Latest / What's Your Problem? / When the Robots Take Over… from Cautionary Tales

Transcript

When the Robots Take Over… from Cautionary Tales · What's Your Problem?

0:15
Speaker 1: Pushkin.
0:20
Speaker 2: Hey, it's Jacob. I recently was a guest co host
0:23
Speaker 2: on another Pushkin podcast shows called Cautionary Tales. The host
0:27
Speaker 2: is Tim Harford, and it was a lot of fun
0:30
Speaker 2: and I thought you might enjoy hearing it while What's
0:32
Speaker 2: Your Problem is on a break, so we'll be back
0:35
Speaker 2: with What's Your Problem in a few weeks. In the meantime,
0:37
Speaker 2: I hope you like this episode of Cautionary Tales.
0:43
Speaker 1: Let's play around Warders Studios. It sounds very elvin.
0:48
Speaker 3: I always think of David Bowie when I think of Wardog.
0:50
Speaker 3: In one of his early songs talks about bright light,
0:53
Speaker 3: soho wardour Street.
0:55
Speaker 1: Bowie himself has an elfin aspect.
0:59
Speaker 3: He said he does is breakthrough gig at Aylesbury, my hometown,
1:03
Speaker 3: which basically is about the only thing remarkable about Aylesbury
1:06
Speaker 3: is well and also.
1:08
Speaker 1: Home of Tim Harford. As the plaque.
1:09
Speaker 3: Says, yeah, they've got a statue for Bowie, not yet
1:13
Speaker 3: for me. I don't why shore we go Okay, let's go, okay,
1:17
Speaker 3: I'm ready. Hello and welcome back to another episode of
1:43
Speaker 3: Cautionary Questions, our first of twenty twenty four. I am,
1:47
Speaker 3: of course Tim Harford. You are our loyal listeners. You've
1:50
Speaker 3: been sending in your burning questions on money, technology, economics,
1:55
Speaker 3: and problem solving, and thank you so much to everyone
1:58
Speaker 3: who's done so. And today is the day I do
2:02
Speaker 3: my best to answer them. And thankfully I won't be
2:04
Speaker 3: alone in this endeavor. Here to help me out both
2:07
Speaker 3: of the questions and I think with some of the
2:09
Speaker 3: answers is the brilliant, brilliant Jacob Goldstein, the host of
2:15
Speaker 3: Pushkin podcast What's Your Problem, author of the book Money,
2:19
Speaker 3: The True Story of a Made Up Thing. Jacob was
2:22
Speaker 3: my inaugural Cautionary Questions co host. Jacob's wonderful to.
2:25
Speaker 1: Have you back, Tim. It's an honor. Hi.
2:28
Speaker 3: So we should just get home with the questions because
2:30
Speaker 3: we always have so many and so much to say. So,
2:32
Speaker 3: what have you got in real big bag of listener
2:34
Speaker 3: questions for me?
2:36
Speaker 1: Let's start with a question from Alex in Melbourne, Australia.
2:40
Speaker 1: Alex writes, I work with artificial intelligence image generation software
2:45
Speaker 1: almost daily now, and I'm quickly seeing how so much
2:49
Speaker 1: of my workforce and processes can either be sped up
2:52
Speaker 1: or entirely replaced by AI, and this makes me nervous.
2:56
Speaker 1: And then he asks about universal basic income, right, this
3:00
Speaker 1: idea of a government giving all its citizens' money, and
3:03
Speaker 1: he says it seems for the first time that computers
3:05
Speaker 1: and software will actually replace jobs in a deeply concerned way,
3:09
Speaker 1: which is both exciting and terrifying. What are your thoughts
3:13
Speaker 1: on UBI universal basic income as a solution to an
3:16
Speaker 1: AI crisis and the widespread philosophical and economic implications of this.
3:22
Speaker 3: I love this question. I mean, it's so big, and
3:26
Speaker 3: I think the first thing to say is we don't
3:28
Speaker 3: really have any idea. If what Alex is thinking about
3:33
Speaker 3: comes true, and if most people just have no economic value,
3:37
Speaker 3: they have value as human beings, have value as members
3:39
Speaker 3: of society, but there's nothing that they could actually sell
3:41
Speaker 3: their labor to do, then that's completely uncharted territory. We've
3:46
Speaker 3: never been anywhere like that before, so everything we do
3:50
Speaker 3: is kind of speculative.
3:51
Speaker 1: We've feared it for a long time, right, We've have
3:54
Speaker 1: two hundred years of being afraid of technological unemployment. And
3:59
Speaker 1: my prior on this is to be somewhat skeptical. Right, Like,
4:04
Speaker 1: clearly there can be some large number of people who
4:08
Speaker 1: lose their jobs. We should be concerned about that, and
4:10
Speaker 1: we should think about how to mitigate that. But the
4:12
Speaker 1: idea of more or less everybody losing their jobs I'm
4:16
Speaker 1: skeptical of for the simple reason that it hasn't happened
4:19
Speaker 1: in two hundred years of incredible technological progress. And right now,
4:22
Speaker 1: after decades of extreme technological progress, in the US, unemployment
4:27
Speaker 1: is below four percent, the share of working age people
4:30
Speaker 1: who are working is near all time highs, And so
4:33
Speaker 1: somehow we keep coming up with new things to do
4:37
Speaker 1: for money, no matter how many things computers can do.
4:40
Speaker 1: And so my first thought is, I don't think we're
4:44
Speaker 1: going to have everybody losing their jobs to AI. I
4:46
Speaker 1: definitely could be wrong, but that's what I think.
4:48
Speaker 3: No, I think that's a good working assumption. If you
4:52
Speaker 3: think back a few centuries, basically, almost all the labor
4:55
Speaker 3: that people did, they might wash their clothes occasionally, Well,
4:58
Speaker 3: that's been outsourced to the washing machines that spend a
5:01
Speaker 3: lot of time moving water around, just drinking water, cooking water,
5:04
Speaker 3: throwing out human excrement. That's all now handled by automated systems.
5:09
Speaker 1: Digging people did a lot of digging, right, pulling a plow.
5:13
Speaker 3: Almost everything we used to do is now done by machines,
5:16
Speaker 3: but somehow we still all have jobs. Let's at least
5:19
Speaker 3: accept the premise that maybe this time it might be
5:21
Speaker 3: different because the robots are doing everything. There's still material prosperity.
5:26
Speaker 3: There's food out there, in all the services we could
5:29
Speaker 3: possibly want. We just have to find some system whereby
5:32
Speaker 3: the humans who have no economic value get to enjoy
5:36
Speaker 3: all this cool stuff that's being produced by the machines.
