Intro. [Recording date: May 15, 2025.]
Russ Roberts: Today is May 15th, 2025. There is no guest today. This is a monologue. As far as I know, this is the 1000th episode of EconTalk. And, to mark that occasion, I thought I’d share some thoughts.
It’s a little self-indulgent. I’m going to talk about some of the logistics and nuts and bolts of the podcast, which people have asked me about over the years, and I’m going to reflect on what this has been like to do this for almost 20 years, 1,000 times.
Russ Roberts: So, to do that–I started to do this. I made an outline of what I was going to talk about and then realized I wasn’t excited about doing it, and I went home and told my wife that it had just kind of fell flat, just couldn’t get any enthusiasm for the idea.
And she said, ‘We should have had ChatGPT [Generative Pre-trained Transformer] ask you the questions.’ And I thought, ‘Well, that’s a good idea.’ So, I did that.
And so, this–I will occasionally repeat some of its questions or just answer them because that’ll make sense. But, that was a fascinating experience in and of itself, and I may reflect on that along the way as we get started.
So, the first question–I love the way that it asked this question. This is the -03, by the way, of ChatGPT, which is really an extraordinary achievement. I’ve done, I think 20 episodes on AI [artificial intelligence], interviewing people about it, its threats, its potential, how it works; and I do think -03 is a step–a quantum leap–in improvement. And, we’ll talk about that maybe at another time.
Russ Roberts: But here’s the wording of the first question:
ChatGPT: What was the spark that convinced you to hit Record? And, what were your earliest ambitions for the show? And, it had said before: When you launched EconTalk back in 2006, podcasting was still a fringe medium; and long-term, long-form economics conversations sounded frankly like a gamble. So, what was the spark?
Russ Roberts: Well, I was late to blogging. I started Cafe Hayek with Don Boudreaux, and it has had a long run under Don’s leadership. In the early days, I blogged alongside him; and I felt we had come late to blogging in whenever we started in 2004, whenever it was.
And so, I didn’t want to be late to podcasting once I discovered it.
And, the way I discovered it was, as far as I know, James Reese, an economist, interviewed me on his podcast, Radio Economics. I don’t think it exists anymore. Jim, if you’re out there, say hello. Or if anybody knows about this, can remember this better than I can.
But, he called me up and interviewed me about–which is how we did interviews in those days, on the phone–he called me up and asked me to talk about Walmart. And at the end, or maybe it was beforehand, I said, ‘How many people listen to this?’ I mean, I’d never heard of a podcast. He said, ‘Well, it’s 2-to-3,000.’
And, when I heard that, I thought, ‘Wow. If I could show up on Monday mornings and there were 3,000 people sitting in an auditorium with two chairs at the front of the room and I could go sit in one of them and a guest could be in the other one, I think I’d come to that room. And it would be fun to have 3,000 listeners.’
Now, at the time, I thought, ‘Well–‘ And, the other reason I was interested was that I wanted to do a–I had considered for a while doing a radio show, a talk radio show, an interview program. And, I had done a lot of commentary at that point, I think for NPR [National Public Radio].
And, I thought maybe someday I could get an NPR talk show–maybe–and maybe I was interested. I didn’t know if I could really want to do it. It seemed like a lot of work.
But, this seemed like a way to try it, and do, like ,a free trial to do podcasting and see if anybody listened. And, how hard was it?
So, in the early days, I think I started off just doing them occasionally. You can look at the archive. But shortly after that, I realized after starting, I realized that people expected to hear one on some kind of regular basis. Because otherwise they’d show up and it would be the same episode as it was up there before, and we wouldn’t be making any progress.
So I was pretty quickly, pretty early on, I decided I would do one every week. And I kept to that schedule. I would take off the last week or two of the year for the holiday season, give myself a break, and then I realized: Well, people like it every week. So I think I’ll just record one in advance. I can take off the holidays anyway.
