Why Experts Keep Getting the Economy So Wrong

0:05Derek Thompson: In 1966, the economist Paul Samuelson said that the stock market predicted nine out of the last five recessions. It's kind of a weird joke. He's saying that every single time the stock market goes down, people freak out, predict a recession, and about half the time they're wrong. 60 years later, I think if we were going to update that sentiment for a modern audience, we'd have to say that economic commentators have predicted something like 100 out of the last two recessions. I mean, think about all of the times you have heard that this economy is doomed just in this decade alone, removing the first few months of COVID, of course, when the government basically put the whole economy in a cryo chamber to stop the spread.

0:47Derek Thompson: In 2021, inflation started to surge out of control. No recession. Then the Federal Reserve responded by raising interest rates faster than any time in modern history. Still no recession. Then the Russia-Ukraine war pushed up commodities prices, but no recession. Then Donald Trump becomes president, announces some totally audacious, unprecedented tariff regime on Liberation Day 2025. No recession. Then AI takes off. Still no recession. Then the war in Iran raises gas prices and the deficit is rising alongside data center debt and the bond market's a mess. And still, according to the most up-to-date estimates of economic growth, things are just kind of chugging along.

1:28Derek Thompson: Consider, for example, GDP growth. Last quarter, real GDP growth was 2%. The last quarter of Biden's presidency, it was 2%. It's average in the entire decade of the 2010s, just 2%. Or take unemployment. In the last month before COVID, during Trump's first term, the unemployment rate was just about 4%. It's average under Joe Biden, just about 4%. And today, after the tariffs and the wars and the AI and the gas prices, 4.1%. I think this is very strange. I mean, think about all the economic news of the last few years. Think about all the op-eds you've read, all the videos you watched, all the podcasts you listen to, maybe including guests on this show, that continually predicted imminent doom.

2:20Derek Thompson: The AI bubble is going to pop. No, it actually won't pop because AI is going to grow and take all the jobs. No, tariffs will crash the economy. No, the Iran war will crash the economy. No, the deficit will crash. It's nonstop. How many shows do you listen to that actually go back and count the recessions we predicted that never happened? How many shows go back and ask, why are we getting the economy wrong and wrong and wrong again? What are the experts missing that they always seem so certain that everything is just about to go to hell? and meanwhile national economic reality as it actually exists is so eerily stable.

3:03Derek Thompson: 2 % GDP growth, 4 % unemployment, like it's practically frozen in place, sealed in amber. I don't know. I don't know why the economy is so resilient to these predictions of doom, but I'm hoping today's guest does. Jason Furman is an economist who often does the heavy lifting of answering my most unwieldy economic questions. Today's unwieldy question is, why do economic experts keep getting the future wrong? Why is economic commentary so pessimistic? And what does it say about the U.S. economy that reality keeps defying the doomsday forecasts? I'm Derek Thompson, and this is Plain English.

4:15Derek Thompson: This episode is brought to you by JPMorgan Chase. In today's unpredictable world, geopolitics isn't just a backdrop. It's shaping markets, business strategy, and risk. The JPMorgan Chase Center for Geopolitics brings together former senior policymakers and seasoned practitioners to translate fast-moving global events into clear, decision-ready insights. From conflict and great power competition to energy and supply chain disruption, the center helps clients anticipate what comes next. Read the latest insights at jpmorganchase.com slash center for geopolitics. Did you know Uber has a range of safety features for riders?

4:58Derek Thompson: Like the Share My Trip feature that lets you send your live location to the people who matter most, your spouse, your kids, your best friend, so they can track your ride and make sure you get where you're going. But the safety doesn't stop there. Uber requires every driver to pass a thorough background check before they can start driving. This consists of a multi-step screening process that checks for impaired driving or criminal offenses, followed by annual background checks each and every year moving forward. Share My Trip and annual driver screenings are just a few of Uber's many safety features that put safety at every turn.

5:33Derek Thompson: Learn more at uber.com slash safety. Annual driving history reruns do not apply in New York City. Jason Furman, welcome back to the podcast. Great to be back with you. So I brought you back on to ask you to be a little bit of a traitor to your class here. That would be the class of economists and economic commentators who have spent the last few years finding reason after reason after reason why the economy is about to implode. Inflation, interest rates, Ukraine, tariffs, Iran, AI, deficits. And in the face of all these predictions, the economy just keeps on not imploding. So at the highest level, why do you think experts keep getting the economy so wrong?

6:22So, it's great to be back with you. I hope no one goes back and listens to every single thing I've said, even just on your show on these topics. But what I think the biggest answer to your question is there's a difference between economic models and economists that you hear. And so, if you read, you know, any of the investment banks have macro models that track what's going on in the economy and do lots of analysis about what happens if the Strait of Hormuz closes or Trump does his tariffs or whatever it is. The Congressional Budget Office does it. Academic macroeconomists do it a little bit, but not nearly as much.

7:02And generally, when you look at those models, you see things denominated in the tenths of a percent. When you see someone on TV, they sound much bigger than 10 tenths of a percent. Now, if it's me, they sometimes, if they're quite worked up, and I really was about tariffs, and to some degree still am, I'll sound incredibly worked up, but if asked about a number, I'll give one in the tenths. Other people might talk about, you know, catastrophe, devastating, everything is going to change. And then that also raises the question of why do the models give answers in the tenths? And the answer is, first of all, that the economy is really big.