5:39
Speaker 1: Yeah, and you know, technological prosperity has given us in
5:44
Speaker 1: the developed world ubi for old people, right in the US,
5:49
Speaker 1: as in every developed country as far as I know,
5:52
Speaker 1: once you get to some age, if you are a
5:54
Speaker 1: citizen and you have worked, yeah, the government gives you
5:57
Speaker 1: money yea every month until you die, right, and so
6:00
Speaker 1: you could imagine a kind of creeping extension of that. Certainly,
6:05
Speaker 1: right now they're not talking about lowering the retirement age
6:08
Speaker 1: in the US. They're about raising it as in many countries.
6:11
Speaker 3: Yeah, raising it in the UK as well, But it's
6:13
Speaker 3: still a long way away from life expectancies. My recollection
6:16
Speaker 3: is that when Bismarck introduced the first pension, which was
6:18
Speaker 3: in Germany in the late nineteenth century, I think the
6:23
Speaker 3: pension started to be paid at the age sixty seven
6:26
Speaker 3: and the life expectancy was sixty three, you.
6:30
Speaker 1: Get negative four years in expectation.
6:33
Speaker 3: People who survive long enough to claim any pension at
6:35
Speaker 3: all are already exceptional. So that would be like having
6:38
Speaker 3: a pension today that started at age ninety. Like some
6:41
Speaker 3: people will get it by most bad road. But actually
6:43
Speaker 3: a lot of people could easily collect thirty years of pension,
6:46
Speaker 3: certainly twenty years. So they're living a large proportion of
6:51
Speaker 3: their adult life receiving money from the state, and they're
6:53
Speaker 3: also receiving money from their own savings, their own investments,
6:57
Speaker 3: which might be a replacement for UBI. Maybe we all
7:00
Speaker 3: just have shares in the robots instead, we have shares
7:02
Speaker 3: in Google or whatever, and that's how we get paid.
7:05
Speaker 1: Yes, whether that is mediated by the government or not
7:07
Speaker 1: looks somewhat different. Right, Either the government is taxing the
7:11
Speaker 1: owners of in Nvidia and Google stock and distributing the money,
7:15
Speaker 1: or everybody owns Google and Nvidia stock. There is the
7:19
Speaker 1: piece of this which is about the non financial parts
7:23
Speaker 1: of work, right, I mean, there's a political economy piece,
7:26
Speaker 1: the sort of bridge. How do we get from here
7:28
Speaker 1: to there? If that happens and that's complicated and maybe ugly.
7:31
Speaker 3: If the robots can do all of the stuff the
7:34
Speaker 3: computers can do it, there's no economic reason why humans
7:38
Speaker 3: couldn't just receive whatever an allowance they're given ten robots,
7:42
Speaker 3: or they're given ten thousand dollars a month to spend
7:45
Speaker 3: or whatever they like. There's no economic reason why that
7:47
Speaker 3: couldn't happen. But I think what you're getting at is
7:49
Speaker 3: what does it do to us if we're in that situation?
7:52
Speaker 1: Yeah? Yeah, Like I don't want to not have a job.
7:55
Speaker 1: I recognize that I am fortunate to have a job
7:58
Speaker 1: that I enjoy, that I derive a part of my
8:01
Speaker 1: identity from. I recognize that for many, in fact, probably
8:04
Speaker 1: most people, a job is not that it's some unpleasant
8:07
Speaker 1: thing they do because they need the money, and if
8:10
Speaker 1: they got more money, they would quit their job. Well,
8:13
Speaker 1: you probably know the kind of empirical evidence on this
8:15
Speaker 1: better than I do. Like, people have looked at lottery winners, right,
8:18
Speaker 1: and my sense is it's not great for you to
8:20
Speaker 1: win the lottery and quit your job. It doesn't actually
8:23
Speaker 1: make you happier.
8:24
Speaker 3: Actually, the evidence on lottery winners is a bit mixed,
8:26
Speaker 3: and I think slightly we hear all the disaster stories
8:29
Speaker 3: of lottery winners.
8:30
Speaker 1: Actually, it's great to win the lottery. That's amazing. Is
8:32
Speaker 1: that true.
8:33
Speaker 3: I think it's fine. Yes, it's just fine to win
8:35
Speaker 3: the lottery. Is not really a problem to win the lottery.
8:38
Speaker 3: In my vague recollection I wrote about this a couple
8:40
Speaker 3: of years ago. But we may not all want a job,
8:44
Speaker 3: but we all want something to do. We all want
8:47
Speaker 3: to feel useful, we all want a sense of some
8:51
Speaker 3: kind of purpose. We all want, I think, that experience
8:54
Speaker 3: of mastery, that experience of knowing that you can do
8:56
Speaker 3: something that not everybody can do. Those things don't have
9:00
Speaker 3: to come from a job, but for a lot of
9:01
Speaker 3: people they do come from a job.
9:02
Speaker 1: Yeah. For me, in my narrow provincial experience of life,
9:07
Speaker 1: it's frankly hard to imagine and getting those things without
9:10
Speaker 1: a job. Yeah.
9:12
Speaker 3: It comes back to my initial reaction to Alex's terrific question,
9:15
Speaker 3: which is this is such unknown territory that we can
9:18
Speaker 3: only really speculate. But I think you and I jcub
9:20
Speaker 3: are in agreement that the fundamental issue here is not economic.
9:24
Speaker 3: It's really to do with our souls. How would we
9:27
Speaker 3: react if our desire for master. We our desire for meaning,
9:30
Speaker 3: our desire to feel useful if that had to be
9:33
Speaker 3: satisfied without having a job, and what would we do
9:37
Speaker 3: and could we cope? And I don't know, I.
9:39
Speaker 1: Mean, Harford, if the robots come and take our jobs,
9:42
Speaker 1: let's just you and me make a podcast for free.
9:44
Speaker 1: I'm in. I'll commit right now in the robot Utopia
9:48
Speaker 1: apocalypse to making a weekly podcast with you for free.
9:52
Speaker 3: Deal.
9:52
Speaker 1: Sorry, we have no more time. You might give in
9:55
Speaker 1: your robot overlord. Give us another question, Hello, Race Ubinov,
10:03
Speaker 1: I would like to ask, with the onset of AI,
10:05
Speaker 1: what is the next cautionary tale you anticipate talking about
10:09
Speaker 1: in the years to come?
10:11
Speaker 3: Hello Ivanough, I have just finished the first draft of
10:14
Speaker 3: a script about the coming of AI, and without giving
10:19
Speaker 3: too much away, the cautionary note is about what happens
10:23
Speaker 3: when automation gets so good that we lose our own skills.
