Koen Smeets: Welcome to the interview series on the socioeconomic consequences of disruptive technologies by Rethinking Economics NL. In this first interview we will examine the topic from a more general perspective, for which I am deeply honoured to be able to introduce two of the most influential economic experts in the world. Firstly with us today is Daron Acemoğlu. He is a Institute Professor at the Massachusetts Institute of Technology, MIT. He's one of the foremost living economist and also one of the most cited. His work focuses on the role of institutions in political and economic growth, combining history with econometrics.
In 2005 he received the John Bates Clark Medal, awarded for senior economist under the age of 40 judged to have made the most significant contribution to economic thought and knowledge. He has written several books including "Why Nations Fail" and "The Narrow Corridor", some of my favorites. He has also written several articles on the economic impact of disruptive technologies as AI and robotics, related to which we'll be focusing today. Also with us today is Martin Wolf. He's the Associate Editor and Chief Economics Commentator of the Financial Times. Known for his independent, unique and insightful perspectives, he is one of the most influential financial journalists in the world. Related to this he was
awarded the CBE, Commander of the British Empire, in 2000 for services to financial journalism. He has also been critical of the current neoclassical and mathematical focus of the economics education. He is the author of several books including "The Shifts and the Shocks", here, "Fixing Global Finance", and "Why Globalization Works". Having similarly an incredibly interesting perspective on the current and future socioeconomic consequences of disruptive technologies, this should make for a very interesting interview. First question of today is for Professor Acemoğlu. Considering your extensive research on how disruptive technologies are affecting the economy, could you tell us more about the past, present and future consequences of this phenomenon? Daron Acemoğlu: Thank you Koen and thanks for organizing it, it's a great pleasure to be here especially with Martin, who is also my favourite economics journalist, so it's a singular pleasure and the topic is really close to my heart so I'm really delighted to be part of it. Look, I think
economic historians have done an amazing job of documenting that almost every technology you can imagine is disruptive. You know, disruption hasn't been invented in the Silicon Valley and at some level there isn't anything qualitatively different about robots and numerically controlled machines and even AI, compared to the disruptive technologies of spinning and weaving that's launched the British Industrial Revolution, mechanization of agriculture, machine tools in manufacturing. And they are having very similar effects, and that's where I depart from many other economists and some commentators, that those effects are always disruptive. If you are a spinner at the end of the
18th-century things are not bright for you. If you're a telephone switch operator as the automatic switches are coming in, in the middle of the 20th-century things are not bright for you. That's not different from the situation that a welder is facing when robots are coming in. What has changed, in my opinion, and what is qualitatively different today, is that in the past, somehow, we managed to have a broad portfolio of technologies that as we were disrupting and destroying jobs we were also creating jobs via other technologies, new tasks, new industries. That's the story of mechanization of agriculture in the US and to a large degree in the UK as well.
New jobs didn't come from agriculture, the share of agriculture in employment went from you know over 50 percent to less than 10 percent in the course of four decades, but we generated a lot of jobs via other technologies at the same time and that's what we've stopped doing. We have put somehow all of our eggs into algorithmic and robotic automation and we're not doing enough to generate jobs, especially good jobs for people with diverse backgrounds and different sets of skills today, that's where the future really hangs on. Because if we keep on doing this, we're going to increase inequality, we're going to stop, we're going to continue to have a pitiful growth for average and median wages, and we're going to have lots of social costs come from a growth model based on only empowering a small fraction of the population. Koen Smeets: What I was curious about is that a lot of economists seem today, of your colleagues, seem to think that this is just another technological change, that in the end everything will be fine. There seems to be a lot of division in the economics field on the consequences of these technologies. What I sometimes like to say, is it almost seems that
everyone is really, really worried about these socioeconomic consequences of these technologies, except for most economists. How do you see this? Daron Acemoğlu: Yeah, I think that's, I think that, you know, first of all I think economics as always is a broad tent and there are many many different views. You know what you characterize is not the view of many economists I work with and I interact with, but it is true that it's it's probably the modal view among economists. And I think it's based on a double misunderstanding. The first misunderstanding is theoretical. Somehow many economists have grown up with models, and I grew up with those models and I included them in my [inaudible], the main models of production function, where technology just augments and increases the productivity of labour. And if you start with this perspective and if you impose this perspective on the data, any productivity improvement will ultimately flow to labour indirectly, or sometimes perhaps directly.
