The Joe Walker Podcast - How To Build A Future More Like Star Trek Than Terminator - Andrew Leigh

Episode Date: November 18, 2019

Dr Andrew Leigh MP is a member of the Australian federal parliament. He is currently Labor's Shadow Assistant...See omnystudio.com/listener for privacy information.

You're listening to the Jolly Swagman Podcast. Here's your host, Joe Walker. Hello there, ladies and gentlemen, boys and girls, swagmen and swagettes. Welcome back to the show. My guest is Dr. Andrew Lee, MP. Andrew is a member of the Australian Federal Parliament, where since 2010, he has represented the seat of FENNA, and he is currently Labor's Shadow Assistant Minister for Treasury and Charities. Before politics, Andrew was a professor of economics at ANU, a corporate lawyer, and an associate to Justice Michael Kirby when Michael was on the High Court. Andrew is the author of multiple books, including Disconnected, Battlers and Billionaires, The Economics of Just About Everything, The Luck of Politics, Choosing

Openness, Random Masters, and most recently, Innovation Plus Equality, co-authored with Joshua Ganz. I had the privilege of working for Andrew for about 10 months in various capacities while I was at university and it was a formative experience. I admire Andrew for his combination of intellectual fearlessness and a utilitarian compulsion to improve the world. I think this marks him out as a special parliamentarian. We had a great conversation about his new book. Without much further ado, I hope you enjoy. Andrew Lee, thank you for joining me. Absolute pleasure, Jack.

The only sitting politician that I allow on the podcast. It's a rare honor. Thank you. I thought we'd begin with a bit of a twist. You have a podcast called The Good Life Podcast, and at the end of every episode, you ask your guests the same six questions. I wanted to take this opportunity to turn the questions back on you, just as a way of sort of introducing you. I mean, I will already have introduced your CV, but you was a person. So the first question we'll start is what piece of advice would you give to your teenage self? Enjoy the moment. I think I was very concerned as a teenager to plan a career and to do things that would be good in the future and less focused on just having great fun as a teenager. I suspect I grew up a little bit too fast and was sort of

one of these kids who was an adult at age 10 and probably if I had my druthers again, I would play and explore a little more, try a few more pathways at that stage of my life. Why did you grow up so fast? Was it your personality or were there things in the environment? Yeah, standard type A personality, oldest child syndrome. Yeah. Next question. What's one thing you used to believe that you no longer do?

I think I used to believe that data and empirics would answer everything. And I'm increasingly drawn to notions of the importance of storytelling. And particularly, I guess this is characteristic of a shift from academia into politics. Academia particularly economics is very much about what evidence you can bring to bear and the rigor of that evidence. Politics is much more about persuasion and the art of storytelling storytelling and I'm drawn much more to the power of stories, not just to make a case, but also to enrich our lives. Great stories are a thing of beauty. And one of the real delights of podcasting has been finding so many good stories out there. Yeah. So stories as a means of persuasion and enriching our individual lives. What about stories as a means of persuasion and enriching our individual lives.

What about stories as a method of understanding the world? Yes. I mean, I think stories certainly unlock a lot. On my podcast, The Good Life, I was speaking this morning to a Zen Buddhist teacher, Frank Ostasecki, and Frank talked about moving towards things you fear. But he didn't talk about that directly. He talked about it through the analogy of putting up telephone poles, where apparently if a telephone pole is put in and it's looking a little wobbly, the best thing you can do if it starts to really wobble isn't to run away, because chances are it'll hit you in the back. It's to go right up to the telephone pole,

hold it firmly and stabilise it. And that analogy is so much more powerful than any intellectual argument you could make for moving towards your fears. Third question, when are you most happy? When I'm with my three little boys in the Canberra bush, which is out the back of our place. I'm in the bush every day. As a marathon runner, I run about an hour a day.

And to have them with me in the bush is just magnificent. We actually took a tour on the weekend through our local bush with an Indigenous guide, a bloke by the name of Tyrone Bell, and that again just opened all of our eyes to what's interesting in the bush, the different plants. Tyrone showed us some of the markings on the trees made by Indigenous peoples in the past that points in different directions. There's something deeply grounding about being in nature. You seem like you'd be a very good father.

Are there any... I wish. Do you remember any lessons you've tried to consciously impart to your boys recently in the last, say, one or two years? The hardest thing I find as a dad is building resilience because obviously you can try throwing them in the deep end, but if that goes badly, then you haven't built resilience, you've built fragility. So that notion of how to create, in Nassim Taleb's phrase, anti-fragility. So that's the thing I'm constantly struggling with, how to push my boys a little outside their comfort zones but not so far outside that they then retreat back again.