10:28
Speaker 3: We hand over controlled the machine, and then how do
10:31
Speaker 3: we respond when the machine says, actually, this one's too hard?
10:35
Speaker 3: Could you take over again? Suddenly we're back at the
10:38
Speaker 3: wheel and we're out of practice.
10:40
Speaker 1: Harford, it's not another plane crash, is it. Don't you
10:43
Speaker 1: have a moratory. I'm undering plane crashes at this point.
10:45
Speaker 1: Because it sounds like that.
10:46
Speaker 3: I'm going to say nothing more fair enough. Let's go
10:49
Speaker 3: for a break.
10:50
Speaker 1: All right, Yeah, we're back.
11:00
Speaker 3: I'm talking to Jacob Goldstein, the host of What's Your Problem,
11:04
Speaker 3: and we are doing a Cautionary Questions Q and A episode.
11:09
Speaker 3: Jacob has the questions. Jacob's also helping with the answers.
11:12
Speaker 3: I'm really sepless to requirements here, but I'm doing my best, Jacob.
11:15
Speaker 3: What have you got for me?
11:16
Speaker 1: Tim? Our next question comes from Adam, who writes, Hi,
11:21
Speaker 1: I have a question about investing in cryptocurrency. Oh dear,
11:24
Speaker 1: oh dear, come on give it to them. Okay. It's
11:28
Speaker 1: an interesting one. Traditionally, when buying shares in a company,
11:32
Speaker 1: you would be supporting that company to grow, create a
11:35
Speaker 1: new product, or enter a new retail space. This would
11:38
Speaker 1: hopefully create jobs and new products. When people invest money
11:42
Speaker 1: into cryptocurrency or any currency for that matter, isn't that
11:46
Speaker 1: money just sitting around not doing anything until you withdraw it.
11:49
Speaker 1: Would investing in crypto be bad for the economy compared
11:53
Speaker 1: with investing in businesses?
11:55
Speaker 3: This is a deep question. I really like it. It is
12:00
Speaker 3: as the author of the wonderful book Money, The True
12:03
Speaker 3: Story of a Maid Art Thing. I am sure you
12:05
Speaker 3: have thoughts, Jacob. They may have a first crack and
12:07
Speaker 3: you can tell me everything I've made.
12:09
Speaker 1: Yeah. Yeah.
12:09
Speaker 3: So let's say you buy bitcoin or whatever, and then
12:12
Speaker 3: the money's just sitting there because bitcoin is not building anything.
12:16
Speaker 3: You're not buying investment in, say a road building company,
12:19
Speaker 3: or you're not buying an investment in Google, which is
12:21
Speaker 3: developing new technology. There's one of two possible things that happens.
12:25
Speaker 3: So most likely is whoever you bought the bitcoin from
12:29
Speaker 3: now has your money. Well, now it's their money, and
12:33
Speaker 3: then they're going to do something with that money. So
12:34
Speaker 3: maybe they then buy shares in Google, or they then
12:37
Speaker 3: set up a business, or they will then go on
12:40
Speaker 3: to do something with that money, or they'll lend it
12:43
Speaker 3: to somebody else and that person will do something with it.
12:45
Speaker 3: Eventually the money will find its way into some productive investment.
12:49
Speaker 3: The fact that you bought bitcoin off somebody does not
12:52
Speaker 3: mean that they won't then do something useful with the money.
12:55
Speaker 3: But then let's say for some reason they just go,
12:57
Speaker 3: you know what, I'm just going to sit on this money.
12:59
Speaker 3: It's just dollars. I'm going to stick them under the mattress.
13:01
Speaker 3: I guess a bitcoin guy might do that. That would
13:05
Speaker 3: reduce inflationary pressures on the economy, which is something we
13:08
Speaker 3: want anyway. And if inflationary pressures are reduced too much,
13:13
Speaker 3: the Federal Reserve could then go, you know what, guys,
13:16
Speaker 3: we should probably print some money or otherwise stimulate the economy. Actually,
13:20
Speaker 3: the worst thing that could possibly happen is that the
13:23
Speaker 3: money gets invested in buying more computers that could be
13:27
Speaker 3: put to better use, but in fact just end up
13:29
Speaker 3: doing bitcoin mining. I guess what I'm saying is, as
13:31
Speaker 3: long as they don't spend it on bitcoin mining computers,
13:33
Speaker 3: it's totally fine. Well what am I missing?
13:36
Speaker 1: I went around the same track as you went around.
13:38
Speaker 1: And you know, different cryptocurrencies have different methods of regulation, right,
13:43
Speaker 1: So Bitcoin by design is this very energy intensive computing
13:48
Speaker 1: process in order to mine bitcoin, and that is, in
13:51
Speaker 1: my opinion, a bad outcome. Socially, I agree, because you
13:55
Speaker 1: could imagine a world where maintaining the bitcoin network was
13:58
Speaker 1: socially desirable. In the world as it has evolved, I
14:00
Speaker 1: don't find it particularly socially desirable, but at this point,
14:03
Speaker 1: I don't think most of the money going into bitcoin
14:06
Speaker 1: is going to miners, right, there's a large stock of
14:09
Speaker 1: coin that exists in the world. I mean, an interesting
14:12
Speaker 1: thing about money is it almost never sleeps. It keeps going,
14:16
Speaker 1: and so you can always ask what happens next, you know,
14:20
Speaker 1: with the stock market, sometimes people talk about money on
14:23
Speaker 1: the sidelines when they're like, well, the stock market can
14:25
Speaker 1: go up somewhere because there's a lot of cash on
14:27
Speaker 1: the sidelines. And there's this famous billionaire investor, Cliff Asnes.
14:31
Speaker 1: He's an interesting guy. You got a PhD from the
14:33
Speaker 1: University of Chicago, studied with Gene Fama, famous economist, and
14:37
Speaker 1: Cliff Asnes gets driven up the wall when people say
14:41
Speaker 1: cash on the sidelines for the same reason you just said, Right,
14:43
Speaker 1: Like every time somebody buys a share of stock, someone
14:47
Speaker 1: else is selling it, right, So like the money is
14:50
Speaker 1: changing hands, the stock is changing hands. But like there
14:54
Speaker 1: are no sidelines with money.
14:56
Speaker 3: Yeah, and even if someone puts just cash in a
14:59
Speaker 3: checking account, well then now the bank has money, and
15:03
Speaker 3: you can bet the bank is going to want to
15:04
Speaker 3: do something with that money, something productive. And even if
15:07
Speaker 3: they doesn't put it in the bank, as I say,
15:10
Speaker 3: it's a dramatris. Well, even then, the Federals could always
15:13
Speaker 3: put money.