But that's not the only way we can conceptualize technology and as I have loosely argued a second ago, a lot of technologies are of this replacing form that displaced workers from the tasks that they were performing. And the second misunderstanding is, you know how really the adjustment took place. It's a subtle misunderstanding in my opinion, but it's a critical one. So, one way you can think about it is to say,
oh you know new technologies come in, they are disruptive, but automatically they will self-correct. And that I think is the view that many economists gravitate towards, partly because of the conceptual framework that they bring to the table, and partly because I think because they haven't really studied the history of these adjustments as much as they should have. Whereas the perspective that I am bringing is, there's a disruption, if you do the right things, both institutionally and technologically, you can rebound and create shared prosperity out of it, but there's no necessity, there's no automatic process that will, it will take work it, it will take work from entrepreneurs, from innovators and from politicians to build the institutions to support it. And I think that's where we are failing and that's why the automatic one will not happen, in my opinion, unless we take drastic action. Koen Smeets: And on the other hand, we also see technologists that sometimes are really, really optimistic on the way that AI and robotics will develop and change society, they really think that, starting with society but then also our economy will fundamentally change. How do you, why do you think that their opinion often diverts such an extent from some of the leading economists? Daron Acemoğlu: Well on this one, you know I think there is a lot of uncertainty, there are many data scientists, AI scientists, computer scientists who are also incredibly optimistic about what AI and future technologies are going to do, you know there are scores of books on singularity, great abundance, incredible richness, superintelligence, and you know, I mean I don't rule out that we may at some point come to superintelligent machines, but certainly, if you think of the current technologies, they would be much more in the in the category of artificial dumbness rather than artificial intelligence, I mean they do a certain number of things well, but they're not intelligent and they're not amazing, the productivity gains that they bring are very, very small. I mean, you know, imagine automated
checkout kiosks. I mean nobody would think, would write a science fiction movie, book about automatic checkout kiosks, and the productivity gains that they are bringing are, you know, small. Koen Smeets: It's very interesting, I think it's interesting to see the diverse uncertainty here. Mr Wolf, I was very curious for your perspective on this and for instance I saw an article from you by in foreign affairs in 2015, labelled "Same as It Ever Was: Why Techno-Optimists Are Wrong", in which you argue that the socioeconomic consequences of the technologies that we're discussing here, are relatively small compared to some of the technologies of the past. Is this still your opinion and could you expand on that?
Martin Wolf: Well I suppose I think my perspective, which actually rather surprises me is, similar to Daron's. I believe they relate fairly well, but I think he's far too optimistic. I'll try and explain what if he's right what I mean. I've been very, very influenced by in particular the work of Robert Gordon on the history of technology in the US. I tend to view things historically and not so much in models. And I think he tells a wonderful story which fits very well I think with Daron's, although he may disagree with that and I will interpret what Daron said in Gordon style way of thinking about it.
Which was that, for reasons which to me seemed to me essentially to some degree accidental. From the beginning of the Industrial Revolution and particularly after the initial displacement period, which was massively disruptive as Daron explains and quite clearly worsened the lives, particularly in Britain, of the industrial working class, I won't go into the whole history of that, and that seems to have gone on for something like half a century, so that was a big deal! To put it mildly, you know, half a century of a working class in miserisation, though that's a complicated story. But after that, we got, and Robert Gordon's work describes this I think magnificently, a vast range of disruptive technologies which rather beautifully offset one another.
And I think that is one way of putting what Daron said. So, massive displacement which was I think the biggest productive revolution in human history, which was the end of the mass agricultural labour force, which was what everybody had done for like, you know, since the agricultural revolution pretty well, for you know since ancient Egypt and earlier. So we got rid of the agricultural labour force, but fortunately all the other technologies, which developed, everything we can see about modern life, modern urbanization, the factory, the immense range of products could be produced in factories because of the electricity revolution, and the internal combustion engine, which of course generated a massive investment boom, the development of the chemical industry and with it the pharmacy, and just an immense range of general purpose technologies which drove growth for let's say 80 years, roughly. And those generated a whole range of new jobs which replaced the ones that were being lost. So there was pull as well as push, I think that's what Daron is saying, you know farmers, workers fled the farms and found lots of factory jobs and it also led to a political revolution. The new industrial working class, because of its hold-up power in factories, was able to generate immense power, extract rents, which were very important I think, and generate by the middle of the 20th century in our societies, the advanced societies, very high incomes for a large part of the industrial working class, which also generated with it a political and social revolution which we many of us, certainly in my age, I'm in my 70s, look back to with some nostalgia.
