% of the S&P, and their market drop scenario, remain host framing.
His aggregate estimate does not show how concentrated the financing is or whether customers can support future spending. The episode gives no firm by firm returns or financing breakdown.
Ed Zitron on The Diary Of A CEO in August 2026 described major tech firms financing AI customers. He said their spending helps justify more data center building.
That comparison suggests checking who funds customers, but it does not establish that the spending Krugman measures is circular. It sharpens a diligence question rather than proving his recession scenario.
“we're talking about something like one to two percent of GDP.”
Our read
Krugman narrows the likely direct damage, not the uncertainty around markets or sentiment. Treat the GDP figure as a measure of scale, not proof that the investment is safe.
Debt markets depend on confidence, not a fixed ratio
Krugman gives no debt to GDP trigger for a crisis. He says the limit comes when markets conclude the country is no longer serious about controlling its obligations.
He points to Britain’s debt at 250% of GDP after World War II. He says it avoided crisis because people expected its leaders to do what was necessary.
That comparison does not give the United States a safe borrowing limit. Krugman says advanced countries borrowing in their own currencies can have room, but political stability and rule of law matter.
He describes a political deadlock between popular programs such as Medicare and Social Security, and an unwillingness to collect enough taxes to pay for them.
The host’s example of $5 trillion in receipts and $7 trillion in spending was hypothetical. Krugman also says historical cases offer little guidance on how a politically paralyzed country’s debt problem ends.
Our read
Fiscal capacity is tied to institutional credibility, not just a ratio. The episode gives no timing or threshold for a market break, so budget plans should not assume one.
A public AI rescue would face a political test
Krugman says an administration could be tempted to support AI investment, but a bailout would be politically difficult. He recalls Congress initially voting down TARP in 2008.
He says an AI rescue would be far larger and harder to justify than a limited bank bailout. He also describes this as an unusually unpopular boom.
The political question affects infrastructure plans as much as the technology forecast. A company that assumes public financing will sustain demand is relying on votes, not just returns.
David Friedberg on All-In in November 2026 said data centers could eventually be treated like utilities. That forecast makes infrastructure access a possible political battleground, not a settled policy.
Sam Altman on Sources in September 2026 said delivering everyday value is the right response to public skepticism. That is a product and acceptance case, not evidence that Congress would approve a rescue.
Krugman also calls tech leaders’ White House courtship crony capitalism rather than coherent AI policy. He says Wall Street is more alert to political risk because it has dealt with government longer.
David Friedberg · All-In · Nov 7AI compute could become a regulated utility
Sam Altman · Sources · Sep 1AI backlash calls for more everyday value from the technology
Our read
Public approval is a constraint on financing, not merely a communications problem. Neither a utility forecast nor a product pitch establishes that government support would arrive.
Climate damage accumulates, while clean energy offers upside
Krugman says greenhouse gas emissions accumulate. Even if emissions stopped completely today, he says temperatures would not start to fall for generations.
He expects substantial property and economic damage if emissions continue. That is a long horizon for many business plans, but not a reason to treat the exposure as temporary.
He separates that slow risk from his immediate political concern. At 42:11, he says he worries whether America will still be a democracy next year.
For example, Krugman says China is taking the lead in green energy while the United States keeps defunct coal plants running. He also sees a path to stabilizing the climate at a tolerable level with new energy technology.
Take his productivity example. He says a persistent 150 basis point rise in U.S. productivity growth would ease many economic problems, but he does not forecast that AI will deliver it.
Our read
Krugman separates the technology’s potential value from the market value of its builders. Climate exposure and productivity gains both belong in planning, but neither validates current AI valuations.
What it means for you
- Map AI demand. Measure how much of your growth plan depends on a small group of AI customers or infrastructure decisions. Stress test a pause in spending by one major buyer.
- Trace customer financing. Ask who funds customers’ AI spending and whether demand depends on further investment. Track committed capacity separately from expected demand.
- Measure access pressure. Track first home costs, borrowing costs and hiring time alongside real income. Krugman says those conditions can reveal pressures headline measures miss.
- Plan without a rescue. Model a slowdown without assuming public support arrives. Krugman’s account points to both an uncertain political process and an unpopular industry.