15:14
Speaker 1: Yeah, the stock of money is not fixed, right. I
15:17
Speaker 1: think there's an assumption in the question I think goes
15:21
Speaker 1: back to the pre modern era, the era before the
15:24
Speaker 1: nineteen thirties when money was based on gold or silver
15:27
Speaker 1: and it was finite. Right, there was a fixed amount
15:30
Speaker 1: of gold and silver in the world in those days. Indeed,
15:33
Speaker 1: if you sat on money, you were effectively reducing the
15:36
Speaker 1: supply of money in the world. But to your point,
15:39
Speaker 1: the supply of money in the world is as big
15:41
Speaker 1: as the Central Bank wants it to be, and so
15:44
Speaker 1: it's not a meaningful constraint on economic growth. So the
15:47
Speaker 1: ultimate answer to the question is if you are just
15:50
Speaker 1: sitting on money, it's not really going to have an
15:53
Speaker 1: effect on the broader economy.
15:54
Speaker 3: Well, I think we've come to agreement. Hopefully that makes
15:56
Speaker 3: sense to Adam. Next question.
16:00
Speaker 1: Our next question comes from doctor Yvonne Couch, who is
16:04
Speaker 1: Associate Professor of Neuroimmunology Alzheimer's Research UK Fellow, Associate Research
16:11
Speaker 1: Fellow at Saint Hilda's College, Stipendary Lecturer at Somerville College,
16:16
Speaker 1: University of Oxford.
16:18
Speaker 3: Yeah, University of Oxford. I know, well, one of my
16:20
Speaker 3: sisters went to St. Hilda's College and one of my
16:22
Speaker 3: sisters went to Somerville College, so this feels very much
16:25
Speaker 3: in the family. So what does Dr Couch got to say?
16:28
Speaker 1: She writes, my question was vaguely economics based, although probably
16:33
Speaker 1: not very is the way we currently fund science feasible
16:38
Speaker 1: going forward? Do we just have too many scientists and
16:42
Speaker 1: not enough resources? All the best, Yvonne, oh love it.
16:47
Speaker 3: Let me throw out a few thoughts. Then I think
16:49
Speaker 3: that probably there is a problem with the way we
16:53
Speaker 3: fund science, But there's no one way that we fund science.
16:56
Speaker 3: You have university research, you have various grands, various sources
16:59
Speaker 3: of funding, philanthropy and so on. You also have private
17:03
Speaker 3: sector research, which is often incentivized by the patent system.
17:08
Speaker 3: And then you have big block grants that are handed
17:10
Speaker 3: out by agencies such as the National Institutes for Health.
17:15
Speaker 3: So there are lots of different ways that science gets funded.
17:19
Speaker 3: A couple of things that worry me is that, first
17:24
Speaker 3: of all, there's a big incentive to make really incremental
17:27
Speaker 3: improvements rather than to take risks. There's a great economics
17:33
Speaker 3: paper studying scientists who are funded by the National Institutes
17:37
Speaker 3: for Health, which is a wonderful institution, very important, and
17:41
Speaker 3: scientists who on paper seem to be the same that're
17:45
Speaker 3: on the same career track, got very similar publication records,
17:48
Speaker 3: and they are instead funded by a private foundation called
17:51
Speaker 3: the Howard Hughes Medical Institute. Listeners who want to hear
17:55
Speaker 3: more about Howard Hughes and the Howard Hughes biography can
17:58
Speaker 3: go to the back catalog of caution details. So the
18:00
Speaker 3: Howard Hughes Medical Institute basically takes risks. It wants people
18:05
Speaker 3: to do something new. It's happy with a high risk
18:08
Speaker 3: of failure as long as there's some chance of a
18:10
Speaker 3: real breakthrough success. And this particular paper studying the results
18:16
Speaker 3: that come from these two funding systems basically finds that
18:20
Speaker 3: the grant funders get what they pay for. So the
18:23
Speaker 3: National Institutes for Health get a high success rate, but
18:26
Speaker 3: it's often quite incremental progress. And the Howard Hughes Medical
18:29
Speaker 3: Institute has lots and lots of failures, but when it succeeds,
18:32
Speaker 3: it really succeeds. And they these are hugely important papers,
18:36
Speaker 3: and I just feel that we're not deliberate enough about saying, well,
18:40
Speaker 3: how much of our funding ecosystem should be aiming at
18:46
Speaker 3: kind of venture capital style for really big wins, and
18:49
Speaker 3: how much should be incremental. I think those are not
18:51
Speaker 3: the sorts of questions that get asked, So that's one
18:54
Speaker 3: of the things that worries me. What's your take, Jacob.
18:57
Speaker 1: One of the things that I have read is that
18:59
Speaker 1: over time the average age of grant recipients from the
19:03
Speaker 1: NIH has gone up and up and up. So kind
19:06
Speaker 1: of you just think of this whole universe as getting
19:08
Speaker 1: all and more risk averse and more kind of bureaucratic.
19:13
Speaker 1: There is an interesting set of counter pressures I think
19:16
Speaker 1: rising up partly out of Silicon Valley. I don't know
19:19
Speaker 1: if you've come across the work of Patrick Hollison and
19:23
Speaker 1: his brother. They're from Ireland. They found it Striped. They're
19:26
Speaker 1: very rich. Stripe is a big company that does basically
19:30
Speaker 1: payment stuff online. They have a really interesting set of
19:34
Speaker 1: kind of philanthropic endeavors around the idea of progress. You know,
19:39
Speaker 1: they're trying to create a kind of field of progress
19:42
Speaker 1: studies that is, it's very meta. What are the conditions
19:46
Speaker 1: that best foster technological and scientific progress?
19:51
Speaker 3: But this question from doctor Couch, he is very massive. Yes,
19:54
Speaker 3: it's very much progress studies, isn't it.
19:56
Speaker 2: Yes.
19:56
Speaker 1: There is this institute in the Bay Area called the
20:00
Speaker 1: ARC Institute that hires leading scientists. The basic idea is
20:04
Speaker 1: give talented people freedom and money, right and encourage them
20:10
Speaker 1: to take big swings. And they are interested, you know,
20:13
Speaker 1: not just in outcomes, but in new tools. Right. Again,
20:16
Speaker 1: continuing on the nerdy thing, like if you take a
20:19
Speaker 1: tool like Crisper. Crisper is an intermediate tool that allows
20:22
Speaker 1: people to cut up a genome. Basically, we just had
20:24
Speaker 1: the first transplant from an animal into a human a
20:27
Speaker 1: few weeks ago because of Crisper. We have the first
20:30
Speaker 1: treatments for sickle cell disease because of Crisper. So I
20:33
Speaker 1: do think there is a wave of people trying to
20:38
Speaker 1: rethink scientific funding. There is a bottom line aspect to
20:43
Speaker 1: this question from doctor Couch that I don't feel like
20:46
Speaker 1: I know enough to answer, and I'm curious if you do.