And so, it turns out, according to artificial intelligence–I forget which one I looked at, I think it was ChatGPT–I have put out an episode of EconTalk for 854 consecutive Mondays. This will be 855 if all goes well. Or maybe it’s a little more than that actually because it didn’t know about the couple that are coming up.
But, December 2008 was the last time I missed a Monday. So, the first Monday of 2009, for the last 16 years, I have done one of these.
It’s hard. I have a job now that’s pretty demanding. I had a job before, but EconTalk was a non-trivial part of it. So, the chance to do this every week has been a lot–the privilege of doing it every week has gotten a lot harder–now that I’m president of Shalom College in Jerusalem. But I’m kind of addicted to it and I like to think some of you are as well.
So, we haven’t missed a Monday in a long time. I worried during the war that we might miss–the war here since October 7th [Oct. 7, 2023]–I might miss one. We’ve managed to not do that. We’ve managed to keep them going. I will have a health issue inevitably, probably, someday while this program is still in my hands, and that’s exciting. It’s still in my hands, but someday it will have a challenge; and we’ll maybe use an old episode to keep the streak going. But, for now, we’ve kept it going for 850-plus Mondays. Which is kind of cool.
Russ Roberts: The next question was:
ChatGPT: How did I think of this 3,000-person interview audience? Was it going to be a lecture, or a radio call and a fireside chat, or something else?
Russ Roberts: And, it’s funny, when I started the program, for some reason I kind of thought it should be about an hour. I don’t know why I thought that, but I soon found out that sometimes it’s hard to keep it to an hour. But, I tried to, in the early days. I remember my dad saying, ‘Oh, an hour? No one’s going to listen for an hour.’ I mean, it’s 2006. He said, ‘Radio pieces are, like NPR [National Public Radio] is eight minutes. That’s long. No one will listen for an hour.’ I didn’t care.
The other thing that was fun in the early days is that people said, ‘Well, you’ll run out of gas because how many economists can you interview?’ Of course, I found out quickly that there were people who weren’t economists who were fun to talk to, and I had this brilliant idea that you could interview somebody more than once. Sometimes they had more than one hour’s worth of interesting information.
So, obviously the program has changed a lot since the early days when it was pretty much all economists all the time. But, I’m now am happy to go over for an hour. And of course, since I started the podcast, there are podcasts that go for three and four hours. I’ve been a guest on some of those. They say, ‘How long are we going to go for?’ ‘Oh, three or four hours.’ I think I can’t stand to listen to myself for barely more than an hour. I can’t do that.
So, I’ve done a few longish ones. But on EconTalk, I think the longest episode we ever did was Davies on Extreme Economies, I think it’s called. We’ll put a link up to it. It’s Richard Davies, and that was two hours and four minutes.
The Haviv Rettig Gur episode is about that long. It’s two hours and four minutes–maybe a few seconds shorter than the Davies one–but they’re both in the two-hour-and-four-minute length. That’s a little long. But the goal now is I try to go a little over an hour and if it’s still interesting, I go an hour and 15, or an hour and 20.
In the early days, the audience I had in mind were people interested in economics, and I was always thinking, ‘How do I reach people who are interested in economics? Should I advertise in The Economist or the Wall Street Journal?’ Of course, we never advertised anywhere in that normal way. But that’s who I saw as my core audience.
And, in the back of my mind, I thought I was going to encourage people to understand free-market principles better. So, early episodes–especially with people that I didn’t agree with–were more or less a debate.
And, if you go back to the early episodes, there’s two things that bother me about them. One is that debate mindset. And the other is I talk too much.
So, if you go back to the beginning, as many people have–which blows me away–there are many people who tell me they start at the beginning or they’ve gone back to the beginning and they’ve listened to every single episode. Some of them more than once. But if you do that, don’t be discouraged: They do get better.
But, I got interested–well, two things. One is I realized pretty quickly that my audience wasn’t people interested in economics. It was people interested in learning something, which is why eventually we added the tagline, Conversations for the Curious. It was for curious people. People were looking for educational content that was somewhat a little bit entertaining. And that was my audience. That is my audience now.