7:42Oil matters less than it used to. Trade is 10 % of the economy. You know, whatever it is you're talking about touches less of the economy than you think. And there's a lot of mechanisms for healing and adjustment. So the price of one input goes up. You shift away from it. You use another input. There tend to not be a lot of things in the economy that are nearly quite as bottlenecky as we think they do. So I think economics, in summary, is a little bit better than economists on this set of questions.

8:15Derek Thompson: I take this as a defense of the models against the economic commentators who are sometimes representing those models on CNBC or Bloomberg or some podcast. I do think that that answer might even be a little bit too nice to the economic models themselves. I think that sometimes the economic models have failed to see surprises in the real world, pockets of resiliency in the real world that have made the U.S. economy even more resilient than some of those models suggested they might be. But we're going to get to some of those examples in a bit. There's a part of me that wants to preview the fact that I'll be arguing against the models a bit, but I don't want to show my hand too much.

8:58Derek Thompson: I think we should be very specific to start out. We may not disagree when we get there. So I was doing the first order difference between people talk about catastrophe. We haven't had catastrophe. The models never said catastrophe. Now we can debate how many tenths off they were. I want to go through a few specific recession predictions to understand why you think they were specifically wrong. And I think we should start with inflation and interest rates. So to catch people up under Joe Biden, annual inflation went crazy. annual inflation rates went all the way up to eight, nine percent, the highest level since the early 1980s.

9:34Derek Thompson: The Fed jacked up interest rates in 2022 to cool off demand in the economy. The pace of interest rate acceleration was the fastest in modern history. Many economists said that a recession was practically inevitable, that we were essentially back in the 1970s, that we'd seen this movie before and we were back to the stagnation film. A recession did not happen. Why not? Yeah, first of all, by the way, let me just say, one place the models did go wrong was the inflation. There they were wrong by several percentage points, which is to say there were some economists that predicted inflation. There were no large-scale economic models that did, whether the Fed, the Congressional Budget Office, the IMF, et cetera.

10:15All of them didn't have it. But then when it came to the sky falling, I mean, the amazing thing there was how bipartisan it was. I remember there was one month when Greg Mankiew, who was chair of the Council of Economic Advisors for President Bush, said, we've raised rates so much. Rates affect the economy with a lag. This is getting increasingly risky. Maybe they should stop. And that same month, Paul Krugman wrote basically the same thing. I believe this was December 2022, and the Fed ended up hiking for another six or so months after that. So, yeah. So the idea that the economy would go into recession was a bipartisan one.

10:54It was very credible, serious economists. In some sense, the last time we had raised rates this much was in the early 1980s, and it did lead to a recession. And so I don't think it was crazy that they were predicting that. At the time, I was a bit less worried, in part because I believe there was so much fiscal stimulus that hadn't yet worked its way through the system, that that would provide enough of a counterweight to the interest rate hikes. But, you know, all of this is sort of instinct and guess and a little bit hard to work out the exact numbers on.

11:29Derek Thompson: But if we want to understand specifically why those bipartisan predictions of imminent recession were wrong, what can we say now, from the vantage point of 2026, to understand why economists as diverse as Mancou and Krugman were wrong in December of 2022?

11:49Part of it is, again, if you looked at like Goldman Sachs' modeling at the time, they were showing something smaller than those two were. So the models were a little bit better, but the models were having a number of these mistakes. So I don't want to say they were perfect. It's possible we underestimated just the massive amount of stimulus that went into a system, and then a year in which people weren't spending the money. So at the time, people were calling it pent-up saving, and there was just a huge stock of that money for people to glide on. The economy has become less interest rate sensitive than it used to be when you had more manufacturing, auto-sensitive economy, and where companies needed to go to the bond market or the bank in order to fund themselves.

12:30Interest rates mattered a lot. when you have tech companies financing themselves with retained earnings and others or services that don't need a lot of capital. Interest rates just don't matter as much as they used to. And then finally, you know, just all sorts of randomness bounces in different directions and maybe some random thing bounced in the positive direction that people hadn't been factoring in.

12:52Derek Thompson: Yeah, I want to recircle something you just said and then add maybe another explanation. So one, I think it's very possible that the economy is less interest rate sensitive than it used to be, that maybe while the economy of the 1980s might have had a much larger manufacturing goods producing sector that was going to be sort of mechanically pinched once interest rates went up. Now the economy is more services. It's health care. It's education. It's government related. It's related, therefore, to government spending. It's things like restaurants. And it's less about manufacturing steel and automobiles.

13:26Derek Thompson: So the way that I kind of thought of it as I was writing up notes for the show is the economy used to be a little bit more like a three-legged stool. And if you cut out one leg with rising interest rates, then the stool topples over. But as the economy has become more diverse and more of a service-based portfolio, it's like a 100-leg stool. And so if you cut off one or two of those stools, you still got 98 legs that are up, and therefore the stool is going to remain upright. And that's basically what the economy was in 2023, 2024, as it was withstanding the blows of higher interest rates. The other thing I'd love to go back on, because I don't know where economists are on this story, is that while you and other folks have said, and I think I believe this, that fiscal stimulus led to, contributed to, I should say, inflation in 2021 and 2022, there was also a supply side story to tell that constrictions to global supply chains that reduce the amount of furniture or cars that were available to the U.S.

14:30Derek Thompson: economy, made it so that prices for these scarce things went up and that contributed to early inflation. And once those supply side supply chain aspects, the global economy were resolved, that there were parts of inflation that came down, even if those weren't interest rate sensitive. So how do you feel about the fact that maybe on the one hand, rising interest rates didn't sort of strangle the economy as much as it might have in 1991? And number two, there were parts of the inflation story that were already resolving themselves because of this supply chain aspect of it. Yeah. So first of all, I am fine with debating the ratio of demand and supply and inflation.