20:48
Speaker 1: I mean, the question, is the way we currently fund
20:51
Speaker 1: science feasible going forward? There is a yes no version
20:55
Speaker 1: of the answer. Do you think you know? Is it?
20:58
Speaker 3: I don't know the answer, but I guess I don't
21:00
Speaker 3: know either. Another thing that concerns me on this, which
21:04
Speaker 3: is potentially really existential is the question of how we
21:08
Speaker 3: fund new antibiotics. I think this points to a real
21:13
Speaker 3: weakness in the ecosystem of research funding. So if you
21:17
Speaker 3: think about the basic way we develop drugs, the fundamental
21:20
Speaker 3: incentive is the patent. So a drug company spends a
21:23
Speaker 3: lot of money, tries to develop drugs. Some of them work,
21:26
Speaker 3: some of them fail, a lot of them fail, but
21:28
Speaker 3: in the end you have a drug and the deal
21:31
Speaker 3: is we give you this artificial monopoly called a patent,
21:34
Speaker 3: and it will only last for a certain amount of time,
21:37
Speaker 3: which is kind of a problem because it takes so
21:39
Speaker 3: long to develop the drugs. Maybe the patent's nearly expired.
21:42
Speaker 3: You can charge an incredible amount for these drugs for
21:45
Speaker 3: a while, and then your patents will run out, and
21:47
Speaker 3: then loads of as people will make the same drug
21:49
Speaker 3: and it'll come down in value. For example, Viagra, you
21:52
Speaker 3: know you could sell this for a huge amount of money,
21:54
Speaker 3: and now Viagra is off patent and anyone can make
21:57
Speaker 3: a generic Viagra. So that's the incentive that we're given
22:00
Speaker 3: to private companies, that you will have this temporary monopoly.
22:04
Speaker 3: So now think about antibiotics. The problem with antibiotics. We
22:08
Speaker 3: have lots of antibiotics work really well, except the bacteria
22:11
Speaker 3: have figured them out. And why have they figured them out?
22:14
Speaker 3: Because we've used them. The bacteria developed resistance. So what
22:18
Speaker 3: we really need is new antibiotics that we don't use.
22:23
Speaker 3: And now think about how the pattern system deals with that.
22:26
Speaker 3: So you basically say, if you develop a new antibiotic,
22:30
Speaker 3: we'll give you a tempery monopoly, but we really need
22:33
Speaker 3: you to just not sell any don't sell any of.
22:36
Speaker 1: This thing except in cases of dire emergency. It's like
22:39
Speaker 1: a break glass in case of emergency. I mean, I
22:42
Speaker 1: feel like a bounty. Like every econ nerd storyteller loves
22:46
Speaker 1: a good bounty. Right, wasn't this done with a malaria vaccine,
22:50
Speaker 1: which is actually coming along quite well. You have some
22:52
Speaker 1: universe of people say we will pay a billion dollars
22:56
Speaker 1: to anyone who comes up with a new antibiotic that
22:59
Speaker 1: meets this set of criteria, that treats this set of
23:02
Speaker 1: bugs that are resistant to these existing antibiotics, and we'll
23:06
Speaker 1: give you the money, and you give it to us,
23:07
Speaker 1: and we'll put it on the break glass in case
23:09
Speaker 1: of emergency. Shelf.
23:11
Speaker 3: You've used the word bounty, but what the term that
23:13
Speaker 3: is normally used is an advanced market commitment. These were
23:16
Speaker 3: proposed most famously by Michael Kramer, who's a Nobel Prize
23:19
Speaker 3: winner in economics. And basically, the way this price, this
23:23
Speaker 3: bounty tends to get paid is as a kind of
23:26
Speaker 3: extra payment on top of each dose you sell. Ah,
23:28
Speaker 3: so we'll give you extra for every kid that gets vaccinated.
23:32
Speaker 1: So or back to the same problem in that universe.
23:35
Speaker 3: Back to the same problem. But the reason you go, well,
23:36
Speaker 3: why does it have to be like that? It doesn't
23:38
Speaker 3: have to be like that. But the reason that they
23:40
Speaker 3: tend to be structured like that is because you need
23:42
Speaker 3: to demonstrate some kind of market demand. Somebody needs to
23:45
Speaker 3: be willing to buy your product. If they are, we'll
23:48
Speaker 3: give you an extra payment for every product you sell.
23:51
Speaker 3: But that wouldn't work for antibiotics. There's so many different
23:54
Speaker 3: ways in which science funding could be said to be broken.
23:57
Speaker 3: But I think doctor Couch is right to raise the issue.
24:01
Speaker 3: We need to do a lot more of this kind
24:02
Speaker 3: of meta thinking about progress studies.
24:05
Speaker 1: Tim that was a lot of answer. Let's take a break.
24:08
Speaker 1: Cautionary to e will be back in a moment Edward Back.
24:19
Speaker 1: I'm Jacob Goldstein here with Tim Harford on cautionary tales.
24:22
Speaker 3: Hello, Jacob, more questions? What have you got for me?
24:25
Speaker 1: All right, Tim, I got another one for you. Comes
24:29
Speaker 1: from Graham in Florida. Graham Rights. I'm fascinated by errors
24:35
Speaker 1: made by falsely identifying correlation as causation. What examples of
24:41
Speaker 1: this error stand out to you?
24:43
Speaker 3: I love the question, but I'm going to slightly sidestep
24:45
Speaker 3: it because I'm worried by this question. But I think
24:48
Speaker 3: that it's generally more complicated than simply some fool sora
24:55
Speaker 3: correlation and thought it was causation.
24:58
Speaker 1: You don't want to talk about sun spots and crap yields.
25:01
Speaker 3: Well, let's talk about stocks and babies briefly, Okay.
25:04
Speaker 1: A classic, A classic of the genre.
25:06
Speaker 3: Yeah, the most successful book have a public statistics. How
25:10
Speaker 3: to Lie with Statistics by Daryl Huff includes this example
25:14
Speaker 3: of stalks and babies and shows that there's a correlation
25:17
Speaker 3: between the number of storks and the number of babies.
25:20
Speaker 3: And there are various ways to demonstrate this correlation. One
25:22
Speaker 3: way to do it is you just look at national populations.
25:25
Speaker 3: You go, hey, countries with lots of storks also have
25:27
Speaker 3: lots of babies, and there's a very very strong correlation.