And, it’s a crazy thing. You’re doing the dishes right now or you’re walking your dog or you’re working out at the gym and you want to grow or learn or have an interesting something to listen to. And, those are the people I think are my main audience, which is–hello, out there. It’s nice to have you. So, that’s the story there.
Russ Roberts: The next question actually at this point–it’s kind of interesting. Just an AI aside: as I mentioned, I think the beginning of EconTalk was when I was interviewed by Jim Reese, James Reese on the show Radio Economics. In the new version of -03, if you watch while it’s thinking, you can see its thoughts, whatever that means: you can see what it’s looking for and what it’s trying to discover. It kept coming back to that question of what was the first podcast I was on that got me interested in EconTalk? It didn’t let it put it down. It couldn’t put it down. Two or three or four different times that was what it started looking at when I asked it to continue and ask me another question. It was very obsessed with that.
And also, it refused to do what I asked it to do, which was to have a conversation back and forth. So, at this point, after I answered the question I just did, it then gave me a bunch of questions:
ChatGPT: How much with the mix of entertainment versus education, engineering or re-engineering the prep for when I wasn’t doing as much debate? How do I get chemistry to work across the screen now that we’re doing it on Zoom? Broadening the virtual hall: How did I get more downloads and people?
Russ Roberts: And, we now probably have something over, by the way, it was something like 100,000, maybe 150,000 listeners.
It’s hard to define. Any one month has different numbers, and my actual total numbers of downloads are down for, I think two reasons. One, there are people who got angry that I did a bunch of episodes on Israel; and maybe we’ll talk about that.
And also, Apple stopped automatically downloading episodes to people who, once they subscribe, they don’t get every episode that comes out anymore. Now you have–if you don’t listen to any of the episodes that have been downloaded, if you don’t listen for a while, it stops automatically downloading. So, that hurt my numbers. Of course, that shows the numbers were somewhat distorted. Anyway, our listenership is something over 100,000, maybe 150,000.
Anyway, so to continue, the question is why did I stop doing the debate? And, a couple of reasons. I realized that if I yelled at people and debated them and challenged them constantly–people who disagreed with me–really[?] wouldn’t be as interested in coming on the program. And that would mean I’d end up preaching to the choir. And I wasn’t interested in that.
And I also came to believe–maybe just convincing myself because I’m biased–that debate isn’t as good as a shared exploration. That’s more like a conversation for discovering truth. If you rant and counter-rant, you’re going to get eyeballs and downloads and hits, but I didn’t think that would be good for my soul. I thought that would be bad for me. And, I also realized a gentler approach was just more pleasant.
I trusted my listeners. I didn’t think you had to hear every counter-argument. People would say, ‘How could you let so-and-so get away with saying that? Why didn’t you answer them? Why did–‘ I thought, ‘Well, that’s your job. Your job out there is to when you hear something you think is a bad argument, to take note.’
And I also would say to people who complained about that: You know, if I answered everything I disagreed with and tried to refute it, we wouldn’t get too many observations in the of course of the talk–of the interview. It would just kind of stagnate.
So, the bottom line is I tried to build what I would call respectful skepticism with people I didn’t agree with. But, I saw my mission when I’m interviewing an author about a book they’ve written, to let them make the case for the book. To try to bring out the narrative arc of the book, or at least the parts that I thought were interesting or that I thought you, the listeners, would find interesting.
Now, I still challenge lots of things. I still raise questions about something an author would claim. But in general, I do not keep going back and forth and argue with them. So, it’s not an argument. I challenge respectfully. That’s sort of my way I think about it.
There are exceptions, even in the later years. Some people say things that are so shocking to me, I feel like I can’t stay quiet. But, in general, I try to be let them have their say. I make it clear when appropriate that I do not agree and why. And then, you, the listener, can make the decision of which one makes sense to you.