What to watch
- Whether data center investment produces measured productivity gains beyond the firms making the largest spending decisions.
- Whether AI demand rests on end users or financing among companies building and buying capacity.
- Whether housing costs, borrowing costs and job access ease or continue to squeeze households.
Send this to your CFO and whoever approves AI infrastructure or policy exposure.
Watch the minute where Krugman sizes data center spending against GDP.
Written by Real Alpha from the episode's transcript and the other moments it links, then checked against the transcript. Quotes are the speaker's exact words.
Minute by minute
- 6:48Tariffs collide with an AI boom
Krugman calls 2025 a highly unusual year: 90 years of U.S. trade policy gave way to very high tariffs as an AI boom gathered force. He says the two developments are interacting, making the economy unusually difficult to read.
- 7:58Affordability persists despite rising real income
Krugman says the typical person’s real income is slightly higher than before the pandemic, so a simple account of prices making everyone poorer is wrong. But higher interest costs, the cost of buying a first home and a frozen job market leave people struggling to get by or find a new job.
- 10:18Poverty needs more than one measure
Krugman says the official poverty measure began as a quick calculation based on food costs and has since been adjusted only for the consumer price index. He favors looking at both whether people can afford essentials and whether they are excluded from the economic mainstream; measuring poverty as less than half of median income shows it has risen with inequality.
- 15:35Debt risk turns on confidence
Krugman says weak demand may have persisted without large budget deficits, and that the U.S. can borrow heavily as long as markets trust its institutions and its willingness to control debt. He says the risk is not an arbitrary debt ratio but whether the U.S. is seen as a serious country, while political deadlock leaves popular programs such as Medicare and Social Security without enough tax revenue.
- 28:04Data-center spending limits crash scale
Krugman says economists have a poor record of predicting recessions and cautions against assuming an AI downturn would match 2008. He estimates data-center investment at about 1–2% of GDP, saying a pullback would be unpleasant but more comparable to 2001 than 2008.
- 29:51Krugman calls tech ties crony capitalism
Krugman says there is no clear AI or broader economic policy in the administration, and describes tech executives seeking favor as crony capitalism. He says Wall Street has been less visible, while Silicon Valley has less experience dealing with government; he also argues that some tech billionaires want to restore the sector’s former public standing.
- 39:38Climate risks persist as green technology advances
Krugman names environmental damage as a major risk, saying cheaper green technologies could limit climate change while the U.S. turns away and China takes the lead in green energy. He says the more immediate worry is whether the U.S. remains a democracy, while cumulative emissions threaten lasting climate and property damage; as an upside, he says AI could raise productivity growth by 150 basis points.
- 45:11Independent media still needs legacy reporting
Krugman says Substack gives him more freedom and lets him publish charts quickly, but independent media cannot replace the reporting he relies on from outlets such as The New York Times and Bloomberg. He calls the Times’ paywall model a success but says it works for relatively few organizations, leaving local reporting vulnerable.
Published Dec 5, 2025.
Transcript · 37 min read
Prof G Markets, hosted by Scott Galloway and Ed Elson, with Paul Krugman.
Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there. Where the podcast’s audio and the video part ways, some times are a minute or so off.
on average people spend shopping online at work. I hate those little reminders that people who bought this also bought. Okay, those people are also emotionally fucked up in the head. Stop Stop profiling me. Cash is trash. Stocks look pretty attractive. Something's going to break. Forget about it. Ed, I made that up myself. Yeah, I could tell. You could tell I made it up myself? I had a great joke on, what was it? Oh, there's the Prof G Show. I don't know if you know this. I do other podcasts, Ed. That's right. And it was that I watched Jaws last night, but instead I watched it backwards, and it's actually the heartwarming story of a shark that helps disabled people recover their lives.
i think that's genius really dialing it back and this was the the question last week should we dial it back i thought the answer was no i guess the answer is yes we've got a dad joke about jaws and about shopping would you trust the judgment of someone who takes the time to comment on a youtube video i mean we love our fans we love our fans love them but i'm not listening to any of you. No shock. I always try to be positive in comments. I'm not getting into fights anymore with people online because I'm usually fighting against a bot. And then I find my erectile dysfunction gets even more dysfunctional. And so I'm telling you how it's coming.