25:30
Speaker 3: But of course the reason is big places like the
25:33
Speaker 3: United States have a lot of room for storks and
25:35
Speaker 3: a lot of room for babies, and small places like
25:38
Speaker 3: in a Luxembourg or the Vatican City don't have a
25:40
Speaker 3: lot of room for babies and don't have a lot
25:42
Speaker 3: of room for storks. So then you go, oh, that's
25:43
Speaker 3: a great example of this mistake. The sting in the
25:47
Speaker 3: tale of that story, as I describe in my book
25:50
Speaker 3: The Data Detective, is that Daryl Huff, the guy who
25:53
Speaker 3: created that story, then went on to tell the same
25:57
Speaker 3: story in congressional testimony, saying that there was no compelling
26:03
Speaker 3: evidence that smoking cigarettes would give you lung cancer, and
26:06
Speaker 3: it was just like stalks and babies, and he was
26:08
Speaker 3: actually high by the tobacco lobby. It was just correlation.
26:11
Speaker 3: It was just correlational. They seemed about the same. Sure, yeah,
26:15
Speaker 3: there are people who smoke, and that people who get cancer,
26:17
Speaker 3: but there's no cause of evidence. One theory was cigarettes
26:21
Speaker 3: are so soothing, and if you have some early symptoms
26:24
Speaker 3: of lung cancer, maybe you soothe that with the soothing
26:26
Speaker 3: vapors of cigarettes. I mean, it's completely ridiculous, but this
26:29
Speaker 3: sort of rhetoric was deployed. And so one of the
26:32
Speaker 3: things I'm very concerned about in my work on statistics
26:35
Speaker 3: is that it's great to be skeptical about statistics and
26:38
Speaker 3: to point out lots of examples of statistics being misused.
26:41
Speaker 3: But if that's all you do, you just get to
26:44
Speaker 3: a canihilistic situation where you're just constantly rejecting statistical evidence
26:49
Speaker 3: because it's just another of those damn lines and statistics.
26:52
Speaker 3: When you look at the real world, I think this
26:55
Speaker 3: gets to be incredibly interesting. So a real hot topic
26:59
Speaker 3: at the moment is our smartphones destroying a generation. Basically,
27:04
Speaker 3: are our kids having their mental health wrecked.
27:06
Speaker 1: There's a new book that basically makes that up.
27:08
Speaker 3: Yeah Yeah, by a prominent ecademic yea by Jonathan Hyde,
27:12
Speaker 3: and there have been others by Gene Twegian. Lots of
27:14
Speaker 3: people have said this, and the evidence for it is
27:18
Speaker 3: mostly correlational, not completely. There are some experiments, but they're
27:22
Speaker 3: not that convincing. None of them are perfect, but there
27:25
Speaker 3: are lots of different ways of measuring it. And the
27:28
Speaker 3: really interesting evidence basically says look, there appears to be
27:32
Speaker 3: a mental health crisis. The kids seem to get really distressed,
27:35
Speaker 3: particularly the girls, when children have access to social media
27:41
Speaker 3: on their phones, and that happens around sometime between twenty
27:44
Speaker 3: ten and twenty fourteen. And at the same time, suddenly
27:47
Speaker 3: you've got this outbreak of suicidal ideation and self harm,
27:52
Speaker 3: poor mental health and so on. And that's correlational evidence.
27:55
Speaker 3: I don't completely believe it, but I don't completely not
27:58
Speaker 3: believe it either. I think that's what makes it interesting
28:00
Speaker 3: and makes it important to engage with. You have to
28:03
Speaker 3: start going, well, what's the alternative explanation? Is there something
28:05
Speaker 3: else that has happened sometime about sort of ten, twelve,
28:09
Speaker 3: fourteen years ago that might explain this mental health distress?
28:12
Speaker 3: So the timing of the Great Financial Crisis probably not
28:14
Speaker 3: quite right. Donald Trump, maybe Donald Trump is upsetting the kids.
28:18
Speaker 3: Timing doesn't work, so partly it does. The pattern of
28:21
Speaker 3: the correlation make enough sense to explain this causal story, which,
28:26
Speaker 3: funny enough, is basically exactly what the scientists finding a
28:30
Speaker 3: connection between cigarettes and lung cancer were doing. They only
28:33
Speaker 3: had correlational evidence for a long time. They weren't running
28:35
Speaker 3: randomized trials saying you know, half of you smoke and
28:38
Speaker 3: half of you don't smoke. I mean, they couldn't do that.
28:40
Speaker 3: They had to look at correlational evidence, and sometimes that's
28:43
Speaker 3: what we've got.
28:44
Speaker 1: This question got me thinking about the rise in social
28:48
Speaker 1: science of what they call natural experiments, right, trying to
28:52
Speaker 1: find instances in the real world where you have something
28:57
Speaker 1: that obviously is not as good as a randomized trial,
28:59
Speaker 1: because you just can't get that with large numbers of
29:01
Speaker 1: people in the world, but that gives you some element
29:05
Speaker 1: of randomization or pseudorandomization, something that allows you to try
29:09
Speaker 1: and make the leap from correlation to causation. And you know,
29:13
Speaker 1: this as far as I know, goes back to the
29:15
Speaker 1: Vietnam War, right, where in the US there was a
29:18
Speaker 1: draft lottery, and there were social scientists after the war
29:23
Speaker 1: who looked and said, oh, look, here are people who
29:25
Speaker 1: are at an aggregate level very similar on many dimensions
29:28
Speaker 1: socioeconomic dimensions. We can look at people who randomly got
29:33
Speaker 1: drafted versus those who randomly didn't, and who appear quite
29:37
Speaker 1: similar in the aggregate, and see how their outcomes differ.
29:40
Speaker 3: And that's a really good example of a natural experiment.
29:43
Speaker 3: Because it's actually very close to a really.
29:45
Speaker 1: Yeah, when you get a lottery in the real world. Now,
29:48
Speaker 1: social scientists flocked to it.
29:49
Speaker 3: Right.
29:50
Speaker 1: Similarly, there was one in Oregon some years ago with Medicaid,
29:55
Speaker 1: which is the healthcare program for low income people in
29:58
Speaker 1: the US, and Oregon got some new Medicaid funds and
30:02
Speaker 1: they randomly allocated them to a group of people over time, right,
30:06
Speaker 1: so that social scientists could say, oh, look, here are
30:08
Speaker 1: people all who are basically identical. Some of them got
30:11
Speaker 1: this health insurance and some of them didn't. And in
30:13
Speaker 1: that case the findings were quite interesting. Medicaid didn't appear
30:16
Speaker 1: to be as helpful as I would have thought, as
30:19
Speaker 1: the researchers themselves would have thought. According to the one
30:22
Speaker 1: I interviewed, lowered mental health problems. People worried less about money,
30:26
Speaker 1: but like their basic health outcomes didn't improve, which nobody
30:30
Speaker 1: would have guessed. Right. And the evidence is quite robust.