Now, the revolution of our time, Robert points out is narrow, in the sense that it's focused in one set of technologies, basically computing and communications [and] now AI, that so that we don't have all these offsetting technologies and all these offsetting industries that are exploding, in fact we're basically pushing the displaced workers into different forms of casual labor, doing fairly menial tasks, looking after ourselves as it were, in which it's very difficult to raise technology, raise productivity, and we've got a relatively limited number of huge winners, I'm staggering with this and I think there are other things. Now that's I think my description, my story, is I think another way, Daron might agree, of telling his story. But the point is, Daron and this is where we agree and that's a problem, because we've got one sector, or one sets of technologies which are dynamic, efficient, they're generating a lot of new opportunities, but they are clearly getting rid of an awful lot of employment. I mean an enormous amount employment and they destroyed the industrial working class, among other things, now well there are other factors. Now where I'm less optimistic is I don't think, it's not at least obvious to me that we will generate these new opportunities in any simple way without some corresponding, economic stroke technological revolution. So the narrowness of what we've got, generates both relatively slow productivity growth overall because what's left is so big and it employs so many people and of course productivity growth in all these labour-intensive sectors is pretty miserable. So that gives us this weak overall growth and we don't seem to be able to do much
about, you know, caring for children is caring for children, and technology doesn't help you much with that, idem dito caring for old people. And then, so it becomes very, very difficult to generate productivity growth, so what I think we end up with is massive disruption and no offsets. Now that's the worst of both worlds, now when I say massive disruption I actually think the disruption itself is smaller than the Industrial Revolution or the Second Industrial Revolution, is the offsets we lack, and that's what makes it so painful. So I actually think with Robert, really it's quite a small beer thing. However, and this is the final point which you raised,
is I did raise the possibility, which Daron just touched on, maybe at some point these new machines will be intelligent and not stupid. My main point was they're very stupid, so overlap with Daron again, and maybe we will find, half a century from now, that we've got machines that can really think better than we can and then maybe humanity goes the way of the horse, we cease to be economically relevant to ourselves and we have a completely and utterly different society, but that's certainly not we are where we are now. I think what I've just said, at rather great length, I apologise, is a journalist's way of saying, more or less, what Daron said but I may be wrong, maybe there's something I missed, but that's the story.
So the tragedy is, this is too narrow, we're not distributing the gains from it widely, pretty obvious, we have generated huge monopoly power in the process, which terrifies me politically and socially, I think that fits very well with "The Narrow Corridor"-thesis, which I buy, and I don't think we know how to generate these new jobs. I'm writing about this right now in a book I'm finishing on the crisis of democratic capitalism and the core is how do we generate these new good jobs which generate lives, livelihoods, meaning for people because to my mind, to bring it up today, it's our failure to do so that delivered what we saw last week in the Capitol. This is a massive political crisis, generated by an economic failure, which to my mind we don't have simple fixes for, other than a hell of a lot of concealed redistribution. Koen Smeets: Yeah, perhaps of importance to mention for those who are viewing this now is that we're recording this on the 11th of January. So yes, but I think it is very interesting indeed how the convergence of your stories, and indeed I saw Professor Acemoğlu very often nod. So, Professor Acemoğlu would you care to comment, and how do you see this and what do you see as some of the most important things in 21st century? Daron Acemoğlu: Yeah I would love to comment, because, you know, yeah indeed as Martin said, you know, our perspectives are surprisingly well aligned, except that he puts it much more eloquently than I do, so thank you Martin. And I think, you know, even the words are really so
similar, the word that you know Pascual Restrepo and I, in our work we use "counterbalancing technologies" and he used "offset" but I think they mean the same thing, and and exactly the way that Martin described it is beautiful. So 99 per cent we agree, I think there's one per cent slight disagreement, but it's useful to highlight it because, not just to distinguish our perspectives, but I think as a way of responding to your question, Koen. I actually don't think that AI is narrow. AI and digital technologies have the possibility to be very broad, but they have been misused in a narrow way, and that is the reason why, I think exactly like Martin emphasises the productivity implications have been poor, and they haven't helped generate good jobs. Let me give you an example. You know, the promise of AI, in some sense is that it would affect many sectors, but how and in what way, I think what we have failed to do is to create good jobs in services, as manufacturing employment has declined, in the same way that, for example, we had managed to create good jobs in manufacturing when agriculture was declining. Take education, for example.