You laugh now. You laugh now. Just wait, my friend. Just wait. Teladnophil, which by the it's a generic name for Cialis. Just get used to it. And it's less embarrassing to say to your nurse who's got, or your nurse, your doctor who always has like an attractive nurse in the room and you have to say, I'm kind of interested in those drugs that, um, what is it? How you say, make dick hard. I don't know who I'm imitating there. I just know that's all kinds of wrong. You know what? Actually, our producer's fucked up. You know what today's number should have been? what 30 why's that ah okay that's right there we go ed elson 30 under 30 according to forbes which by the way i did not know forbes still existed but it is still very prestigious very prestigious and katherine uh we're on another podcast rage moderates she came on this is such exciting news ed just got 30 under 30 on forbes and i literally reflexively shout it out that bit shows me everything it's true i do everything tell us how does it feel to be the 30 over 30 or 30 under 30 yeah it's nice you know the secret of the 30 under 30 is it's actually 600 under 30 Keep on increasing the number of us every year.
Read the whole transcript
So, I mean, that's the sort of little footnote that we should consider, but still very exciting. Well, you joined past alumni, including Mark Zuckerberg, LeBron James, Miley Cyrus. LeBron James. He and I have a similar career trajectory, I think. I think you're much more similar to Miley Cyrus, for that just mean the guy who started Spotify. That's pretty good company. Also, Elizabeth Holmes and also Sam Bagman-Fried. I mean, if anything's going to happen to me, I'm going to end up in prison in the next two years, if we're being realistic here. Let's bring this back to me. I peaked pretty early.
I came out of the Gates Strong out of business school. I started a strategy firm. Someone got me invited to Davos. I was literally 29 at Davos, and I thought it would just keep going uphill, just up and to the right. And by the time I was 33, I was divorced, broke, and basically living like a caveman in New York, like occasionally leaving to try and go to the ready teller and get food and pursue sex unsuccessfully and then going back, retreating back to the cave. So that's what's coming next to you, Ed. That's what's coming. That's next up. Next up. I can't wait. You just got it. You can't let it get to your head.
Oh, that's too late for that. You're one of those quiet, conceited guys. You're always like, oh, well, I don't think about me, me, me all the time. I have to make you think that I don't think about me all the time. That fake British accent is about says it all. That says it all. Claire and I know you're from fucking Alabama. Anyways, what do we got going on today? We got a conversation with Paul Krugman, which is going to be great. And I would also add that we are recording our annual Ask Me Anything episode next week. So please send in your questions to markets at proftymedia.com or drop them in the comments.
Very excited about that episode. I think that's all I got before we got this conversation with a Nobel Prize winning economist. More importantly, along the same lines, did you see me on Oprah? I didn't see you on Oprah, but I saw the photo of you and Oprah. Is it out yet? No, she and I just rolled together. What do you think? Yeah, I was on her show. No, I had to send you on the show. I just haven't seen the interview yet. We've decided to come public with our relationship. We were talking in the side channels. It strangely looked AI generated, that photo of you and Oprah Winfrey. Yeah, it did look fake, didn't it?
Everyone looked so happy. It didn't look like repute. I had no idea what to expect. I know you want to know more about this. I do, actually. walk into a room, and they have filled the room with like 100 young people, 80 of them are men. Wow. And they all proceed to ask me questions about mental health and what should I do? I'm having trouble connecting to relationships. And I'm like, dude, I don't know. When I was here, I was just getting fucked up and approaching people in Irish bars. I don't know. It was, I was literally sweating. And I had to basically start every sentence with, I am not a licensed therapist.
I'll just give you my experience and if it's helpful great but and at one point I said to this woman I'm like I don't know I have no idea it sounds like you're really you know it sounds like you're struggling I have no idea it was so uncomfortable I think someone told them I was Dr. Drew or something or Esther Perel you got to assume that position if you're going on Oprah you're not going to be doing hard-hitting economics you're going to be talking about feelings, right? My friend Gail King got me on the show. Gail King's my new best friend. She's super cool and nice and funny. Anyways, I was on Oprah. That's very cool.