30:34
Speaker 1: When you don't have those lotteries. As you say, the
30:36
Speaker 1: world is just hard to understand. I mean, even in
30:39
Speaker 1: some instances where you do have randomized trials. People are complicated.
30:43
Speaker 1: The body is complicated, society is complicated, and so you know,
30:47
Speaker 1: correlation in a certain way I think is underrated, Like, yes,
30:51
Speaker 1: obviously it doesn't equal causation, but it's a place to
30:54
Speaker 1: start looking, right, it's a place to start asking questions.
30:57
Speaker 3: I think that's absolutely right, and we need more randomized experiments.
31:00
Speaker 3: There are more opportunities to run them than people seem
31:03
Speaker 3: to think. For example, one of the things that John
31:05
Speaker 3: Hyde in his book is arguing for is, you know,
31:07
Speaker 3: shouldn't have smart funds in school Schools should be phone
31:11
Speaker 3: free zones. I could completely imagine a state saying we're
31:14
Speaker 3: going to introduce a rule whereby in all of the
31:17
Speaker 3: schools in the state, no smartphones strictly forbidden. You have
31:21
Speaker 3: to put them in a locker when you show up
31:23
Speaker 3: and then unlock them at the end of the day.
31:25
Speaker 3: You could introduce that rule and just go, well, we
31:27
Speaker 3: introduce it for a semester at random in fifty percent
31:31
Speaker 3: of the schools, and then we'll introduce it in the
31:34
Speaker 3: other fifty percent of the schools in the next semester,
31:36
Speaker 3: and we'll just randomize that because we want to know
31:38
Speaker 3: whether there's any point in this experiment or not. That's
31:41
Speaker 3: not very difficult to do. It would create so much
31:44
Speaker 3: information about children's performance in the classroom, their mental health,
31:49
Speaker 3: that could inform policy across the world.
31:52
Speaker 1: That would be the dream. You know, there is now
31:55
Speaker 1: a robust set of methods essentially where social scientists could
31:59
Speaker 1: look at that state and compare it to neighboring states
32:02
Speaker 1: and see the difference in differences, as they say, see
32:05
Speaker 1: the change over time. And that's an instance where it
32:08
Speaker 1: wouldn't as elegant as randomizing within a state. But I
32:12
Speaker 1: feel like you could start to get pretty good data
32:14
Speaker 1: even if you did one state compared to other states.
32:18
Speaker 3: Yeah, if you've got no experiments, we should run them.
32:21
Speaker 3: If we don't run them, there are still ways of
32:23
Speaker 3: making correlation talk. Should we have another question?
32:28
Speaker 1: Okay, Tim, we got one more. We had to have
32:30
Speaker 1: one game related question for right, Yes, you're welcome. This
32:35
Speaker 1: one comes from Yost, who writes, Hi, Tim, diving into
32:40
Speaker 1: the board game shaped flank you've now exposed for discussion.
32:44
Speaker 1: That is a very gamersh way in what board games
32:48
Speaker 1: or board game mechanisms that you enjoy are particularly insightful
32:51
Speaker 1: on some aspect of real world economics offer a unique
32:55
Speaker 1: angle to look at a maybe niche problem and what
33:00
Speaker 1: non D and D games are you particularly enjoying right now?
33:04
Speaker 1: Love the show? However, many episodes you produce that is
33:08
Speaker 1: a listener? We all yeah, yeah, so kind your.
33:12
Speaker 3: There are a couple of mechanisms that I do see
33:14
Speaker 3: used quite often that shared important light on economics. So
33:17
Speaker 3: one is trading. A lot of games involved trading. Now Monopoly,
33:22
Speaker 3: that classic of sadly not a very good game. Oh
33:25
Speaker 3: there's a cautionary tell about the history of Monopoly if
33:27
Speaker 3: people want to listen. Monopoly in theory involves trading, but
33:31
Speaker 3: in practice, not a lot of trading happens. It turns
33:34
Speaker 3: out that you need to give people a reason to trade.
33:39
Speaker 3: The great modern game Settlers now about thirty years old.
33:42
Speaker 1: By the way, Settlers, I assume that Settlers of Catan
33:45
Speaker 1: as an outsider, not on a first name basis, with the.
33:47
Speaker 3: Game, Settlers of Katan is the game. It's super It's
33:50
Speaker 3: the game that Monopoly wishes it was. It's like if
33:53
Speaker 3: Monopoly had been designed with a modern eye to be
33:57
Speaker 3: super exciting. The people need different resources, and the supply
34:02
Speaker 3: of resources fluctuates, so sometimes you've got loads of wood,
34:05
Speaker 3: sometimes there's no wood. So there's an active incentive to
34:07
Speaker 3: trade all the time. By the way, if you don't trade,
34:09
Speaker 3: every now and then the robber Barron comes and take stuff.
34:11
Speaker 3: So it's an interesting insight that trading doesn't just happen
34:15
Speaker 3: because it's allowed. There needs to be some difference in
34:18
Speaker 3: value and perhaps some incentive to get on with it.
34:21
Speaker 3: So my personal favorite game is a gricler, a wonderful
34:25
Speaker 3: game about developing a farm, which doesn't sound promising, but
34:28
Speaker 3: it's really really good. But one of the clever things
34:30
Speaker 3: about a grickler is it uses an auction in an
34:33
Speaker 3: interesting way. But it's sometimes called a descending clock auction
34:37
Speaker 3: or a Dutch auction. The prize gets more and more tempting,
34:42
Speaker 3: so in a grickller there's just more and more good
34:44
Speaker 3: stuff on the board. In a traditional descending clock auction,
34:47
Speaker 3: basically the price is going down and down and down
34:49
Speaker 3: and down, and everyone is just sitting there and then
34:51
Speaker 3: it's a question of who grabs it first. The longer
34:54
Speaker 3: you leave it, the better it gets, but then only
34:57
Speaker 3: one person can have it. It's very very elegant. They
34:59
Speaker 3: sell flowers at Alsmere in the Netherlands and it's just
35:03
Speaker 3: incredibly quick, much much quicker than selling say art. You
35:07
Speaker 3: know you're selling a van god price rises and rises
35:09
Speaker 3: and rises.
35:10
Speaker 1: So auction design is a whole thing, right, Yeah, When
35:12
Speaker 1: would you choose the traditional prices going up eBay art
35:16
Speaker 1: style auction in the real world, and when would you
35:19
Speaker 1: choose a Dutch prices going down auction? What do you
35:22
Speaker 1: get in a relative sense out of each one of those.