15:14It wasn't 100-0 and it wasn't 0-100. The only people who I have relatively little time and patience for are people who say the demand side was zero. I think it was probably about two-thirds. Maybe you could convince me it was one-third. You're not going to convince me that it was zero. So absolutely, supply played a role in inflation. And conversely, as supply started to heal, it played a role as well. Part of why I place more emphasis on the demand side, though, is that throughout that episode, the economy was growing quite well. You didn't really see supply problems in terms of the performance of the economy.

15:52And so, you know, how do you adjudicate supply and demand? To some degree, if real GDP is having a hard time, that says you're having a supply problem. If it's all in nominal GDP, which is roughly where it all was, that says it was demand. That's part of why I diagnosed this as a bigger part. But I should just also step back. There's a bigger thing you said about switching from a three-legged stool to a hundred-legged stool that I think is really true, important. And we started to understand that in the years before the financial crisis, and then we sort of forgot it because of the financial crisis.

16:28Back then, it went under the heading, the great moderation. It was a term, I'm almost positive it was Ben Bernanke, certainly made it famous. There's a lot of research that went under that heading, and there was a lot of evidence that just the economy was smoother. It fluctuated less month to month, quarter to quarter, year to year. Then people made fun of that research after the financial crisis hit and it became like a poster child of a thing that was true in retrospect, not prospect. In some ways, we're back to the great moderation. Take COVID out, which is not, COVID isn't an economic failure.

17:00It's a massive, massive thing. Take sort of a year and a half out of the data and the great moderation data looks really good in the period after the financial crisis. Right now, we're at the longest period ever of unemployment rate at or below four and a half percent. So a lot of the things that made the economy more moderate, and some of it is a shift to services, some of it is better inventory management, some of it is actually better credit availability that in normal times can smooth shocks and make it so you can still get money and spend. Although if you have a big problem with it. It can cause a big crisis like the financial crisis.

17:37But I think we're sort of in some ways still in the great moderation, which had one, you know, two big exceptions to it. One which raises issues about the thesis, the financial crisis, and the other is COVID, which, you know, has no bearing on the great moderation thesis. If anything, it reinforces it that we came out of it as quickly as we did.

17:57Derek Thompson: Can I do a quick detour on the great moderation before we get back to the actual subject of this interview, because there's an aspect of the great moderation thesis that I wanted to ask about separately. The great moderation is this idea that since the 1970s, monetary policy has been better at and economic diversity has been better at creating this world of steady, if somewhat slow growth with low inflation. and we're heading into a period where overall inflation is still somewhat elevated. The deficit is extremely high. Demand for debt from private sector actors like the hyperscalers and the AI companies is extremely high.

18:45Derek Thompson: Bond yields are soaring. And it seems like we might be headed into a period that is at the very least great moderation 2.0. So maybe we're not like back to the 1870s where it's like we're growing at 7%. Nope, it's deflation. Nope, we're back at growing at 8%. Nope, it's deflation. Like I don't think it's gonna get, it's gonna be that crazy in the 2030s, God willing. But don't you think that we are headed into some kind of different paradigm if the cost of money is going to be elevated for so long with a period of high deficits, high demand for debt by the private sector actors and overall consumer inflation for a variety of reasons.

19:27Derek Thompson: Doesn't that seem like something slightly different than the 1990s and 2010s? It's slightly different. It worries me, and we should do everything we can to make everything you said false by behaving differently with our policy, including things by managing our deficit. But I do think it is worth every now and then pausing and talking about the ways in which policy has improved. Our central bank is just much, much more credible than it was. The early 1980s, it needed a massive recession to convince people it meant business. Even though it lost complete control of inflation, the market, for the most part, never lost faith.

20:08And the fact that expectations stayed under control helped a lot. Things like inflation targeting, technocratic central banks, etc. Now, we can talk about nervousness, about whether or not that will persist. I think you appreciate an independent central bank. I appreciate an independent central bank. I'm not sure our current president does quite as much. So even these things I'm nervous about. Even if you globalize this discussion, we're talking mostly about the United States, go back to when COVID hit. The amount of discussion about this being a massive global financial crisis was really high.

20:44In February 2020, there was huge capital outflows from emerging markets. And so you could be forgiven for thinking this was going to be another replay of, say, the Asian financial crisis or something like that. Turned out emerging and poor countries weathered COVID relatively well. And when I look at their policy environments, they actually have more credible central banks than they used to. They're borrowing less short term. They're maintaining more reserves. They have more flexible, correctly priced exchange rates. It's just a lot of the basic macro advice that we've gotten a little bit better understanding over time and better at implementing over time has made a difference.

21:27Now, I should say the big caveat to everything I just said, which is both for emerging markets and the United States, the one measure where things are less safe and less good than they were before is the debt. That one does matter, but it's not the only thing that matters.

21:44Derek Thompson: Yeah. Yeah, this actually might have sounded to some people like a detour, and that is how I described it. But in a way, it goes directly to the question I posed at the beginning of this podcast, which is why has the economy been so surprisingly resilient? I think one very good answer is that the economy is more diversified than it used to be and therefore more resilient to shocks. And monetary policy is much, much better. And if you're not an economics nerd and you want a very clear way to think about that, it's kind of like the body of the U.S. economy is much healthier than it used to be. It works out more.