That's it. And now Paul Krugman. Let's get into it. Here is our conversation with Paul Krugman, Nobel Prize winning economist and distinguished professor of economics at the Graduate Center of the City University of New York. Paul, thank you very much for joining us for the very first time on ProfG Markets. Oh, hi there. Good to be on. Scott Clillian has an issue with the intro. We talk about Paul so much. I assumed he'd been on seven or eight times. It probably feels as if we know you. We're parroting your data so much. Well, we're slightly short on time, so I want to get straight into the questions. I have so many questions for you, Paul. And the first is, you know, we're coming to the end of the year here and a lot has happened. We've had a new administration. We've had Liberation Day. We've had tariffs. AI has exploded. You have been writing about economics for years. You're a foremost leader in economics. Just at a very broad level, what has stood out to you in 2025 and how do you think 2025 is going to be remembered in economic history i mean it's a weird year and that's all i can say i mean um history doesn't usually give you you know clean experiments it's usually more than one thing going
on um but this year is really extreme you know the one And you have 90 years of U.S. trade policy abruptly thrown into the waste bin and replaced with very high tariffs. And on the other side, you have this AI boom, which is the biggest sort of, well, it's the biggest thing since the housing bubble. But the housing bubble was kind of on its own. And here we have the AI boom sort of colliding with and interacting with the tariffs. And it's a very, very bizarre economy right now. One of the things you've been talking about lately is affordability. And that's something that's been very prevalent in politics, most notably probably with the mayoral election.
Where do we stand in terms of affordability in this economy right now? Clearly, we're seeing some growth in the overall economy because of AI. But the affordability crisis seems to be a kind of a different thing. The problem that we have in talking about affordability is that there is a simple thing, which is just your real income, typical person's wage divided by the consumer price index, which is actually up a little bit since pre-pandemic. So we're actually not seeing the simple version, oh, prices are way up and everybody is impoverished, is not true. Way too much reporting talks as if that was the story.
But that doesn't mean that everything is okay. So once you start to look under the surface, you see, first of all, there are some real cost of living issues or just generally getting by issues. There's interest costs are way up, certain really critical things. Above all, I would say that the cost of buying your first house are way up. So people are not imagining that there's an affordability crisis. And then if we just sort of take affordability as kind of a proxy for how are things, well, the economy is in some important ways worse than it looks. We don't have high unemployment, at least not yet.
We don't have high inflation by historical standards. But we do have this frozen job market where, you know, if you got your job, OK, but if you lose it or you're new to the labor market, it's very hard to get a first job or a new job. And again, I do think that, and I will be writing myself some more, I think that the way that some things that are key markers or have been key markers of middle class. status are receding out of reach matters in a way that the normal numbers don't quite capture. We read a really interesting Substack article recently, which has kind of gone viral by this guy, Michael Green, this asset manager, where he brought up, and I don't know if you've seen this, but he brought up this poverty line question where, you know, the current poverty line in America that we all use is around $30,000, a little over $30,000. And he made the point that this is an old metric, an old measurement that we used that worked in the 60s, but no longer works anymore. And he came up with kind of an adjusted calculation, and he landed on $140,000 as the poverty line. It's been very controversial. Some people said that that number is all wrong.
but the point kind of stands it's like well we're all talking about affordability we're all trying to figure out what's going on for the american uh economy particularly on the lower end and this has really resonated with people i just want to get your reactions to that article and this idea
“Most countries actually measure... poverty quite differently. They'd say that poverty is having less than half the median income.”
If someone said we're taking in $5 trillion in receipts, tax receipts, and spending $7 trillion, I would naturally assume that we were either at war or trying to jumpstart or reverse an economic decline, a malaise, get us out of a recession or a depression. And yet we're not. I mean, I don't think either of those are true. And the fact that the economy is not surging with $2 trillion in additional deficit spending, to me, says that masks an underlying sickness, that the economy is actually much weaker than people think.