I think everybody agrees education is critical, and if you think of AI being used in education there are so many different ways of using AI. For instance, there is an growing body of evidence from educational scientists, showing a lot of children are left behind because they can't follow some of the core subjects in which they are being instructed. And that then has huge consequences for their lifetime earnings, participation in society and everything. But AI
is a perfect technology for that: You can use adaptive technologies to find out which parts of the curriculum a student is having a hard time with, change the method of instruction, find the right remedial things targeted to that student, student-centric teaching. So why aren't we doing that? If we did that, what would we do? First of all, we would actually increase good jobs in teaching, because to do that, to use that technology and to actually build on that technology and provide more student-centric teaching, you would need very well-qualified teachers. That's exactly which what we should be excited about: generating new jobs in a sector that's critical. But of course, there's zero chance of doing that. Why, because if AI is going to be used in education, and probably will be in the next 10 years, we're going to do it in order to reduce teacher salaries and teacher employment, things like automated teaching, automated grading, automated exam setting. So it's a mentality of how we are going to use the AI whether it goes in the narrow direction or the broad direction. And that's why I agree with Martin, that we need a technological and institutional and I would say a vision revolution, but I don't think it's impossible. What stands in front of us, is actually first diagnosing the
problem and finding a broad enough constituency who understands it and pushes us in the right direction. That revolution won't happen, if a handful of companies and a handful of very similarly educated, with very similar ideologies managers are in charge of where our future goes. Koen Smeets: And then I would love to hear Mr Wolf's response to that, and I was very curious on how he sees that we can tackle some of the problems that are also I think underlying this, such as the immense inequalit and also other problems we've been generating last fourty years such as the climate crisis.
Martin Wolf: Let's go to the point that Daron just made, which sort of fits into another strand of literature which he will of course be incredibly familiar with, which is sort of Baumol's law type stuff. Which is the notion that, oh well, this has many aspects, but one notion is that a lot of the sectors in which we would like to see this enhanced productivity, are sectors which are either directly part of the public sector or are very heavily dependent on payments from the public sector. So, to take, here's one beautiful example, education, you know, children don't have incomes of their own, and many parents don't have very adequate incomes. And that's pretty obviously one of the biggest challenges of our society, of the inequalities in our society, which is, a relatively small proportion of parents have the resources to provide their children with all the enrichment possibilities he's just talking about, including highly paid teachers, and many aren't. So if we're going to do what he suggests, indeed pay teachers
well, hire very highly qualified teachers who can be augmented by technology rather, as we now see fascinatingly, grand master chess players aha they get better because they complement the machines, then somebody's going to have to pay for it. And the somebody is the taxpayer. And the related point, which he didn't explore so much, is how the distribution of gains come from those activities in which, and they're clearly very important activities, in which there is a platform monopoly of some kind, which does all the organisation, which may have just a few hundred software engineers and and owners, and then a very large number of casual workers. So that this generates, obviously, you can see it, just look at the stock market, staggering rents, which go to a very small proportion of the population. So, and in a way that is similar to what happened with the creation of the industrial working class, the difference was that the industrial working class, as I indicated to you and this was what was seen as a "problem" in the late 19th-century, could organize itself and fought very hard politically and economically to get some of the rent for itself, and interestingly all things there's a very good recent paper co-authored by Larry Summers on exactly this theme, about worker power. So this becomes, and this is one of the many respects in which I've become less of an economist quote unquote, this is Marxist, you know, who gets the rent, is not determined in economics. You know, the perfectly competitive model, it doesn't exist, but it does.
So, I think one of the implications of this, and I do think it is an important implication, is that if we want to ensure that these good new jobs and good new opportunities are exploited properly, and health is another sector, pretty obviously, okay, and I'm sure there are many others that I haven't mentioned, then we have to think about how these activities, which are going to be so important in future, culture, the arts, entertainment, everything like this, it's going to generate incomes, which are incomes that make people feel they're generally part of society. And that I think becomes politics. It was politics before, but it's even more now, because these activities are so centrally in the public sector, or at least involve public sector, engaged the public sector and so I now realize that 30 years ago or 25 years ago, when I was sort of very supportive of the market revolution, and in some respects still are, there is something I got absolutely wrong, which is, these things are not going to happen automatically in a desirable way without an active public sector engagement and without taxes. And they're going to have to be paid, raised and that raises all sorts of very profound issues about corporate taxation and tax havens, and we can go on forever, but I think we have to confront the reality, which and again I think we align and I think "The Narrow Corridor" is by that, that what he refers to as the "civil society reaction", is now desperately needed, because if we don't have a right civil society reaction, we're going to end up with civil war. That's why I mentioned this recent events, and it's a failure of economists, I think, to recognize those dangers, because and here again I think we agree, because there's a certain blindness about the natural harmonies of economic life and that comes from some very simplistic assumptions about how production is organised. Koen Smeets: Professor Acemoğlu, would you care to respond? Daron Acemoğlu: No, perfectly, I'm completely in agreement and, you know, I think I was just going to add something to what Martin said and and he added it himself. But let me just elaborate on
that. You know I think the role of civil society is really critical here, because I don't think this transformation that Martin is talking about can happen by government fiat, because we have enlightened people in government. First of all it won't happen if it's just left to government officials for a variety of reasons, which we can get into. But second even if it did happen it would lack legitimacy and it wouldn't happen in the right way unless society got involved, and to illustrate that let me actually give you one example which Martin didn't talk about but you hinted at, is climate. You know it's another one of our urgent challenges, I really think we have very little time left, but on the other hand I also think that there is an inspiring studies, this story from the climate change response over the last 40 years, you know in the 1980s, there was no way that any renewable technology could be used, anything close to cost-effectiveness. So despite the fact that government responses have been, to put it mildly, half-hearted and we don't have a carbon tax in the US, we don't have a global carbon tax which are, you know, musts, you know, we've seen an amazing surge of technological creativity that have brought renewables, to cost-effectiveness with fossil fuel-based technology. How did this happen?