I don't think I've gotten it right. Perhaps when the last of them leaves home, I'll feel as though I've mastered fatherhood. In a world of uncertainty, that seems like it'll always be a judgment call. Yes, exactly. And I make so many mistakes as a dad. Like I'm just every day saying things, doing things where, you know,

I put my head on the pillow at the end of the night and think, oh, I could have done that better. But I'm a more conscious dad, I think, than I was 12 years ago when I started the job. So, you know, like any job, you get better with experience. The challenge is getting better fast enough to have a good career. Fourth, the Good Life podcast question. What's the most important thing you do to remain physically and mentally healthy?

Oh, got to be running. Yeah, I could have answered that for you. Do you have any guilty pleasures? I have been going overseas to run the Indigenous Marathon, sorry, the World Marathon majors over the last couple of years. There's six of them, Boston, Berlin, Chicago, London, Tokyo, New York. So not everyone would think of running a marathon as being a guilty pleasure, but dashing overseas on a frequent flyer flight

for a weekend to run a marathon is pretty fun for me. Have you done the New York yet? I have, yes. I've done all six of them, all wearing the Indigenous Marathon Project supporter singlets. So Rob DiCostello got me involved in that and being part of that community and that network brings a special pleasure to running. I did the New York two years ago with a mate as part of the Aussie contingent for Movember. And coming from Queens back into Second Avenue, you feel like an Olympian.

The crowd is just roaring, handing you bananas or Gatorade or whatever and it's such a special event. It's astonishingly good. And what's amazing about Rob is he makes it the capstone event for the Indigenous Marathon Project, despite the fact that New York robbed him of ever being a sitting world champion for the marathon. So as we know, Rob held the marathon world record for a couple of years in the early 1980s. But that was only recognised after the fact, because New York had been setting a course which was slightly too short, and so the world record time which had been run in New York was regarded by most people as being the world record.

As a result, Rob didn't get the appearance fees that he should have gotten for being the world champion because the New Yorkers had the course too short. So what an act of grace to then every year take a dozen Indigenous runners to the New York marathon, the very marathon that dudded Rob back in the early 1980s. Final The Good Life podcast question. Which person or experience has most shaped your view of living an ethical life? Oh, my parents.

They're great. Great. Just such relaxed mum and dad and that real ability to love fully, to encourage, to be curious and to guide by asking questions rather than by telling. I don't ever remember my parents shouting at me, raising their voices in the household. And the opportunities they gave my brother Tim and I in terms of travel and the rest were astonishing. So I suppose my parenting is to a large extent just an attempt to live up to what my mum and dad gave me. Let me add two of my own questions. Please.

The first is sort of a bastardization of peter teal's famous contrarian truth question so what's what's one thing you hold to be true that very few other people believe uh this is a tough one for a for a sitting politician yeah Yeah, yeah. I'm just less worried about the politics of it than just thinking through. I mean, for all that I've said about the power of storytelling to persuade, I think I do have a pretty strong scepticism of the efficacy of any given social program. So if you describe to me a government program that sounds good, my intuition will be to say that sounds great, but it probably

won't work. And this is known as Rossi's Law after the sociologist who first coined it. And one of my heroes, the late great US Senator Daniel Patrick Moynihan, used to adhere to it as well. But I don't think most people in public life or the general public do. I think there's a tendency to believe that most programs are efficacious. My reading of the evidence is that while I would love it to be so, I don't believe it is, which is why I'm a passionate advocate for more randomized trials in social policy. That's right. The wicked thing is that you never truly know whether or not a social program is efficacious if you don't have a true counterfactual. Precisely, yes. And the story of

medicine should give us pause on this. The majority of drugs that look great when they come out of the lab do not pass through stage one, two, three clinical trials. They fail at some point along that way. And I'm not sure that our social policy labs are any better than the cancer drug labs in producing effective ideas. So test, learn, adapt ought to be the mantra for how we do a whole lot more government policy. Or if you need another reminder, the medical journal, The Lancet is named after bloodletting. Exactly, exactly. A practice discontinued thanks to a random randomized trial just a couple of decades after the Lancet began publication. So the day after the 2019 federal election, I sent you a message of commiserations. I was feeling a bit down that day. I wanted Labor to get up. I prefer Labor's policy