My prep hasn’t changed much in a long time. I script about 15 questions–maybe 12, maybe 18. I sometimes will have quotes from the book that I want to use as question generators or that I might read out loud to let you, the listener, try to understand what the book is about. But, I script about 15 questions that I know will get us enough material for the hour, or hour-plus.
Of course, I don’t just go down the list. We digress. We go off into side-tangents. And, a lot of what is challenging about being the host–which, it wasn’t clear to me until I actually had to do it–is you’re doing a bunch of things at once. It’s very hard to listen–because you’re looking at the clock, you’re trying to figure out how much time you have left, how much questions you have left, should you ask a follow-up question? Should you switch gears? Should you challenge?
And so, it’s funny: many of you probably have a better idea of what I said or what the guest said than I do, because my attention is always being pulled in those different directions.
And, every once in a while–without naming names–I’ll get a guest who is not very chatty or not very conversational or very taciturn in general or shy. And I realize early on that I’m in trouble. Fifteen or 20 questions are not going to be enough to fill the time. And you’ll hear me even in later years talk more than I might normally do in those situations. So, if you notice that, that’s what’s going on there.
Russ Roberts: The next set of questions–again, ChatGPT wanted to ask me a bunch at once:
ChatGPT: How do I know when to bite my tongue? What about those questions? What about tangents? What about leaving economics and doing other things?
Russ Roberts: A whole bunch of others; there’s a bunch more. But it came up with about 14 questions at that point, so I’m going to try to answer some of them.
So: When it comes to biting my tongue–that is, not responding to something–I don’t remember particular moments. Sometimes I hesitate internally: I’m debating whether I should answer or not.
What else? Oh: Do I share the questions in advance? Never; almost never. I have guests who ask me to, but I always worry that if I do that it’ll sound scripted and less spontaneous.
So, I will occasionally give them–when they’re uneasy or nervous, I’ll give them an overview of the topics we’re going to talk about; but I don’t give them the exact questions.
A handful of times I’ve said the first question to the guest because I want to get off on a good start. I don’t want them to be all rambling and lost when they start.
And, I should also say that if you go back to the first episodes of the program, I am often fumbling around trying to get the conversation started. And, I realized pretty early on that that’s when listeners leave the room: you switch channels, turn it off, go to find something else. So, I always script the opening question and read it verbatim.
But after that, often I’ll improvise and go with the flow.
What else? Was there a conscious moment when I broadened the focus from economics?
I don’t think so, but it was gradual.
It was pretty clear to me when–a couple of things were clear to me at various points. After the Financial Crisis of 2008, it was pretty clear to me that I had to understand it. And I knew I didn’t. I knew almost nothing about finance. So, I did 20, 30 episodes to try to get tooled up on that. And, you came along for the ride. You heard those as well. And I figured you wanted to understand it as well as I did. And, as a result of that, I wrote a book Gambling with Other People’s Money, trying to understand the incentives that may have contributed to the Financial Crisis. So, that was really a wonderful intellectual journey that I couldn’t have done without the different guests that I did, that I interviewed in those days.
And then, starting around the rise of Trump in the mid-2010s–2015, 2016, it was also clear to me at that point that standard economics was not what people cared about at the moment, and I started trying to understand what was going on in America that led to Trump.
And by that I meant–and I’ve said this many times–I don’t think of Trump as much as, until recently, I would say he was much more of an effect rather than a cause. Right now he’s doing some causes, he’s causing some things. It’s a very different time. We may do a future episode on that, a follow-up to the one I did with Mike Munger on DOGE. But, the general broadening was the realization that economics, part of it was economics is too narrow to understand what’s going on. This isn’t just about ‘the economy, stupid’; it’s about other things.
And, the other part was that I found economics less interesting. I found economics less able to contribute to my understanding of the world. There were many, many other things I was interested in. And, every once in a while I reflect on the reality that while that is true–that I am a little tired of economics–it’s also the case that a little economics goes a very long way in preventing bad reasoning.