35:24
Speaker 3: There's a couple of differences. One is that in certain
35:28
Speaker 3: types of ascending auction you get to see people dropping
35:32
Speaker 3: out as the price rises. It depends on the way
35:34
Speaker 3: the auction is designed. But you could imagine an ascending
35:37
Speaker 3: auction where you just see people going no, I'm out,
35:39
Speaker 3: I'm out, I'm out.
35:40
Speaker 1: Sort of the classic kind of art auction in a movie.
35:43
Speaker 3: Yeah, and so you're learning information as the price rises.
35:46
Speaker 3: How many people are still interested? Are there's still ten
35:48
Speaker 3: people interested? Is it just two people? So that generates information.
35:51
Speaker 3: By generating information, it might raise the final price. Overall.
35:56
Speaker 3: The advantage of the descending auction is it's just so quick.
36:00
Speaker 3: You can run an auction in ten seconds. It's one
36:04
Speaker 3: hundred thousand tulips. Here are the tulips. We're going to
36:06
Speaker 3: sell them for whatever, ten thousand years, nine hundred and
36:11
Speaker 3: thin thousand, and it's literally a clock. The clock just
36:14
Speaker 3: runs around showing what the price is and then someone
36:17
Speaker 3: presses the button. Sold, okay, now bringing the roses. So
36:21
Speaker 3: it's very fast.
36:22
Speaker 1: You would want to have a good idea of the
36:24
Speaker 1: market clearing price if you were going to run that
36:26
Speaker 1: kind of auction, so that you didn't sell it for
36:28
Speaker 1: too cheap. Right, it's fast, but you run the risk
36:31
Speaker 1: of selling it for too cheap. But if it's just
36:32
Speaker 1: commodity toolips, then you basically know the market clearing price. Anyways.
36:36
Speaker 3: Yeah, it's almost like a stock market for tulips. That
36:38
Speaker 3: I think is why it works so well. Whereas if
36:40
Speaker 3: it's a unique work of art, you need to give
36:42
Speaker 3: maximum information, maximum comfort to people, maybe a little bit
36:46
Speaker 3: of theater as well. We're not in a hurry to
36:48
Speaker 3: run this auction. The cost of the auction itself is
36:51
Speaker 3: trivial compared to the value of what's being sold.
36:54
Speaker 1: Now we go over to Agricola and have a descending action.
36:57
Speaker 1: It feels exciting, and it also feels like you're trying
37:00
Speaker 1: to guess what is in your opponent's head, right, because
37:03
Speaker 1: they're not revealing it as they would in ascending price auction,
37:06
Speaker 1: you have to think, oh, what are they willing.
37:08
Speaker 3: To pay for? Absolutely, and it works slightly differently, but
37:11
Speaker 3: the functionally what's going on is that every turn you
37:14
Speaker 3: take you get to grab some resource on the board,
37:17
Speaker 3: and then once you've grabbed the resource on the board,
37:19
Speaker 3: no one else is allowed to grab it until the
37:22
Speaker 3: next round. And there are several moves you can make
37:24
Speaker 3: in each turn, so you might be able to get
37:26
Speaker 3: two or three things each turn. And so you go
37:29
Speaker 3: backwards and forwards. It's your move and it's my move.
37:31
Speaker 3: And each time we want to grab something, and you
37:33
Speaker 3: might go, well do I want to grab it now?
37:34
Speaker 3: What do I want to wait until next turn where
37:37
Speaker 3: it might be better? And you're trying to figure out, like,
37:38
Speaker 3: what is the thing that the other person is desperate
37:41
Speaker 3: to have? Can I just let this pile get bigger
37:46
Speaker 3: because they can't afford to take it because they've got
37:48
Speaker 3: to use their move on something else. It's a very
37:50
Speaker 3: very clever game. I do enjoy it. You just also
37:53
Speaker 3: ask what non D and D games I am enjoying.
37:57
Speaker 3: I am enjoying a modified version of Blades in the Dark,
38:01
Speaker 3: Whi's a classic role playing game, very fast moving, you
38:04
Speaker 3: get to do Ocean's eleven or other kind of heists,
38:07
Speaker 3: and it's very modifiable. We're speaking on Monday. Just yesterday
38:12
Speaker 3: on Sunday, had a whole bunch of old friends around
38:15
Speaker 3: to my house and we just played this game all
38:18
Speaker 3: day and we had an absolutely terrific time.
38:21
Speaker 1: I feel like you're living the dream, Harfred makes me happy,
38:23
Speaker 1: living my best life.
38:25
Speaker 3: Jacob, this has been such fun. Thank you so much
38:27
Speaker 3: for doing this.
38:28
Speaker 1: Ah, it was a delight. I'll come back anytime.
38:31
Speaker 3: We would love to have you back. Thank you everybody
38:35
Speaker 3: who sent in a question. Sorry, we weren't able to
38:36
Speaker 3: answer all the questions, but we are going to be
38:38
Speaker 3: back with another Cautionary Questions episode later this year. Please
38:42
Speaker 3: do keep your queries coming, send them into tales at
38:45
Speaker 3: pushkin dot fm. That's t a l e s at
38:49
Speaker 3: Pushkin dot fm and I'll be back with another Cautionary
38:53
Speaker 3: Tale in two weeks time. Jacob Goldstein has a wonderful
38:56
Speaker 3: podcast called What's Your Problem? Thank You, Jacob, Thanks Jim
39:08
Speaker 3: Caul is written by me, Tim Harford with Andrew Wright.
39:12
Speaker 3: It's produced by Alice Fines with support from Marilyn Rust.
39:16
Speaker 3: The sound design and original music is the work of
39:18
Speaker 3: Pascal Wise. Sarah Nix edited the scripts. It features the
39:23
Speaker 3: voice talents of Ben Crowe, Melanie Guttridge, Stella Harford, Jemma
39:28
Speaker 3: Saunders and rufus Wright. The show also wouldn't have been
39:31
Speaker 3: possible without the work of Jacob Weisberg, Ryan Dilly, Gretta Cohne,
39:36
Speaker 3: Vital Millard, John Schnaz, Eric's handler, Carrie Brody, and Christina Sullivan.
39:42
Speaker 3: Cautionary Tales is a production of Pushkin Industries. It's recorded
39:47
Speaker 3: at Wardoor Studios in London by Tom Berry. If you
39:51
Speaker 3: like the show, please remember to share, rate and review,
39:56
Speaker 3: tell your friends and if you want to hear the
39:58
Speaker 3: show ad free, sign up for Pushkin Plus on the
40:01
Speaker 3: show page in Apple Podcasts or at pushkin dot fm,
40:06
Speaker 3: slash plus