agenda. I can only imagine how you would have been feeling given you were part of that team. You strike me as a very naturally optimistic person. Do you ever have dark moments and what do you do to stay strong? That was certainly a dark moment. I think not only because I wanted our ideas to get up, but also because I think it posed a challenge to anyone who believes deeply in oppositions taking a serious policy agenda to the election. And in the same way that John Hewson's loss in 1993 led to small target politics for a decade, my fear is that some will be urging both sides of politics not to take serious agendas to the next election. That, I think, would be a profound disservice to the Australian people. I think elections should be a battle of ideas

and governments should be ready to govern if they win. So the work that we had put into developing a carefully costed, well thought out policy agenda was what made me most disappointed. I'm ambitious for Labor, but I'm even more ambitious for the country. And so I think it's vital that Australia has an ambitious government. You look around the world right now and the lack of leadership from the great powers on big global issues creates an opportunity for a middle power like Australia to play an outsized role. But I don't see us doing that at the moment. And I think that's a real pity. Sorry, you asked me about staying, about positivity. The biggest silver lining to me of the election loss is that I've watched many ministers and parliamentary secretaries and seen

what that does to their family lives. Rule of thumb, you become a parliamentary secretary, you lose one of your weekend days, you become a minister, you lose both of your weekend days. So for the next three years, I will get to have two weekend days every week with the three little people who are most important to me through some really important years of their upbringing. They're now seven, 10 and 12. So that's pretty amazing to have. And one of the things that I guess keeps me grounded when I think, well, it would have been nice to be implementing a positive progressive policy agenda right now as part of a Labor government. Hmm. My final general getting to know you question. You studied arts law undergraduate

at Sydney Uni and then later you did your PhD at the Harvard Kennedy School. I once heard you give a speech while I was at uni which kind of changed the course of my life, which is that you would rather take the incentives-based framework of an economist than the rights-based framework of a lawyer as a means of affecting social change. It reminds me of a quip by Paul Samuelson, who I think he once said something like, I don't care what the laws or the constitution of a nation are if I can write its economics textbook. But if you could go back in time, or if you could give some more advice to your teenage self, would you have preferred to study a quantitative degree undergraduate?

Probably, yes. So my advice to young people when they're choosing courses is twofold. Choose professors rather than course descriptions because a bad professor can write an enticing course description. So talk to people and find out what the reputations are of good professors and go and study with them even if their course descriptions don't immediately grab you. Secondly, don't study things where you would otherwise read the course books for leisure on the weekend.

So if you're an American politics buff, for goodness sake, don't go and study a course about the American presidency. Just go and sit there on weekends and read Gore Vidal's Lincoln and Remnick's Obama biography, and that will give you enough of a flavour of things. But very few of us wake up on Saturday morning and think what I really need to do is open Maskell L. Winston and Green's

microeconomics textbook or wouldn't it be great if I was getting into learning some more matrix algebra and so I could really nail a particular econometric function. Those skills though are enormously important. One of the reasons that economics has made huge inroads in understanding a whole range of social science questions is its high degree of technical rigour. So acquiring that, I think, is pretty important for anyone who wants to make a difference as a social scientist.

About four years ago, late 2015, maybe early 2016, I'd started thinking a lot more about automation and the future of the workforce. I was working part-time and casually in your office at the time. So, of course, I was thinking about these ideas because of the heady intellectual atmosphere. And it struck me that a lot of people say that to worry about automation or AI replacing jobs is just a modern version of the Luddite fallacy. And at every step along the way, people have resisted technological change, but technology has created new unexpected jobs and the world moves on more productively than ever before. But I was thinking to myself whether maybe this time really is different and there's something so qualitatively different about the coming technology and AI and automation that might mean that we see a different response in the workforce. And I remember asking you, you know, the most extreme example of this is the singularity where we reached this point where technological change

becomes, you know, explosive and recursive and human life is changed forever. And I remember asking you, you know, what happens if we reach the singularity? And you said to me, at that point, all we worry about is redistribution of wealth. A lot of people have this anxiety. I recently had Bob Schiller on the podcast. He just has a book out called Narrative Economics. A great episode if anyone hasn't listened to it. Thank you. And he argues that one of the things that worsened the Great Depression were these narratives about technological unemployment. That was actually the phrase that people used back during the 1930s. And that if this narrative ever rears its head again, we could fall into a deep recession.