22:14Derek Thompson: It eats its fruits and vegetables. And medicine is much better than it used to be. That's monetary policy. We're just better at dosing for what's wrong with the economy to keep inflation more or less in the right zone and keep unemployment more or less in the right zone. So fitter bodies, better medicine. That metaphor, I think, applies to the last 30 to 50 years of the U.S. economy, which is sometimes described as the great moderation. Back to the main highway of this interview. We talked about the first way that economists got the future wrong when they predicted that inflation and interest rates would crash the economy.

22:50Derek Thompson: The second way that I think economic experts got things fantastically, sensationally and loudly wrong were the Trump tariffs. You go back to 2025, April 2025, the Liberation Day tariffs. Trump releases this plan to raise tariffs for all of America's trading partners. Democrats predict sometimes joyfully that a recession is imminent. Recession doesn't happen. Why doesn't it happen? This one, I think, really is 100 % the models got it right. And people either talked about something outside the models or talked in a tone that was inconsistent with the models. So I don't think the models get everything right.

23:30But on trade, they're pretty good. And part of why they're pretty good is they have the discipline of you raise the tariff on such and such by 10%. What fraction of our GDP is it? How much of an input is it to other production? When the price goes up by 10%, you know, how much does demand go down? And they sort of add all those things up and take them through the economy. And if you looked at, you know, Yale Budget Lab, you know, I think they're both very good at what they do, but there's certainly no one could accuse them of being predisposed to be enthusiastic about President Trump. They generally had numbers that were sort of like half a point off of growth.

24:07And by the way, I think it's possible that you said our growth rate was 2%, which is correct, that it would have been 1.5, I'm sorry, it would have been 2.5, that our growth rate would have been 2.5, but for the tariffs. And then, by the way, there's other things that have helped our growth rate and gone the other way. So in terms of growth, I think that's where the models got it right. Then look at the public discussion. I would include myself in this. I'm not sure whether or not I regret it, but I would write articles and talk in a way that sounded bigger than that. But then when I would include a number, that would be the number I would include.

24:45And by the way, I don't think we should sneer at half a point. You know, if you have a household that has $100 ,000 income, that's$500 for them. It's$500, by the way, maybe on the level. So$500 every year. It's basically like setting fire to it. It's a pretty stupid thing to do and not something you want your policymakers doing. And we need just some better way of talking about really unnecessary, unforced, harmful errors that aren't recessions.

25:19Derek Thompson: You know, there are several reasons to provide public comment if you're an economic commentator, if you're a talking head. One is to explain the present and predict the future. The other is to shape the future as a kind of advocate. And I wonder if you think economic talking heads who predicted that the Liberation Day tariffs or any increase in the overall tariff rate would destroy the economy, even if they were

25:53Derek Thompson: quantitatively wrong about what happened in the future, were their histrionics maybe useful for getting Trump to back off, useful for building an environment that scared Republicans and forced them to make the White House back off the original Liberation Day tariffs? Because certainly one reason why the economy did not respond to the tariff rates that were announced on Liberation Day, April 2025, is that those tariff rates weren't law for more than like 18 hours. Trump changed them very, very quickly. So is there a way in which you have like a little bit of sympathy for some of the economic commentariat that is often squealing and predicting that the sky is going to fall because they're trying to change policy so that the policy that they hate doesn't make this guy do something that is less dramatic than fall?

26:53Yeah. So I call that a self-unfulfilling prophecy. And it's sort of the ideal thing. I mean, you could even argue some of it happened with inflation. You know, those of us that were warning a lot about inflation and then inflation started to come down. Well, I think that's because the Fed did this historically rapid increase in rates. They listened in part to our warnings as well as their own understanding of the world. And that became a little bit of a self-unfulfilling prophecy. Maybe there's some of that on tariffs too. I worry though about, and I'll take the other side of that because I think what you're saying is sometimes true, but just to do the thing on the other side that I think is also sometimes true.

27:31If you're constantly saying the sky is falling, people tune you out and you end up discredited. I think the climate movement, maybe that's happened to sometimes, and maybe that's happened with the anti-tariff movement at other times too.

27:46Derek Thompson: We'll see about the AI safety movement, you know, come back in one year. We will see about the AI safety movement. I agree with that. And, you know, so in general, there's a part of me that thinks trying to deliver the numbers without sort of huge amounts of spin on the ball is the best way to remain credible. And I remember even in the beginning of the Trump administration, the GDP tracking numbers started to go negative for the first quarter. And there were a lot of people on X. Actually, these were not economists. They tend to be relatively political people like, oh, look, Trump's already caused a recession.

28:23And I was out there saying like, that's crazy. Like, you know, people don't cause recessions by turning on a dime. And most of why I was saying that is because I like to inject what I think of as truth into the world. I'm not saying it is truth, but I like to try. But also, I didn't think that we'd be that well served the people that were trying to make Trump's policies out to be bad if all of a sudden, you know, we had 1 % growth instead of minus 1 % and everyone declared that a victory. So I'd sometimes worry that when you move and set a goalpost in a certain way, you're ending up hurting yourself.

29:00So I try, and I'm going to try even harder going forward, to be more moderate about these things. But yeah, that means maybe a little bit less self-unfulfilling prophecies coming from me.

29:37Being rich is knowing what counts. Let's find your rich. Edward Jones, member SIPC.