Well it happened exactly through that civil society prodded government action. First, a significant fraction of the population understood the facts, understood the urgency of the issue, and started being mobilized on it. They started talking about it, they started writing articles, they started changing their consumption patterns, for example towards cleaner products recycling, you know, hybrid cars rather than gas guzzlers, and they started putting pressure on governments, which then did something critical: they started giving research support, and that's where the technology nexus is critical as we were talking on the earlier issue of job creation, they started giving technological support to clean technologies.
And even though that wasn't, you know, anything of the order of a New Deal or anything like that it was transformative. It's just solar panels, wind, geothermal, carbon sequestration, all of these things improved so rapidly in about 25 years, and we are now in a position that with really, quite feasible action on the part of governments we could actually turn the corner on climate change, all we need is continuation of the research support and a meaningful carbon tax, and accordingly international coordination. Not easy but but certainly feasible, something that would not have been feasible in the 1980s. So that's why I think civil society is really critical moving forward and the climate response demonstrates that in my opinion. Koen Smeets: Mr Wolf, are we still in agreement or did we reach a point where you diverge? Daron Acemoğlu: My optimism might have gotten in the way. [Laughter]
Martin Wolf: No this is a subject on which I agree very strongly. So I have been thinking, so Daron raised two issues. One touched on something I said and this is as it were how politics works. Now, it's pretty easy, and perhaps I'm guilty of this, to get pretty depressed about some aspects of how politics globally and in the developed world have been going. Particularly if, like me, you are rather a passionate believer, as I think Daron is from everything I know about him in the ideas and ideals of democracy, and of a democratic society. It's pretty obvious that some quite some things have gone quite badly wrong in many parts of the world, including in Daron's home country and in his prison country and in my country. [Laughter] So, widely shared, and this has a lot to do with these failures and challenges, and to put it very brutally, I think, although this is a more complicated story, because bringing Turkey and the US together is a difficult thing that won't go but anyway, part of it is clearly elite failure and an elite rent extraction, I'd be brutal about it, and the great upheavals of the middle of the 20th-century led to at least a temporary resolution of that problem in many countries, which has been unwound, always being unwound. So this very big story worries me a lot,
I mean terrifies me actually, and I could go into that much further. But there are stories where civil society responses work very well, and it's interesting to think why and who they are, because it's both a good story and a bad story. I think it's absolutely true, that civil society pressure has led governments to do some things, but also scientists and technologists themselves, have been influenced and business people have been influenced, by what they see going on around them. And then there's China, let's remember China has been a big deal in all this, because of the sheer scale of what it's done in some areas, like, you know, when the Chinese decide to do something they do it the way the US used to do it, right, in the middle of the 20th century, with solar panels. So these things together have brought about a technological revolution in renewable energies, which I have to say and I've been following this very closely for nearly 20 years, has surprised me, really surprised me, and I believe it surprised most people, how far the cost curve has got. If you go back to the famous Stern report of I think 2007 or 2008, I mean he was much less optimistic and that was based on the technologies of the sky. So this is a tremendous
progress, and I think we may get this revolution, despite the backlash which we're seeing, have been seeing in the US. And the interesting question is, what this tells you about what bits of civil society work. So basically, this is why, I think this is fair, this is a young middle class movement. Educated people, mostly graduates or people who are soon going to be graduates, like Greta, I think we can be safe ensure that, who are clearly the driving reforming radical group in our current society. You can see this in things like this, where I agree with them and things and other areas where I disagree with them less, of what's sometimes called the woke agenda, but these are mobilized, they get together, they make things happen, and they're a very, very important group in civil society. The problem with this group, is whom exclude. So basically they are completely separated from and completely uninterested in the old working class and it's plight, they are quite interested in some other causes, but the core causes their own. And this is related