So there seems to be this anxiety that technology promises this future that we want to be a part of, but we're not sure whether we'll get the benefits of that future. You have a book just out with Josh Gans titled Innovation Plus Equality. That sounds a lot like a recipe for a future that I want to be a part of, but is it a utopian vision? We don't think so. We think it's a realistic vision. And our vision is, to be clear, one that has uncertainty at its heart. So the economist Frank Knight famously distinguishes between risk, which are things about the future that we don't know, but we can parameterise. So for example, you might think about a bag of apples, which is

half rotten apples, half crisp apples. When you put your hand into the bag, you don't know whether you're going to get a rotten one or a crisp one, but you know that it's 50-50 rotten crisp. And he distinguishes that from uncertainty, which is where you know the bag has some rotten and some crisp apples, but you don't know the percentages. It might be 90% rotten, might be 90% crisp. You simply don't know. We think policy needs to take into account uncertainty, and one of those uncertainties is we're simply not sure as to whether the future is going to be hunky-dory on the jobs front or might generate the concerns raised by your singularity example

in which suddenly artificial intelligence is able to do everything better than humans currently are. You have various indicators pointing either way. There's a lovely study of the impact of ATMs on bank teller jobs in the United States, which finds that, in fact, banking jobs don't disappear after ATMs. The jobs simply go into serving customers in other ways. But then you have the example of AlphaGo, where now Google's Go playing machine is so much better than the world's best human player that the gap between the world's best human player and AlphaGo is as large as the gap between the best player and an amateur player. So there's certain areas in which computers look on, if they were self-aware, would look on humans' ability much the same as we look on the intellectual powers of our pets.

So we do need to recognise that the future may have an optimistic scenario for jobs, which is what we've seen right through the 20th century, contrary to predictions of people like Keynes and Leontief, that there would be mass unemployment resulting from technological change, but also the possibility that perhaps some of these technologies really do take off and take off more rapidly than other industries are able to generate jobs. And so Joshua and I argue that it's important to have policy settings that are a form of social insurance that allow society to be ready for either of these possible futures. Let's go back to knighting uncertainty. So this is one of two ideas which pervade the entire book. Have you heard Keynes' articulation of the same idea?

No. How does Keynes put it? So this is from 1937 in a response to critics of the general theory. And the quote is something like, when I speak about uncertainty, I'm not referring to the sort of uncertainty you see in a game of roulette. I'm referring to the uncertainty in answer to the question, what is the probability of European war? Or what will the price of copper or interest rates be 20 years hence? About these matters. What a 1937 question. Exactly. And he goes on, about these matters, there's absolutely no scientific basis on which to form any calculable probability, whatever. We simply do not know. So that was 1937. But I think he actually, that quote was from 1937,

but I think he originally discussed the idea in his treatise on probability in 1921. So he, because Knight's book, Risk, Uncertainty and Profit was published in 1921. I think they independently came on this idea. I was so happy to see it in your book because it's an idea that economists have sort of forgotten in the last century. Well, it is striking when we look at job loss that people put a lot of faith in this Oxford study, which says that 47% of jobs can be computerized and fail to kind of look at the fine print where the authors start off by hand coding occupations into will be computerised and won't be computerised. In other words, they put 100% probability on some jobs being computerised and a 0% probability on others. And jobs like marketing and accountancy are in the will be computerised category, despite the fact they've enjoyed pretty rapid growth over recent years.

And then they take that hand coding, use an algorithm to apply what they've done to 70 jobs to 700 jobs, and publish the whole table. I don't share any of that confidence that we can put a precise percentage on job loss. I think trade and technology tend to have discontinuous effects. You look at the impact of China's entering the WTO at the start of the millennium on the US manufacturing sector, significantly larger than people anticipated. Conversely, anticipated job loss caused by driverless cars has been slower than we expected. You've got driverless trucks that BHP and Rio are driving around mining sites in the Pilbara. But most experiments with driverless cars right now still have a safety driver sitting in the passenger seat, which means there is still exactly as many jobs. Yeah. So let's dig a little deeper on why the Oxford study, for example, is a futile exercise. Why is the future radically uncertain? Because that's the way in which job change and technological change tends to work,

in jumps and uncertain leaps. And so if you're a young person planning for that, you don't want to take the example that is given to Benjamin Braddock, the star of The Graduate, just go into plastics. Plastics is the future. I've got one word for you, plastics. But instead, you want to look at a different Ben, Benjamin Franklin. Benjamin Franklin was not only one of the great polymaths of America and the founding fathers, but also somebody who placed a radical emphasis on self-improvement and on constantly learning new skills. So you want to have, in a world in which the labour market may change on you, you want to have a diversity of skills and a taste for lifelong learning. And we as a society want to engender institutions like MOOCs to ensure that people are able to continue learning.