29:44Derek Thompson: This episode is brought to you by Jira by Atlassian. Your work is scattered. A thread here, a DM there, five different docs related to the same project. And now it's not just your team hunting for context. It's your AI agents too. Jira taps Atlassian's teamwork graph to bring that context together across Jira, Confluence, Slack, GitHub, and more. and makes it easy to share with the agents you already use, like Claude, Cursor, and GitHub Copilot. Assign them work directly or connect your agents through MCP. The result, less token waste, more accuracy, and more time moving work forward. Learn more at jira.com.

30:21Derek Thompson: That's J-I-R-A dot com. This episode is brought to you by Lincoln. What if your vehicle could help give you a reset? Lincoln Rejuvenate is designed to do just that. This multi-sensory feature combines expansive visuals, digital sense, and a seat massage for the driver and is available in the 2027 Navigator SUV. Visit lincoln.com to learn more. Lincoln Rejuvenate does not operate unless the vehicle is in park when the vehicle is off or if any door is open. Do not use the feature in a closed garage or in other enclosed areas. See owner's manual for important operating instructions. So we've talked about why economists were wrong about inflation, interest rates, tariffs.

31:06Derek Thompson: Let's talk a little bit about wars in the Middle East. When the U.S. started the war with Iran and the Strait of Hormuz shutdown, there were a lot of people that I know, that I trust, who said something along the lines of the supply side shock of shutting down the Strait of Hormuz. It's akin to tearing the ACL of the global economy. This is going to be catastrophic, not just for East Asian countries, but for the United States as well. It's going to take us back to the 1970s oil crisis. And, you know, Trump has just authored a new economic crisis. We've been at war with Iran now for months. The Strait of Hormuz is not by any definition of the word open, fully open.

31:52Derek Thompson: Gas is now creeping up towards$6 nationally. But the economy is not falling apart. GDP growth last quarter was about 2%. The GDP now caster from the Fed Atlanta says something closer to 4%. What's going on here? How did another oil crisis, rather than create the conditions of the 1970s, seem to sort of like almost ricochet off the United States economy in a way that if you look at some statistics, some big statistics, GDP and employment, you can barely see the impact of the war in Iran. Let me talk about fertilizer first, and then I'm going to come to oil. Because I think fertilizer is a nice story.

32:38You could, in March, sound really smart by saying everyone's talking about oil, but the real issue is all the ingredients for fertilizer that are going through the Strait of Hormuz. I'm sure I should read that. Yuria. Interruption of, yeah, exactly. Yuria, an interruption of that will dramatically raise the price. That would sound really smart. And you could call that, what is it called? Whatever brain. Anyway, now we'll get to galaxy brain. Galaxy, yeah. Galaxy, no, no, no. One level up from that is to quantify it. If fertilizer prices double, turns out people use, I'm making these numbers up because I don't remember, 20 % less fertilizer.

33:17When you use 20 % less fertilizer, you produce 2 % less crops. And crops are 1 % of the global economy. And all these numbers are made up. But you multiply all of them and you got an extremely, extremely small number. And so that, again, is the difference between, you know, an economist when they're doing their job is going to just quantify the different stages of it. And by the way, they might be wrong. It may not be that the price goes up by 50 percent, maybe goes up by 80, maybe goes up by 40. But there's at least some basic accounting discipline, including the stages in the argument that involve just how much does it affect?

33:54how much is produced, and just how big is that in the economy. And so that's a place where the sort of smart people almost can have it worse than the, not smart, like IQ, but like people paying avid attention to every twist and turn can almost get it more worse because they're not doing the galaxy brain thing of quantifying. Now back to oil, to some degree, there's two pieces of it. One is given that the Strait of Hormuz has been closed for, you know, more than six months, and this was like almost the black swan geopolitical event that was on everyone's list as, you know, the worst thing that could happen.

34:37Are we surprised by how little oil has gone up in price? I am somewhat surprised. If you had told me this set of inputs, I would have thought more like$150 a barrel than, you know, the$100-ish a barrel we are right now. Then there's a second question, which is conditional on. If you had told me the price path of oil and asked me the economic effects, at$100, I would have been barely worried. That's roughly the 75th percentile inflation adjusted of where we've been in the last 50 years. It's just not that high. Even at$150 a barrel, and this one I did tweet out, inflation-adjusted, that's where oil prices went after the Libya strikes that Obama did in 2011.

35:25And they stayed at$150, sorry, in today's inflation-adjusted dollars, they stayed at$150 for about two years. So we went through that experience, and people were talking about it much less than they were talking about this one. I think maybe they actually understated that drag on that for the economy. Part of our slow recovery from the financial crisis was our fault for not doing enough policy. Part of it is financial crises were terrible, but part of it was actually a really terrible oil shock and a really terrible Eurozone crisis, which was sort of out of our control. So we probably talked too little about it 15 years ago, but conversely, I'm like looking at the numbers, thinking maybe we're talking too much about it now, even at$150 a barrel.

36:12Derek Thompson: There's a theme here that I think is really important to Recircle, which is that when the world changes, the world changes. People change their behavior in response to the world changing. So if fertilizer prices go up, farmers might do their best to grow a similar number of crops using less fertilizer. They might try to use the more expensive thing more efficiently. When the Strait of Hormuz closes and oil prices go up, China might reduce its oil imports. I believe China cut its oil buying by roughly 5 million barrels a day, which has played an enormous role in essentially capping global commodity prices, especially compared to some of the doomsday scenarios that were being bandied about when the Strait immediately closed, people talking about$200 barrels of oil.