to a profound social change, which i don't think we discuss enough, and I've only begun to realise in the nature of our society, which is for the first time in the whole history of the world, we have societies in which close to half of the population has been to university, has been in tertiary education, there are a few developed countries which are there and some that are getting very close to it. And that creates a much more profound class rift than I had ever realized, and it comes again all the work done by Anne Case and Angus Deaton on life expectancy, this is all divided by education, it's as if though, so we focus very much on the top one percent and that is actually I think quite important, the rent extractors as it were, but there's this bigger rift, so what Daron is saying is I think, that group, the university educated are becoming a very powerful, mobilizing, culturally dominant group and they will and they are mobilized and mobilizable and they will get their way, and I think climate is a success for this reason. Unfortunately, it leaves a very large part of our population out, and they are clearly feeling deeply displaced in this, because in addition to having no jobs and very poor incomes and no real prospects, they also feel profoundly disrespected by these people, because, to quote one famous politician, "they're deplorable". And, so, I would just say
Daron has described absolutely correctly what I think is a very important social development, which is also very positive social development, but one of the reason Trump got the support, is because he put two fingers up to these very climate conscious, climate mobilizing, middle class and upper middle class, university-educated people, it's our new, deep social rift. And the Democrats obviously represent them, and the Republicans nowadays, you know, there's almost nothing I agree with Josh Hawley on, nothing except the correct statement that, which is incredible if you believe you look at what it's proposing, that the Republicans and our conservative party and other examples are now working class parties. So there's a very important positive thing and there's something negative. The most important point about what he's, Daron's saying is that what I'm saying is that all these really big issues are politics. You can't separate our discussion of economic issues and economic policy from politics, and i think one of the tragic mistakes of economics, and Daron has done something remarkable in his academic career and bringing this back, and there are a few others, Douglass North and others, was this separation of economics from politics, which in a part was a reaction against what Marx tried to do, and I think now, I didn't think this 50 years ago, that was a tragic error.
Koen Smeets: I would love to move towards the way to economics education afterward, but before that I would love to hear what professor Acemoğlu thinks of what you said. Daron Acemoğlu: Well this is going to get boring, but Martin and I again completely agree, I think he again amazingly eloquently explained it and I'm really, you know, in awe of Martin all the time, but, you know, he put it so succinctly and what strikes me, is that what Martin explained isn't the conventional wisdom. It's actually a very scantily heard story. The person, I think, who captures some of the ideas that Martin said and I've been impressed by him because he says something that's not in the conventional wisdom, is Michael Lind, who wrote a couple of articles and a new book that, sort of, perhaps puts it somewhat differently than Martin and my perspective is closer to Martin, but I think Michael Lind as well as Martin gets a lot of credit in my mind for capturing this. And I also think that Martin's story, really, gets to the heart of why nationalism has been so resurgent, because when faced with the exclusion of the more educated and vocal classes, and faced with the globalization's discombobulating effects, nationalism has become a very powerful, potent force for the people who felt, who feel left behind.
But that's exactly why, I think, the role of civil society moving forward is not an easy one, but is a feasible one. It's because if we really start having a conversation about what's going on in our society, why it's very much in grave danger of fissuring, because of the effects of technologies, but that we also have a saying where these technologies go, I do, perhaps optimistically believe that the educated classes that have taken up causes that are not relevant for the other half, may actually take up causes that are relevant for the other half. It's perhaps a dream, but I don't think it's infeasible because it has happened in the past. You know, in the British Fabian society tradition, the most elite have sometimes been a very good conduit for the plight of the less fortunate, and I think, moving forward I don't see any reason why it couldn't happen if we start having the right conversations and if we free media from the many problems that it has, that's another topic that we probably don't have time to get into, but I think the media really deserves some responsibility here as well. Koen Smeets: I think it's very, very interesting to go into all of these areas and that we could keep doing this for hours, and I think I might have now a question which may cause some disagreement. Mr Wolf, you've been also quite critical on the economics education,
especially its lack of, for instance, political considerations and focus on neoclassical, mathematical models. Could you tell us more about your criticism of this? Martin Wolf: Well, probably being unfair because, let's be clear, I've been outside the academic community, though I'm a consumer of course, an avid consumer and moderately wide reader but not at the most technical stuff, but I've been out of it since 1971, which is a very long time. So I'm probably going back to my own education, or failures of education maybe, I'll make just two points. One is, I think, sort of relatively straightforward and I suspect there will be no real disagreement.