Right now, for example, we don't have a lot of accreditation of MOOC courses. There's not a lot of portability. There's not a lot of recognition by traditional higher education institutions. But if we want people to continue lifelong learning, then better incorporation of MOOCs into our education system would be a natural way to go. Because MOOCs bring down the cost of training very substantially. They're going to be an important part of a world in which we've got a bit of uncertainty about the future of jobs. So have anti-fragile career capital. Yes, exactly. Yeah. And so Joshua and I talk in the book about different models for vocational training. People typically turn to the German model of very specific training.

I, when I was in Berlin, met with the Confederation of German Industry and talked to them about their 200 Handwerk categories in which people are trained very precisely for a narrow technical occupation. Turns out that's terrific if your aim is to ensure that you have very low rates of youth unemployment and good labour market outcomes for people in their 20s and 30s. But then when economists like Ludger Wussmann have looked through what happens to people in their 40s and 50s, it turns out that more general technical training is more effective. The Swiss have a model which Joshua and I are more drawn to. The Swiss place more emphasis on broad

skills, which allow people to move as the labour market shifts. So whether you're talking about the value of having a classics base in a university education or a broad base in a vocational training setting, again, as you say, you want to be anti-fragile in your training process. Just to come back to this question of why is the future radically uncertain, have you read Karl Popper's book, The Poverty of Historicism? I've read about it, but I haven't actually read it. What would I learn if I did? So all you need to read is the preface. And he has this lovely passage of a priori reasoning where he explains why the future is radically uncertain. And it basically goes like this. The course of history is in large part determined by new technology we can't imagine the technology of the future since if we could we would already

have invented it therefore the future is fundamentally unpredictable right yeah no i love that i think it's uh it neatly encapsulates the sort of innovation that Joshua and I think we need and the way in which we ought to deal with it. And even when these things are invented, we're not very good at figuring out how they're going to change the world. So the iPhone comes along and it's referred to by one critic as a not very good email machine. People are scathing about the fact that unlike the BlackBerry, which dominates the smartphone market at the time, it doesn't even have a keyboard. Fast forward, BlackBerry's lost nine-tenths of its market cap and the iPhone's dominated the world's smartphone market.

At about the same time, the Segway comes out. This stand-up scooter with gyroscope, which people say will revolutionise personal transportation, will change cities, everyone will get around on Segways because they're so easy to manoeuvre on. And we fast forward a decade and Segway is essentially nowhere. It's used by tourists and a few cops. And it's not even obvious to brilliant people at the moment after those things have been invented what impact they're going to have, let alone the inability of most of us to think of inventing

such a thing in the first place. And the irony in those two stories is didn't Steve Jobs, the guy who invented the iPhone, think that the Segway would change the world? Precisely. Exactly. So even Jobs couldn't forecast the way in which the Segway would have an impact. And it's not even obvious why it didn't. You know, it does what it says on the box. We speculate, Joshua and I speculate that perhaps it's just that people thought it was a bit weird and felt a bit strange standing up there. But it's not like they're dangerous. I mean, police officers use them.

You wouldn't put newbie tourists on them if they kept on killing people. But it's just a thing that didn't take off the way in which we anticipated. And, you know, you look back at the forecasts, the forecast of the impact on cities and on jobs as drivers are as radical as any of the forecasts that people are making for driverless cars these days. I remember the first time I visited San Francisco in 2016 and I had this image built up in my head of this fantastic,

futuristic place. I was very interested in tech and entrepreneurship. And we first got off the bus from LA and stepped onto the pavement and a guy came past on a handleless Segway. I just thought, this is so cool. I've arrived. Yeah. A lot less daggy than the very tall ones though. Yes. They just didn't seem to interact with the market. Yeah. No, that's totally right. I mean, there's a guy who lives near me who commutes on one of these sort of electric skateboard type things, which, again, are just operated by force forwards and backwards.

They look very cool, but they just haven't taken off. Yeah. So some people have a very grim Malthusian vision of the future, and they think that we've picked a lot of the low-hanging fruits when it comes to new technology and that productivity growth is going to continue to flatten. Other people are what you and Joshua describe as tech optimists. Which camp do you broadly belong to? We're tech agnostics. We can see a lot to be said for the tech optimist view. That view is essentially the view that innovation is driven by basic science. So if you look at the history of Galileo's discoveries in astronomy, a lot of

that comes from the quality of glassmaking, which led to the telescopes that he was able to use. And so if you look at what the fundamental technologies of AI and CRISPR can do, for example, you might well have an optimistic view as to where technology is going to go. And the tech pessimists take the view that we're promised flying cars and we've got 140 characters. This is the slogan of Founders Fund, Peter Thiel's. Exactly, exactly. I'm stealing Peter Thiel's line here.