37:01Derek Thompson: That didn't happen. And it's largely because of a Chinese policy change that was a reaction to the world changing when the Iran war started. And so I think that in addition to everything that you're saying, one thing that's hard for economic commentators, especially those that are motivated to tell a somewhat catastrophic story, one thing that's hard for them to anticipate is the way that actors will change their behavior for self-preservation reasons that will help the world achieve a kind of equilibrium, even if there's no official global coordination. Like we're not we didn't ask China to stop buying all stop importing so much oil.

37:42Derek Thompson: But the fact that it cut its oil imports so much absolutely reduces the price of American gas. And so that's like another hard thing. I wonder maybe how you would you would put this somewhat unwieldy comment. That's another really hard thing, I think, for commentators to do and economists to do is to anticipate how the world changes in response to change. Because in many cases, that is ultimately determinative of like the final price of oil, the final effect that it has in the world. Yeah. And there's two types of responses. One is a policy response, which we should think of as endogenous. People are making that policy change because of the thing that happened.

38:24And the other is just economic adjustments that always happen. And we tend to understate both of those. So, for example, after Russian gas for Germany was cut off, there was a big debate in Germany. And there the economists were actually saying this isn't going to cause a massive recession. And it was a lot of the non-economists, including the business community, saying this will cause a massive recession. It turned out not to. Some of that was policy adjustment. They kept, you know, more coal-fired power. They found other sources of energy. and some of it was some adjustment within the economy.

39:03The more energy-intensive things did suffer a bit, but activity shifted over to less energy-intensive. I should say as a side note, the fact that they retired their nuclear reactors at the same time suggests that the rational endogenous policy response, as judged by me, is not always what happens. But by and large, it did. And I think that's why Germany didn't have a huge response here. With China, in part, it is a policy choice about how much oil to buy, and they make more policy choices at the level of their economy as a whole than we do here in the United States. But it's also technological developments.

39:45And I think that's another theme that shows up in oil, but shows up in the moderation more generally, which is we all know just how much less oil intensive the economy is than it was in 1979 when we had the Iran oil shock. It's also much less oil-intensive than it was even in 2011 when we had the inflation-adjusted$150 a barrel when Obama was president. And that's because we've made a lot of progress in fuel efficiency, even for non-electric cars, electric cars. China is recently the fastest and most, you know, and then the most extensively utilized progress. And that's put us in a better position to weather what used to be one of the main sources of recessions, which is oil price fluctuations, into one where both the oil price itself moves less because demand can adjust.

40:41And for any given oil price movement, less happens to the macroeconomy.

40:48Derek Thompson: So we're thinking about the themes we've developed in walking through inflation, interest rates, tariffs, war in Iran, also Ukraine and commodity price increases. I have a few that I really want to remember. One is this idea of the U.S. economy as the 100 leg stool. The U.S. economy is really big, extremely diverse and therefore extremely resilient to changes that affect certain industries. because if certain industries get hit, well, consumers can shift their spending toward other industries and GDP can still grow. Workers that might lose their job in those industries can shift their, can lose their job and be hired by another industry.

41:30Derek Thompson: And the labor force can still grow. And this enormous diversity of the U.S. economy is really, really important about thinking or in trying to understand why the body is so resilient to crises. Number two, I do think it's really important that monetary policy is just much better than it used to be. It's a boring one. It's a vegetal answer, but we just did a podcast on the economic crises of the 1870s. Monetary policy was absolutely horrendous, not only in the US, but also in Europe in the 1870s. We caused a lot of global panics. We're a lot smarter than we were 150 years ago. And then finally, I love this last theme of adjustments from both consumers and from policymakers around the world.

42:11Derek Thompson: The world changes and then the world responds to those changes. I want to spend a little bit of time talking about media criticism, because in addition to the fact that the U.S. economy might be stronger than we often rate it, I do think that sometimes economic commentators are more hysterical than they sometimes should be. Um, you know, I, I think that in many cases, uh, it seems quite clear that people understand if you want to, you know, get on television, or maybe if you want to have a podcast that has more downloads or something, it makes sense to just be constantly predicting bubbles and crashes.

42:49Derek Thompson: I know how I feel about this as a journalist. I wonder how you feel about this as an economist. Do you sometimes have sympathy for these people? because you're like, well, they're in the media and this is just, they're doing their job? Or is there some other feeling that you have? First of all, you write about trends as opposed to just immediate events. And some trends are bad, things I've learned about from you because I don't do work on it myself related to loneliness and isolation. That wasn't like somebody woke up, pressed a button, and that happened. It's sort of a long-running trend. A lot of trends are really positive, like more electric cars and lower emissions, and you're right about that.

43:29An awful lot of people are in the events business. And while I can occasionally think of a good event, most events are bad. There, you did some tariffs, you tried to fire the Fed chair, you launched a war, whatever it is. Rarely is there some dramatic thing where you press a button that day and everyone's debating, will the economy double in size because of that brilliant button that was just pressed? Whereas sometimes there's a button that we're pressed where we're debating whether it would fall in half. Now, I do think we overstate the magnitudes of those events, as I've said before, but I do think most events actually probably are bad and trends are more mixed.

44:09And, you know, given the steady pace of U.S. economic growth over, by the way, the last, you know, more than 100 years, probably those trends are mostly good and mostly our friend. And how do we have a way to talk more about trends, less about events, while putting those events in the proper perspective and making sure we're still doing our self-unfulfilling prophecies? I don't know exactly how to get that whole mixture right, but I actually do think you do better than most.