Your thoughts? Before COVID, a lot of us were talking about secular stagnation, which is a totally impenetrable set of words. It gives you no clue of what we're actually talking about. What it really was that we actually had persistently inadequate demand, persistently inadequate spending in the economy, and that interest rates were always extremely low, and even that wasn't enough to really jumpstart the economy. It may well be that we would still be in that state if it weren't for these big budget deficits. And so that's a concern. It's also the case that advanced, stable, rule-of-law nations that borrow on their own currency have a lot of leeway to run up large debts. They can go a long way up that debt-to-GDP ratio without getting into a financial crisis.
Is that still a description of America? Are we, I mean, we're large, but are we stable? Are we a rule of law society? If I were an investor, wouldn't I worry about some current or future autocratic presidents deciding that to just arbitrarily write down our obligations? You know, it's so that these numbers are troubling. And there are, but we all know where it comes from. We basically have a political deadlock where we have immensely popular programs, I think deservedly so, like Medicare and Social Security. And we also have a complete unwillingness to collect enough taxes to pay for them. To me, this feels unsustainable when we're operating at what feels like it's not a good economy, not a bad economy, just what you call, I don't know, somewhere in the middle.
and we've decided that it's a standard operating procedure to go 40 percent you know to spend 140 percent of our tax receipts to me a i feel that's unsustainable do you agree and be when and it's hard to tell when but how does the music stop is it just when i've never bought the people won't show up for our treasury auctions they'll just demand more payment in exchange for the risk is there a limit do we go to i mean japan's much higher than us okay but at what point does the debt become an issue? Britain had debt that was 250% of GDP at the end of World War II. But it wasn't a crisis because people said Britain is a serious country run by people who are not idiots and they will, in the end, do what is necessary, as they did, to control their debt load.
Now, the trouble is that we may not be a serious country. And more to the point that the markets may conclude that we're not a serious country. And that's when the limit hits. It's not an arbitrary number. It's kind of a, who are we? Stein's law. Herb Stein, chairman of the Council of Economic Advisors, whose son was a much less funny comedian. But Herb Stein's law was if something cannot go on forever, it will stop. And Stein's law clearly must apply at some point. But so far we haven't hit that, though. I have to wonder. I mean, the character of American governance has changed so much in this past year that you have to wonder whether any assumptions based on the kind of country we used to be still apply.
But I put a lot of work on this. I actually was looking for at various times. I've looked for examples of countries that, like the United States, borrow in their own currency, have a high level of debt and are kind of politically paralyzed but are not hyperinflation territory. or whatever, and for crises. And it's really hard to find them in history. I mean, we can talk about the French Frank in 1926, but it is really hard to... We don't have a lot of historical examples that will help us understand how this ends. Sooner or later, it has to end one way or another, but how, I don't know. We'll be right back after the break.
And if you're enjoying the show, send it to a friend, and please follow us if you haven't already. It makes it easier to understand the markets and to make informed decisions. AlphaSpace by Yahoo Finance utilizes a Yahoo Scout powered assistant to build a personalized view around exactly what you want to analyze, then syncs it with your portfolio for real-time tracking. You can compare multiple tickers, explore everything from candlesticks and line charts to fundamental data and layer in indicators like moving averages, Bollinger Bands, RSI, MACD, and more. So I'm in that space in between what I'd call your standard retail investor and a hedge fund manager. I think a lot about stocks. I spend a lot of time managing my money. I advise hedge funds. This suits me perfectly. I want something more robust, better charting, that's AI enabled, that I can ask questions, that I can create baskets of stocks to track. And it has the same intuitive visual feel that I really like from Yahoo. You can explore alpha space with 50% off your first year of Yahoo Finance Gold at yahoofinance.com slash prop g. That's yahoofinance.com slash prop g. Build is $239.70 for the first year, then $479.40 thereafter. Offer valid for
new subscribers in the U.S. Cancel any time. Offer ends December 31st, 2026. Support for the show comes from Alpha Space by Yahoo Finance. Instead of bouncing between Between countless tools and tabs to manage your portfolio, AlphaSpace by Yahoo Finance brings your entire investment workflow into one easy-to-use platform. With advanced charting, market data, in-depth Yahoo Finance news, and AI. assistants all-in-one experience. It makes it easier to understand the markets and to make informed decisions. AlphaSpace by Yahoo Finance utilizes a Yahoo Scout powered assistant to build a personalized view around exactly what you want to analyze, then syncs it with your portfolio for real-time tracking. You can compare multiple tickers, explore everything from candlesticks and line charts to fundamental data, and layer in indicators like moving averages, Bollinger Bands, RSI, MICD, and more. So I'm in that space in between what I'd call your standard retail investor and a hedge fund manager. I think a lot about stocks. I spend a lot of time managing my money. I advise hedge funds. This suits me perfectly. I want something more robust, better charting, that's AI enabled, that I can ask questions, that I can create baskets of stocks to
track, and it has the same intuitive visual feel that I really like from Yahoo. You can explore Alphaspace with 50% off your first year of Yahoo Finance Gold at yahoofinance.com slash prop g. That's yahoofinance.com slash prop g. Build is $239.70 for the first year, then $479.40 thereafter. Offer valid for new subscribers in the U.S. Cancel anytime. Offer ends December 31st, 2026. Support for the show comes from Quince. The world is an uncertain place these days. If there's one thing you want to be able to depend on, it's your pants. Quince focuses on high-quality wardrobe stables made with premium materials, including 100% Mongolian cashmere, organic cotton, and Marino wool.