And that the astonishing innovations of the 20th century in terms of everything from indoor toilets to cars to planes to penicillin were far more life-changing than what we've seen over the course of the last 40, 50 years. That since the early 1970s, we've had a productivity slowdown and that has been due in large part to a slowdown in innovation. There's challenges in measuring productivity well, which I think makes this even tougher. We know, for example, that there's been huge social welfare benefits that have come from having technologies like Gmail and Google Maps

and a range of free online products. You look at all the apps on your smartphone, which many of you listeners are probably using now to listen to this podcast. Those things aren't well captured in GDP. So perhaps we're missing some of the productivity, which makes it even harder to work out whether to go with the tech optimists or the tech pessimists. Again, we think because you're not sure, you want to embrace uncertainty. How do you do that through an insurance-based approach? Although you say

you're agnostic, I feel like you lean slightly towards the optimist camp. I think to be a good politician, Joe, you probably need to be a bit of a congenital optimist, much as in the way that if you want to be a good journalist, you probably need to be a little cynical. Well, I mean, in support of the optimists, tell us this idea of a combinatorial explosion. So this is the notion that if you look at the elements that have been discovered and you start multiplying together the possibilities of those elements, you end up with more possibilities than there are grains of sand on all the beaches and all the world,

and that we've only just begun to look at those combinations. And you can tell a similar story around gene editing, for example, and the possibilities that we might find if we look at different gene combinations. So there's a huge amount out there to be discovered. And as computing power goes up, our ability to mine those intellectual shoals is pretty significant. In the book, you discuss an example of Paul Romas where he uses the periodic table and assume you have 100 elements. You can kind of get a sense of how exponential the combinatorial explosion really is. If you want to mix just two elements together, that's 100 multiplied by 99. And then if you want to mix three elements together, that's 100 multiplied by 99 multiplied by 98, and so on. And it gets out of hand pretty quickly, doesn't it? Yeah. So I think as he puts it, once you get up to 10 elements,

there's more recipes than there are seconds since the Big Bang created the universe. Wow, yeah. And then, you know, potentially the opportunities are limitless. That's a pretty exciting notion and certainly suggests that there's reasons to be optimistic for the future. Yeah. And also reasons to ensure that we have the policy settings right to encourage innovation. So we're very big on the idea that you need to have a patent system that doesn't lock up innovation for too long, that you might

even think about dual-length patents, a model of short patents which have a lower novelty bar and longer patents, 20-year patents that have a higher novelty bar so you don't get Amazon's one-click ordering locked up for 20 years. We also think that there's a value to having competing research bodies, because the more uncertainty there is in the innovation process, the more value there is to having a couple of different research funding bodies competing to try and create slightly different incentives for scientists and innovators. Yeah, let's talk about the patents for a moment, because I think this is really interesting. In the book, you cite research by Alex Tabarrok, who talks about a hump-shaped curve when it

comes to the impact of patents on innovation. Can you explain that? Well, Alex's notion is quite simple. If you have no patent system, then your incentive to create and publicise a new invention is quite low. We talk in the book about the impact of the introduction of forceps, a technology which saves lives, will save lives today. There will be babies born today and mothers that live today thanks to the invention of forceps. But it stayed locked up for decades because the family that invented forceps couldn't patent them at the time. So clearly you want a patenting system to allow those ideas to be brought out

in the public domain and for the creators to enjoy some period of monopoly rents as a reward for their invention. But go too far, lock up things for excessive periods and follow on innovation becomes very tricky. Things like the evergreening of pharmaceutical patents, I think, has damaged innovation in those contexts. It's important that people are able to build on the ideas of others. So 20-year patents, I think, are appropriate. And I like Alex's idea that perhaps you might have an eight or a ten year patent where you haven't managed to clear quite as high a bar for novelty and inventive step. So it's kind of like a technological version of the Laffer curve.

Precisely yes yes and he contends that in some areas, again, taking the example of Amazon's 20-year monopoly that it got over one-click ordering, that that might be something that's on the wrong side of the hump. We're allowing them to lock up that pretty straightforward business process patent for two decades, impeded innovation rather than enhancing it. Now, one of the mysteries of Western economic growth is that beginning around 1974, productivity growth started to flatten out. You mentioned the slogan of Peter Thiel's Founders Fund that we were promised flying cars and instead we got 142 characters. Of course, the irony of that motto is that Peter thinks that Twitter is actually pretty good and that flying cars are just totally unfeasible in practice. But now we get 280 characters. But you get the basic idea, which is that there seems to be so much, you know, the iPhone 11's just come out.