So, even though I sometimes have become more skeptical of economics as a way of thinking, or I’ve softened my views on what is some of my dogmatism–I’ll give you an example of that in a few minutes–even though I’ve softened my views on that, it really is useful to understand some of the basics.
Russ Roberts: But having said that, I have a joke that–I have listeners, I’m sure some of you are still out there who would love to have an episode on Bitcoin every week; or on monetary economics generally or some topic that you’re passionate about. Which I respect. The problem is, is that I’m sure there are things I don’t understand about Bitcoin, quite a few things, but I understand what I want to understand. I’ve kind of learned what I think I need to learn about Bitcoin.
Just like I’d learned what I needed to learn about the Financial Crisis. I’m sure there are things I don’t understand about it. I’m sure there are other theories. But, many of those theories are just not interesting to me. And, maybe I’m wrong. Maybe I’m missing a chance to learn about some incredible theory of the Financial Crisis or some aspect of Bitcoin that’s revolutionary that I’m neglecting. But, what I’ve done in those topics is I’ve done a bunch of interviews. Same with AI right now. I’ve done a bunch of interviews on AI. There are more things to learn, there are different perspectives, but I’ve learned a lot, and I hope you have, too; and it’s time to move on to other topics.
I would say the same thing about Israel. I moved to Israel in 2021. I’m Jewish. You’d think I’d know a lot about the founding of the state and the Arab-Israeli crisis and what became the war in Gaza. But the fact is, I didn’t know a lot about it when I moved here. I knew something about it. So, a lot of those episodes–and I don’t know, we did 14 or so on Jewish history, Israeli history. And I tried to bring in voices from across the spectrum. I struggled to find Palestinian voices. I have some in the archive, but the truth is some Palestinians said they would be on the program and then we couldn’t ever find a date. I tried a few times. It was clear for whatever reason, they either didn’t want to be on the program or had bigger fish to fry. And most of them just never responded to my emails. For whatever reason. There’s no point in speculating what the reason is.
I want you to know out there that I did try. But, even without those voices, I tried to get some variety in the perspectives on the Arab-Israeli relationship and on the history of the founding of the state and the post-state problems that Israel has had with its neighbors and October 7th and its aftermath and so on.
But a lot of people didn’t like that. I’ve lost listeners who were angry I was too pro-Israel, which I’m sorry about that. I am pro-Israel, but I am eager to learn other things. I’m eager to understand the Palestinians’ narrative; and I certainly understand that it’s complicated, which of course is sort of the motto of the show. It’s complicated.
Russ Roberts: ChatGPT asked:
ChatGPT: Are there moments that left me emotionally wrung out?
Russ Roberts: Sure. There were a couple episodes where people yelled at me, insulted me. I mostly kept my cool, I think. I try not to lose my cool. I would often get letters in those days, people saying, ‘How’d you stay calm?’
And, they’ll see me in person; and if I give a talk, if I’m giving a talk somewhere publicly and people come see me, they’ll always often remark how different my persona is on stage giving my own talk versus as host of EconTalk. And, they’ll say, ‘Wow, you’re so passionate or intense, but on EconTalk you’re so calm.’ And, I always say, ‘Well, yeah, when I’m doing EconTalk, I’m wearing my EconTalk hat.’ And, my EconTalk hat, which is what I figuratively wear when I’m the host is–and you can get an EconTalk hat, by the way. EconTalk swag is available. But I don’t wear that hat when I’m podcasting. That EconTalk hat means: when I’m the host, I’ve got to stay calm. And, it took a while. You’ll hear it: in the early days with people I don’t agree with, I’d get worked up. And it got to the point where I got less worked up and I got calmer even when I was insulted, which, fantastic skill to have, I’m very happy to have it. It’s not the case that I can always remember I’m wearing my EconTalk hat, because sometimes I’m not the host: I’m out in the world. [More to come, 26:33]