And one is much more profound, but maybe impossible to deal with. So the the one that is pretty clear, that when I was being taught, which was sort of as Keynesianism was dwindling, we had not just a political but an academic resurgence of pretty simple-minded free-market thinking. I think that would be fair, you think of rational expectations, real business cycle theory, some of them were simple-minded, what was come to be called market fundamentalism. And I mean looking back on that, it seems pretty obvious that these were embarrassingly simplistic ideas about how economies work. Now here, the point I would make is, in order to do social science at all,
which i think is a very different activity from the physical sciences, and particularly you know the canonical physical science physics, you have to do heroic simplification. Economic models involve heroic simplification, and the question always is how misleading are your heroic simplifications, and I think we got to a stage in which the model of humanity, the models of human beings, the model what matters to them, the models of technology, assumptions about society, were just hopelessly naive and therefore profoundly misleading. And I think most economists who I respect, well all the economists I respect would agree with that, and economics has done a fairly good job of rectifying some of that, by a combination of the immensely detailed empirical work, model-based empirical work of people like Daron, and by introducing new ideas, like behavioral economics and all the rest of it.
So I think that, the way professional economists approach things is less bad than that, right, quite a bit less bad than that. Of course, the consequence is that it's much more difficult to be confident when you're talking about economics, but it's at least more realistic. So that's, so I think there are improvements, I still think they didn't go anywhere close, far enough and I've discussed that my "The Shifts in the Shocks" in relation to the financial sector. I mean, which is pretty obviously a very, very big and well-known problem, but economics has made, there's a bigger problem and I don't know what to do about this and it relates to your second question, which is politics, but it's not really politics. It's the question to which I haven't got an answer, but I have been spending four years now on this sort of big political economy project of mine, trying to relate what the Greeks thought about democracy to what we think about the economy how does all this fit together.
In order to make social science work, our forebears back basically in the 19th-century, Schumpeter has a wonderful discussion of that in his "History of Economic Analysis", one of the books that's had tremendous influence on me. Our forebears, divided up society as it were vertically. They said, we're human beings, so we've got this cultural aspect of humanity, and there's the political aspect of humanity, and then there's the social aspect, and then there's the economic aspect of human behavior, human life, and we'll have studies of each of these. We'll have political scientists and
political theorists, and we'll have sociologists and social anthropologists, and we'll have, so I haven't even put in human beings, so, and we'll have psychology and psychiatry, and then we'll have economists looking at economics, and we will behave, because we are proper academics, as if the bit of humanity that we're looking at can be looked at all independently of the others, but that's complete nonsense, isn't it? It's not how the sciences work, the sciences work levels up. Chemistry is based on physics, chemistry is a form of applied physics. Biological sciences are based on chemistry and physics. So their model works. Now I think, I don't have any answer, solution to how to do this, none. But I think the basic way we went about this, sliced up humanity and human society in a way that obviously is bound to lead to grotesque outcomes. Because as it were, we did econ - I'll give you one very quick, I know it's gone on for too long, but one wonderful example.
Economists, I was one of them, went to the former Soviet Union and talked about economic reform. Did we start off by saying, well how does a legal system operate in this context? How, in a society which has gone through the soviet system for 70 years, how does law work? And how can we discuss economic reform we can't answer that question, which we didn't even ask? With very few exceptions, I could give you many, many other examples. So the problem is, I have come to the view, and Daron's work with this, particularly with his colleague, in these works, is a wonderful example because he's got people with two different expertise to do this. I'm doing this crazily on my own, but the only excuse I give for myself, is I spend a lot of time doing classics and politics, so I know a bit about that, I read a lot of literature in that area, before I ever started economics, which is why I'm a very bad economists who can't do math properly. But anyway, the point is, there's a deeper problem, I think, which is that, which is I don't think we have a good model for studying society. And that goes back to the invention, the separation out, in 1800 people didn't think these were separate subjects.