There's a lot of technological change, yet at the same time, it doesn't feel like our lives have become materially better than our parents or our grandparents, kind of the progress they experienced in the course of their lives. What do you think? We've mentioned a few things, but what do you think broke in the 1970s? It's not obvious to me from either introspection or reading the literature why you see the turning point at that stage. I mean, obviously you had stagflation and the challenge that that posed to policy makers, sort of, In economic terms, the 70s are a much darker period than the booming 1960s, where you see the sense of possibility and promise encapsulated not only in social movements

for racial and gender equality, but also in the sense that you can go to the moon. And then when you move from the Beatles decade to the decade of the Doors, you get a sort of a darker sense about that. One notion is just that as you get close to the frontier, you need to throw more and more person power at it to keep on moving. So we have continued to see Moore's law in effect, the doubling of computing power every 18 months. But if you look at the number of people in research labs that are required to continue to get that doubling every 18 months,

it's increasing massively. Right, so diminishing returns. Precisely, yeah. You have to have an awful lot of scientists because the problems are now much more complicated than they were. The heat build-up in these tiny little slices of silicon is a much more substantial problem than it was when you were in the early stages of doubling.

So, you know, I feel a bit like that Robert Barrow line that every discussion of productivity ultimately devolves into amateur sociology. But I do think there's some combination of the zeitgeist and the diminishing returns point that you mentioned before until you move into a new frontier where you see the age of innovators fall and the gains much more rapid in the early years. Hmm. Frank Knight wrote in his 1921 book that uncertainty renders entrepreneurship inherently tragic. The other big idea that runs through your book is Joseph Schumpeter's Creative Destruction, which is also a poignant idea.

Tell us what Creative Destruction is and how should policymakers handle it? So Creative Destruction is the notion that new firms contain the seeds of demise for many of the existing firms. Those of us who care about workers undergoing painful transition are naturally worried by the destruction aspect of creative destruction. But it's very much there. You can't create new ways of doing things without putting pressure on the jobs of

some people. The challenge is how to make sure that we have those transitions happening in a smooth way. And I think about, Jo, the transition from transportation by horse to transportation by automobile. There is a massive reduction in the number of horses used for transport in the United States. Many horses are not bred. Others are sent off to the knackery prematurely. But there is not an outbreak of unemployment as a result. Those who have been working as farriers and driving horses, they manage to re-skill and the transition is gradual enough

that they're able to find new opportunities. So the challenge for policy makers is to ensure that when a new technology comes along, that the impact on the workforce is more like what happened to the farriers and less like what happened to the horses. I want to ask you a couple of speculative questions. And the first is, so people like Elon Musk, Sam Harris, Nick Bostrom

are very famous for worrying about artificial general intelligence. I've heard Marc Andreessen describe this as a modern version of the Promethean fallacy, which is closely linked to the Luddite fallacy we spoke about earlier. And in the myth of Prometheus, Prometheus was a titan who stole the secrets of fire from the gods and gave them to mortals, and then was condemned to be strapped to a rock and every day have his liver pecked out by a crow. It would regrow overnight, or by an eagle, and the eagle would come back the next day. And there seems to be, you know, Mark Andreessen,

founder of Netscape and Mosaic and now a venture capitalist, argues that this is a deep-seated psychological fear in humanity that will, in our striving, overreach and create the thing that destroys us. He thinks there's no cause for concern when it comes to artificial intelligence. It's the latest manifestation of the Promethean fallacy. My concern, again, like with its impact on jobs, is that maybe this time really is different because we're dealing with a technology that is transformationally and qualitatively different to anything we've seen before. Where do you fall in that debate? Do you have concerns about artificial general intelligence? I think there's a set of tail risks to which we're not paying

sufficient attention, risks that ultimately could destroy humanity. And therefore, even if they're of, they have quite low probabilities attached to them, they're worth investing money in reducing those probabilities simply because the consequences are so large. And the consequences not just for us, but for the accumulated future generations. If you take the approach that the lives of your children are as valuable as your own life, then effectively you're applying a zero discount rate to the utility of those in the future. That, I think, is the right approach. So therefore we ought to be concerned about a whole set of existential risks.