These styles are timeless, versatile, and designed to become the dependable pieces you reach for day after day. Quince is able to price all these pieces 50% to 80% lower than similar brands because they work directly with factories, so you're paying for high-quality, not brand markup. I have my first Quince piece. I bought a cashmere sweater. I really like it. Reminds me a little bit of, I won't say what brand, but a very high-end Italian brand. It's a staple, and the price point is great. Find your next fall favorites at Quince. Download the Quince app for app-exclusive offers, or go to quince.com.
Get free shipping on your order and 365-day returns. Now available in Canada and the UK, too. And when Quince asks where you heard about them, let them know it was Prop G Markets. It's the best way to support the show. we're back with prof g markets so one of the big themes of our show are talk tracks is that our economy has become more fragile because 40% of the S&P is represented by 10 companies. And that if NVIDIA announces, a big company announces they're not getting the return they'd hoped regarding their investment in AI, NVIDIA doesn't meet expectations. NVIDIA goes down 60 or 70%, which I will remind people, every one of the big tech companies has been down that much in a 12-month period at some point in the last 10 months or 10 years. And basically overnight, the S&P is off 10, 15, 20 percent, which represents a half of global market capitalization, and immediately we're in what feels like something like a global recession, if not something worse.
Our sense is that that feels more likely than not in the next 24 months. Your thoughts? I think you and I, not Ed, but you and I remember the 90s. I knew it was going to be that we're old. It definitely is. And there was, there's a lot of markers that feel like the dot-com, actually more telecom in terms of economics. But the excesses of the tech bubble are, you know, details differ quite a lot. And I have to say that this bubble, if it is a bubble, is an amazingly joyless bubble compared with the 90s. But sure, I think there's a, and we are in a situation now where, you know, but for all of that spending on data centers, we would probably be in a recession right now.
And so it's not at all hard to see this going south quite suddenly. And it's also, by the way, it does worry me that, and this was not true back then, that a lot of this is not only there's a few companies, but really the economy is being driven by the CapEx decisions of like 10 guys. If a handful of people have a mood swing, they can take down the whole world economy. To me, all roads lead to the same place. And the thesis is that the cloud cover, I think the S&P and the NASDAQ are the worst metrics invented in modern economy because they create the delusion of prosperity and don't really say how people are actually doing or feeling.
And without the cloud cover of the S&P being up, double digits. It'd be much more difficult for the president to be sending a mass police, secret police into big cities and that he has a huge vested interest in the continued kind of Lollapalooza champagne and cocaine of the AI bet driving the markets. And to me, this all leads to one place. When I look at the expectations built into the forward earnings of these companies, I don't think it's sustainable. And all roads for me, or not all roads, I think there's a very big likelihood that in 2026, we see some form of a bailout in the form of government-backed debt such that these guys can continue this capex, which seems to be propping up the market and the economy.