Political economy included everything we would now include, including Loki and psychology and [inaudible], they did everything, Hume did everything. Can't do that now. I think that is an incredibly deep problem, which makes it very difficult for us to talk intelligently as academics and journalists, for that matter, about society and we're struggling with that. Koen Smeets: If I can ask a very short follow-up on that - Martin Wolf: I apologise for being so long, I really do. Koen Smeets: I thought it was an amazing answer and I really, really agree with you, but I was wondering, what could we practically change? Is that including more interdisciplinary or are you thinking something else? Martin Wolf: Is that for me or for Daron? Koen Smeets: If you have, like, a couple of thoughts? Martin Wolf: I have no idea, except, well I do think, that the idea you can take an 18-year-old from a school, give him pure economics, which is basically a lot of maths, mass models, nothing else, knows no history, no psychology, no philosophy, no political science, no sociology, no anthropology, I think you're producing somebody who cannot deal with nearly all the problems that will be actually faced in doing economics well in the world. I would go for a broader education and then, of course, you would have to accept that many people will have to, if they're going to be a real economist, they're going to be even longer as a graduate studies and that probably doesn't work. But
at the moment I feel a lot of economist, Daron is a great exception, there are a few others, the most brilliant ones actually, who are just ludicrously narrow. Koen Smeets: Professor Acemoğlu, do you agree on this, and how do you view Mr wolf's comments? Daron Acemoğlu: Well, these are very deep and challenging topics, but I'm going to be brief. First of all, I agree with Martin. There are deep issues and I don't know the answers to them. I also agree with him that there are serious problems in the instruction of economics. I wouldn't put the blame on mathematics. The way I see it, simplifying it and making it brief, since I know we're running out of time, there's the good and the bad, hopefully not the ugly of economics. The good is that it's a doggedly empirical,
sceptical, fact-driven profession, and I think mathematical-statistical training is critical for that. It's also a broad tent, you know, all of the issues that Martin has raised are correct. But economics is a very broad tent, people do many different things and in some way that makes it much more open-minded, in a way. You know, you cannot imagine, you know, things like behavioural economics, political economy, institutional economics, social economics, spreading so rapidly in many other disciplines as they did in economics within the course of about 20 years or so. On the other hand there is the bad. The bad is both at the graduate and the undergraduate level. There are many studies that show, and I'm doing some myself, that people with economics education or business education are less cooperative, they're stingier, they're tougher managers, they cut wages. And I think some of that is selection, perhaps different people are attracted to economics, but it is a problem, that is related to what Martin pointed out, that many economic students at the undergraduate and the graduate level learn from economics a type of vulgar market fundamentalism, selfishness, greed is good, you don't need government, you don't need redistribution, the invisible hand works, and the wilder the better. And I think it is a failure of our instruction, but it's got nothing to do, in my
opinion, with mathematics, because there are graduate students who get the very mathematical version of economics who end up with that vulgar market fundamentalism, and there are many, many more undergraduates who don't get the mathematical, especially in the US, completely non-mathematical undergraduate education and they end up with the same conclusion. I think we have to have a rethink about how do we communicate the beauty of economics, as envisaged by Adam Smith, but also with the nuance that Adam Smith himself understood about the invisible hand, and its moral foundations and its limits. And how do we do that without sacrificing the good of economics, I think that's an open question, certainly a broader set of perspectives would be useful, but I'm not sure whether that would be by itself enough. I think we need to do much more soul searching and people like Martin, from a broader,
different perspective, sort of coming up with ideas and criticism, I think that's super useful. Koen Smeets: I think we're running out of time, I think we could do this for hours, but my main question, my closing questions in the interview series is the following: If there's one thing you could say to students in economics watching related to the topics we discussed today, what would that be? And I think it will ask the question first to Mr Wolf and then to Professor Acemoğlu and then we'll have to close. Martin Wolf: So I think I'll just have two words, three. Rigour is important. So Daron is completely right. The rigour of economics is wonderful. Scepticism is crucial. You have to ask yourself, does this simplification actually work in the real world and why might it not? I mean, that's very important, we know much less than we would like to know. And the third thing is, I would say, and this may be a private thing, is try to be as broad as you can.
Economics is a social discipline, which means it's a moral discipline. It's a political discipline. You must not allow yourself to become narrowed away from that awareness of what it means to be human even if you're an economist. Koen Smeets: Professor Acemoğlu, what will be for you - Daron Acemoğlu: Again, I agree with Martin, I would say question, question, question. Question conventional wisdoms,
question the boundaries, question authority, but also question yourself. I think that last one, perhaps also questioned the boundaries which has led to narrowness, I think are the two that we haven't done enough and we have to do more. So question yourself in economics means, you know, question all of the assumptions that we have come to make about markets, about individual motivations, about social cohesion. Many times these assumptions may be okay but there will be many, many instances in which they are inappropriate and that can really lead us astray. Koen Smeets: I think that it is a beautiful way to close this interview. Over the next weeks we'll be
posting the rest of the interviews in this series, thank you so much for your time. Daron: Thank you Koen. Martin: Great pleasure. Daron: It was great pleasure to have the stage with Martin. Martin: Great pleasure for me too, I learned a lot. It was very inspiring. Koen: Thank you.
2021-04-08