So I like the fact that there are groups, particularly at Oxford and Cambridge, who are thinking about these sorts of problems, but also increasingly they're becoming mainstream. Asteroid strikes, I think we've actually managed to do a reasonably good job of tracking asteroids. Nuclear war is one that we could probably invest more resources in, particularly securing loose nukes. Bioterrorism is worth investing a little more in curtailing those risks. And the risk of the singularity,

while I would say it is as improbable, has such large costs that it's worth having these sensible conversations now about ethical safeguards around artificial intelligence. Would you recommend any books for the average person interested in that topic? Well, since we're podcasting, let me recommend Rob Wiblin's 80,000 Hours podcast, which has extraordinarily long, thoughtful, and in-depth conversations

with a range of people thinking about existential risk. He's been a guest on both our podcasts. Absolutely. Yeah. Let me ask another speculative question. So national housing bubbles are a creature of the post-war era, really from the 1960s onwards. And I suspect that's for a couple of reasons.

One is financial liberalisation. The second is that National House Prices indices only began to be published around the 1970s. We actually didn't have any indices before that, which enabled the media to report on prices and attract speculative attention. I wonder whether it's mere coincidence that that period has overlapped with secular stagnation and whether there might be some causality in either direction. I have to obviously bring up housing bubbles in every episode I do, but have you considered this idea? Yeah, you do worry that overinvestment in property speculation can't be good for creating more innovation in Australia. That placing more resources in the hands of optimistic entrepreneurs is probably more use than a society which says the way to get rich is to buy an investment property whose value you think will increase. And you worry too when there's not only money but smarts going into that where you

see too many bright people going into housing speculation rather than startups. Or construction. Well, you do want to make sure that you've got a housing stock which is facilitating the productivity which comes out of cities. Sure. When I wrote Economics of Just About Everything, I did a little analysis looking at the impact on productivity of moving from a regional area to a city and estimated that it was about 30%.

And Ed Glazer's research would suggest that that's mostly causal. So cities have been one of the big drivers of Australian productivity and building the appropriate housing stock in cities is I think a good way of boosting productivity in Australia. So I'm more concerned about speculation on existing homes than construction of what I would regard as needed new homes, especially in a country which has had pretty rapid population growth compared to other advanced countries. You don't think there's been over-construction in the last five to ten years? I don't see it no I think there's appropriate appropriately rapid increases in the

housing stock and now you we're missing that kind of mid-level density that you see especially in Paris but also in a lot of other European cities those kind of four to six story walk-ups we tend to have a mix of single family homes and apartment blocks and getting more of that that that comfortable medium density housing is not only a good way of ensuring that you're able to have people having relatively short commutes to work, but also ensures that communities can be pretty liveable, that local shops and cafes and parks can be appropriately funded by the property taxes of those living in proximity to them.

So as a social capitalist, as well as a regular capitalist, I like that medium density model. How confident are you that the West will solve its productivity growth problem in the next, say, five to ten years? I'm pessimistic about five to ten. I'm optimistic about 15 to 20. The current populist emergence, I think, is distracting many policymakers from the core productivity challenges, which I would put into three categories. I think they're around institutions, particularly improving the quality of competition, improving competition and ensuring that you don't get an excess of monopolies and making sure the tax system is appropriate for a world of weightless production. Infrastructure, where you want to have more public investment

in public transport and high-speed broadband, and individuals, where I think the quality of human capital has declined over the course of the last two decades. On the OECD's PISA tests, Australia has gone backwards in every subject. Every year we've been tested to the extent that our year nines are now well behind where their counterparts were at the turn of the century.

So improving the quality of vocational training, high school training, universities is pretty fundamental to boosting productivity. All of that is doable. The challenges I've just outlined, I think, are common to many advanced countries, not just Australia, but I don't see an ambition to solve them in many countries right now. What are you going to be doing in the Australian Parliament to make that happen? Well, sitting on the wrong side of the speaker, there are certain limits as to what one can do. But the right role of a professor turned politician is to be out there in the ideas debate. And increasingly in the nine years I've been in politics, I've found myself

reverting to type as an economist, being aware that it is important to be part of the transmission belt from academia into politics. Politics is the confluence of power and ideas. I'm way down the ideas end of the spectrum. And so I want to be spending as much time as I can talking to bright researchers and bright thinkers about what ought to be done, crystallising that into books like Innovation Plus Equality and engaging in conversation on a splendid podcast like yours. Andrew Lee, I've said it before and I'll say it again. I'm very glad that we have somebody like you in Parliament. Thank you so much for your new book and for your time. Absolute pleasure, Joe. Thank you.

Thank you so much for listening. I hope you enjoyed that conversation as much as I did. For show notes and links to everything we discussed, you can find those on my website, www.josephnoelwalker.com. That's my full name, J-O-S-E-P-H-N-O-E-L-W-A-L-K-E-R.com. And you can find me on Twitter. My handle there is at josephnwalker. Until next time, thank you so much for listening. Ciao.