A bailout of AI in the form of some, quote-unquote, government investment or government-backed continued buying or financing a CapEx. Your thoughts? I think maybe tempted. We certainly shouldn't think of this administration as having any free market principles. And so it's quite possible that they might want to do it. I remember 2008, and it's worth remembering that in the face of the absolute meltdown of markets, the first time that TARP, which was a limited bailout of banks, the first time that TARP was put in front of Congress, it was voted down. I don't think it's that easy to engineer a bailout.
I think that, and among other things, people worried about the budget and the deficit and all of that. And this is really big. It's one thing to bail out Silicon Valley Bank or something like that. That's a fairly small thing. But this would be really, really huge, really, really hard to justify And one thing that is really clear also is that people hate AI. They hate the companies. They hate the people. This is the most unpopular boom. As I said, this is a joyless bubble. Compared with the 90s, everyone was kind of feeling some good vibes about the companies, even the ones that failed. So, I mean, it could happen, but I think that politically it would be extremely difficult.
It's really interesting. I feel like two big themes here, or one big theme in two stories, is that it feels as though we are artificially propping up the economy right now. And we've talked about it in relation to deficit spending, which is already kind of out of control. And now with the big, beautiful bill, it will get more out of control. And the predictions are that it's going to grow our economy, perhaps the stock market, because we're going to spend so much money. And this is coming from the administration that said, we're going to balance the budget. So that's a whole other can of worms.
And at the same time, you've got massive AI CapEx investment, which is basically keeping our GDP growth positive. And we're not yet seeing the underlying demand in the consumer economy that would warrant the amount of investment that we're seeing in AI, which I think leads me and Scott and probably a lot of our audience, to believe that whenever the music stops, because of the amount that we are artificially inflating everything in our system right now, it's going to be an extremely aggressive and shocking downturn. I just want to get your reactions to that notion. The IMF once did a systematic study of how successful are economists at predicting recessions. What is the track record of economists at predicting recessions? And the answer is zero our success. Might as well use a magic eight bowl as ask me or anybody else on this. There's just too many damn things happening in the world to be very good at this. Now, maybe it's worth saying that all the CapEx on data centers are related, and I've been, the measures are kind of, but we're talking about something like one to two percent of GDP. So although it's a, that's not chicken feed, but that's not enough to produce a 2008-level crash. So think more like 2001
than 2008. So it's not all that catastrophic if you try and do the math. Now, what the effects on sentiment might be, what the effects of falling stock prices might be is harder to say. But this does look crazy, but I think we probably don't want to go overboard in terms of the scale of what's at stake. It will certainly be unpleasant, and there will be demands that we bail this thing out and also furious pushback against those demands. But, you know, it's not the end of the world, which I can say because I remember the end of the world, right, which did sort of happen in 2008. So it's worrisome, but let's not get too over our skis on this.
“Silicon Valley used to think that it was a libertarian paradise and didn't need government”
We'll be right back. And for even more markets content, sign up for our newsletter at profgmarkets.com slash subscribe. Support for this show comes from Anthropic, the team behind Claude. Somewhere between the third meeting of your day and the 14th browser tab, there's not a lot of room for the work that matters most to you.
Claude makes room. It takes on the forms, the follow-ups, and the deck that needs reformatting. You get back to the parts that actually need you. Bring Claude into your work at claude.ai. Now for a quick message from Anthropic. the team behind Claude, the AI built for businesses. Every business wants to move faster. Few can afford to be wrong. That's why nine Fortune 10 companies count on Claude for the work they want to get right. Safety and accuracy aren't just boxes to check off for the research team at Anthropic. It's kind of their whole mission. See how Claude works at claude.ai. When it's time to scale your business, it's time for Shopify get everything you need to grow the way you want like all the way stack more sales with the best converting checkout on the planet track your cha-chings from every channel right in one spot and turn real-time reporting into big-time opportunities take your business to a whole new level switch to Shopify start your free trial today We're back with Prof G Markets.
I think some people would say, I mean, we're seeing what appears to be certainly crony capitalism. Others would just say flat-out corruption. And then on the other side, some people would say, oh, well, America has always been corrupt. If it's Silicon Valley now, it was Wall Street 10, 15, 20 years ago. You know, our policymakers are always in bed with whoever has the most money. When you look at this administration and the economic policies in 2025, Where does it rank in terms of corruption? How corrupt is this administration compared with history? It really is on a different scale. I think even the Gilded Age looks relatively clean by comparison.