44:39Derek Thompson: Well, that's actually, I've never quite thought of it like that. I'm trying to think if I, I'm trying to ask myself if I agree with that idea, that events are disproportionately bad and trends are disproportionately good. Is that your thesis? Let me give you a macro, yes. Yeah, go ahead. In the macro economy, at least, just macro. And in some sense, that comes out of understanding the macro economy, which is the economy has a certain amount of potential. At any given point in time, you have a certain number of factories, you have a certain number of workers, you have a certain level of technology.

45:11And if you do everything right, you produce at your potential. But if you mess up, your unemployment rate can go to 10%. If you really mess up, like in the 1870s, you know, it can go to 25%. And so you can subtract a large chunk of your GDP in a press of a button if you really mess up. If you want to add to your GDP, you need to build more factories. You need to develop new ideas. You need to train your workers better. It's all supply side. It's all more gradual. And so the idea that the economy operates sort of at trend or below trend, depending on whether you messed up or not, says events can take you down but not really that far up.

45:51Whereas over time, we have unlimited potential in terms of capital, ideas, people, but that takes a while. So I think there actually is a good macro reason for thinking what I think. Whether it applies to other domains, I'm not totally sure off the top of my head.

46:07Derek Thompson: I'm stuck on this because I think this idea is more profound than maybe even you intended it to be I wonder if a really important lesson for journalists or for anybody commenting on the news is to remind them to write or talk more about trends than events because if let's just again keep this following comment bounded by macroeconomics. If recessions are rare, then almost the entire macroeconomic story of America is one of growth, right? And so the trend line will be about growth and the events, both those that we see and those that we fear, are likely to be interruptions to the trend line, which is growth.

46:58Derek Thompson: And so even there, I think it is fair to say the trend is positive and the events are negative. And I kind of love this idea, even as advice for me, that when I feel like my podcast or my writing is getting too negative, it's like I should talk about a trend. I should focus on trends more to test the proposition that the most important stories in the world right now are bad because it might be the case if I focused more on trends than events, I would see some better stories. It's very interesting. I mean, as you said at the beginning of the statement, I'm very interested in a long-term trend that I think is bad, the fact that people spend more and more alone time.

47:46Derek Thompson: But I do think that especially in the realm of science, it is the case that the trend lines tend to be positive. In the realm of macroeconomics, I don't think you could point to inequality or maybe something else that's a negative story, but it seems like most of the stories about growth are stories that are positive. Yeah, go ahead. Yeah, yeah. And just to give an example, an event that was really bad for the economy and even worse for other things than the economy was COVID. But if you took an economist in the year 2019, put them to sleep for two and a half years, woke them up and showed them the macro data, don't show them inflation, but just show them like where the economy is, number of jobs, et cetera, they'd have no idea.

48:31that this massive thing had just happened. So, and that, by the way, is an event. There's no event that could have come along that could have taken 10 percentage points off the unemployment rate. We couldn't have the unemployment rate fall to minus five because some super great thing happened. We had a terrible thing and the unemployment rate rose to 15, but it fell back down again pretty quickly. And we were back on those trends. So that would be, you know, sort of one of many examples of this. And by the way, you could take this back even further. An economist in the year 1925, trying to predict where the economy was going to be in 2025, just doing a linear extrapolation of per capita GDP growth would get pretty close to the right answer.

49:12So there's something almost miraculously amazing about these trends in the U.S. economy.

49:18Derek Thompson: You're referring to a famous, among maybe 14 people, graph of GDP per capita growing at about 2 % a year every year between the early 20th century and the early 21st century. Devin, I don't know if maybe we can find it to flash on the screen, but it is amazing that if you were like an extremely basic and stupid macroeconomist in the middle of the 1920s, and you were like, I kind of just want to create a diagonal line that goes up with red marker between now and the early 2020s, and that's the future economic growth. Like, yes, you miss all of the events. You miss the Great Recession. You miss the World War.

49:54Derek Thompson: You miss the Cold War. You miss the invention of, you know, the electronics industry. You miss the internet. You miss all the events, But you do get this very specific number of GDP per capita in the 2020s, more or less accurate. Do you think artificial intelligence is an event or a new trend? Is this something that you think is with us now that's going to create a kind of like period effect? Or do you think we're ultimately going to have to think about this as a general purpose technology that sort of spreads its influence on the U.S. economy over the next few decades? Yeah, this one feels like both.

50:30And certainly, you know, insofar as part of it is in the trend category, it is a trend we're talking about very, very early compared to many other trends. And appropriately so, because it's such a massive development. Just from a macro perspective, so far you're seeing it clearly in demand. Now it's also crowding out other stuff. So we build data centers and we cut back construction of other things. In supply, you can debate whether it's adding a tenth or two or three to productivity growth. Maybe you could debate whether it's taking a tenth or two off the unemployment rate, but those are hard to tell apart from zero.

51:09So the trend-type stuff, an economist would look for trends in the productivity growth rate. The trend-type stuff from it is not in the data yet, but I'm reasonably optimistic that it will be. I'm reasonably optimistic that it'll make a difference. Well, a difference probably for the positive. There are days I pause, and that's why I paused right there, even as I answered that. But, you know, again, TBD and, you know, if I were predicting productivity growth over the next decade, I'd probably put down a number like 2.3 or 2.4%, which is maybe seven-tenths of a percent higher than it's been since World War II, which to me is a massive deal.

51:55But in Silicon Valley, you know, if you said four and a half people would think you were incredibly dark and pessimistic about this technology. If you said 15, they'd smile and nod. And if you said 100, they wouldn't really dismiss you as completely crazy.