But certainly compared with—look, we went through. I was very involved intellectually and certainly talking with all the players during the post-2008 financial crisis and the bank bailouts and all of that. And there was no question that big money on Wall Street carried a lot of political clout. But it was not, there was very little outright bribery, as far as we know, as far as I know. There wasn't, people weren't handing over bags of cash. They weren't, you know, buying meme coins or whatever, which didn't exist yet. It was more, there was a subtle, not that subtle, but a relatively subtle influence.
There was the revolving door. Government officials, you know, were always were thinking a little bit about their next job. There was just the prestige that comes with money. I mean, I did have a couple of meetings at the Obama White House where a couple of people like Joe Stiglitz and me were arguing for a harder line against the banks and the bankers were, of course, arguing against. And I came out at both meetings saying, you know, we have a fundamental disadvantage, which is that they have much better tailors. It was never the case that we were a deliberative democracy based on principle. Money always talked.
But billions of dollars in personal rewards for the first family, that's something new in American history.
“a persistent 150 basis point rise in the rate of productivity growth in the United States would make a lot of our economic problems just melt away.”
the New York Times for many, many years. You now kind of made the leap out into independent media, the new media sphere, which we kind of live in. You're on Substack. Your Substack has exploded. People are absolutely loving it. General takeaways on what it's like leaving legacy media and now being in new media, what do you make of it? I will always be grateful to The Times for offering me that platform. And the new media can do lots of things, but one thing it can't do is the kind of steady pound-the-pavement reporting that we still, I mean, I, if you look at what, you know, what do I reference in the Substack? And it's, it's a, it's a lot of stuff that's being reported by the New York Times and Bloomberg and, you know, it's, uh, uh, so we still need all of that. Um, but I will say that there is a, you know, there's a lot more, it's not just freedom to say what you want to say without, without editors and being able to, you know, not be polite sometimes, but also just format.
The kind of stuff I do, which often involves a mixture of charts and statistics and occasionally gets a little wonky for, you know, I can do that, which I really couldn't. I was just talking with some uh actually i just talked with martin wolf about this the the great liberation of not having to have my charts look pretty and so not having to wait three days for the graphics department to produce uh a chart that looks up to time standards and being able instead to just bang a crude thing out of excel and powerpoint that that that makes the point and and i can do within an hour or two do you have any thoughts on just where media is headed i mean the new york times is actually under fire just this week because they put out this article about david sax and all of his all the ways in which he's compromised and then suddenly you had this massive outpouring of um pushback largely from david sax and his his mates on the right saying the New York Times is compromised. It's woke. It's, you know, hasn't.
gender um what do you just make of like what is happening to media at large and what do you think where do you think it's all headed and then and we'll get you out of here you know assuming that we do manage to head off the authoritarian threat then i think the times should actually feel grateful i mean it's kind of a fdr i welcome their hatred i mean this is um the and the times almost uniquely i think uh in different ways a couple of other organizations have managed it but But they've managed to make media a financially sustainable proposition with a paywall that people are willing to actually subscribe to.
So they're actually a success story. I mean, in some ways, the problem is that only a handful of organizations can do that. And that is the great concern and how much we can rely on. I mean, it's great that, I mean, what I value from Substack is mostly actually the people who actually do reportage of some kind, people who actually know something and can tell me about it. And that's great, but it's got its limits. And we don't know. I mean, we used to have a media, you know, that did its job based upon advertising. Now we have a media based upon paywalls, which is works, but for many fewer organizations. I mean, you know, local reporting has become is a casualty. The fact that it used to be that classified ads were the great savior of local newspapers, and now they're gone, and so are the local newspapers.
So I don't know how all this shakes out. I mean, eventually we have to find a model. But look, everything, in the end, technology and the internet have just shaken up everything. Paul Krugman is the distinguished professor of economics at the Graduate Center of the City University of New York. He was a columnist for the New York Times from 2000 to 2024. In 2008, Krugman won the Nobel Memorial Prize in Economic Sciences for his contributions to new trade theory and new economic geography. Krugman was previously a professor of economics at MIT and later at Princeton University.