52:09Derek Thompson: So this is one case where I like to say, I think the pessimistic prediction is secretly the optimistic prediction. Because when I hear these folks say, I think GDP growth on an annualized basis is going to go to 5 % or 10%. I say, do you realize how much change that implies over the course of decades to the U.S. economy? 5 % GDP, 10 % GDP growth every single year. That level of change is going to have social and political consequences that we can't even imagine. I mean, the amount of physical world change that that would demand. It's hard to build a data center in Wisconsin. How many data centers do you have to build for 10 % annualized growth?

52:53Derek Thompson: It's hard to build houses in parts of this country. How many physical structures do you need for 10 % annualized growth? The implications even for something like Balmels cost disease, if you have productivity growth that is so massive in the digital economy that the overall economy is growing at 5 % to 10%, percent. Like, doesn't that imply that movie tickets and concert tickets and restaurant meals are seeing just absolutely berserk inflation because we have to increase the wages of less productive workers to keep up with the productivity of the rest of the economy? Like, you're talking about a level of societal change that I think would be, in some cases, quite wrenching and extremely unpopular.

53:35Derek Thompson: So I tell them, like, if you want to be optimistic, like actually optimistic about this technology, you want it to add half a percentage point of GDP growth on an annualized basis. That is the Goldilocks spot of America's getting richer, but the world isn't transforming so quickly that we get some kind of, you know, what is it called? You know, butlerian jihad that rises up against AI. Yeah, I completely agree with you. I've actually been asking my economist friends for about a year now, if you could pick the rate of productivity growth, it will be whatever you say it is. You say five, you say 10, you say 100, that's what you get.

54:13But you don't get to pick anything else. Everything else is what comes with that productivity growth, whether that's whatever the employment effects are, whatever the social dislocation, whatever the political dislocation, what number would you pick? Some of them say that's nuts. You know, 100 is better than 10, 1 ,000 is better than 100. And if I could have a million percent productivity growth, I would. Why do you want to keep people in poverty forever, Jason? Others have thought about it already. But I'd say the most common is like, huh. And then you get a number like a little bit higher than yours.

54:47Mine, I'd sort of pick a number, 3 % productivity growth, which is almost a little bit more than a point faster than we'd had in the past. I think we could handle that perfectly fine if I could do that. But if I could pick five and press a button for five, I don't think I'd press that button because I'd be afraid of everything else that came along with that five. And part of why this is such a weird question that you haven't normally asked is most policies that we consider, our productivity growth is 1.7. If you do this, it'll go all the way to 1.8 or 1.9. And if you're in that localized region, more is good, less is bad.

55:24And that will almost always give you the right answer. But when you're talking about really large jumps, the shape of the objective function, whether you're going up in goodness as productivity goes up, to me becomes much more complicated and confusing. And I worry about the downsides and mitigating them.

55:41Derek Thompson: Well, here again, to steer us back to the main road and maybe close things out, this is why I think it's useful to think of the economy as a hundred-legged stool. If you think about AI being this big, massive, powerful thing that's going to change the economy in the abstract, you're like, oh, okay, fine. It's going to happen. We're going to get 10 % productivity growth. But if you're really forced to count all the stools in the leg and remember that the U.S. economy is registered nurses and it's home health aides, and it's people bringing your French fries to your table when you go out for a meal, and it's people dropping off packages from Amazon.

56:19Derek Thompson: And it's people in this services, physical world economy, that is not going to be immediately automated and made more productive just because there's a data center that holds a set of inference that can solve a bunch of millennium prize mathematical problems. You realize there are a lot of bottlenecks between the models are really, really smart. And now the economy is suddenly growing at 10 % a year and we're in the singularity. Like there's, there's, there's a lot of steps between any change, tariff, uh, or GPT-7 and, uh, changes in the actual economy. Um, so Jason, I, I really appreciate the conversation.

57:01Derek Thompson: Um, I appreciate you being willing to criticize and only economic commentators like me, but also, um, some folks in your own field. Um, and I'm sure we'll have you back very soon when we, you know, get the next recession. Great. Love being with you. Love this conversation. I have one thing to say about what you just said that I'll say, and then you can decide whether you cut it or use it. No, yeah. Say it. Great. Yeah. I think this is great full circle for our conversation because we talked about how events are usually bad. Trends are usually good. Then we came to this trend where for the first time we're worried, actually, maybe is this trend bad?

57:37And part of why maybe we shouldn't be as worried about this trend being bad is all the same reasons why we've consistently overstated it for events. So you listed a lot of them, the 100-leg stool. The other piece of that is just adjustment, which is when one price goes up and another wage goes down. People buy less of the thing with the higher price. They move out of the job with the lower price. Maybe there'll be a policy response. It's a little bit hard to see that today. But just the economy itself is flexible at figuring out how to reallocate between those 100 different stools. And prices and adjustment are something that we consistently underestimate the importance of, whether it's for events or, in the case of AI, I think, for trends.

58:24Let's keep it.

58:26Derek Thompson: Jason Freeman, thank you very much.

58:39This episode is brought to you by Whole Foods Market. Fall is here and Whole Foods Market has everything you need for cozy meals, seasonal treats, and easy entertaining. Elevate your hearty dishes with tasty bone-in beef short ribs or New York strip steaks with no added growth hormones or antibiotics.

58:54Derek Thompson: And as the fall evenings draw in, why not get ahead of cold season with immunity shots, herbal teas, lozenges, and supplements? Make sure you look for sales throughout the month. Shop fall flavors with higher standards at Whole Foods Market.

59:33price plan options available. Taxes and fees extra. Default terms at MintMobile.com.