Creative Destruction

Ann Mossop: Welcome to the UNSW Centre for Ideas podcast, a place to hear ideas from the world's leading thinkers and UNSW Sydney's brightest minds. I'm Ann Mossop, Director of the UNSW Centre for Ideas. The conversation you're about to hear, Creative Destruction, is between distinguished economists Philippe Aghion, from the Collège de France, and Richard Holden from UNSW Sydney, and was recorded live. They're talking about Philippe Aghion’s recent book, The Power of Creative Destruction: Economic Upheaval in the Wealth of Nations, co-written with Céline Antonin, and Simon Bunel. I hope you enjoy the discussion.

Richard Holden: Hello, and welcome to tonight's event, Creative Destruction. My name is Richard Holden, and I'm a Professor of Economics here at UNSW Sydney. Tonight's event is presented by the UNSW Centre for Ideas, and we're thrilled to be joined by esteemed economist Philippe Aghion. Firstly, I would like to begin by acknowledging the Bidjigal People who are the traditional custodians of the land on which I'm speaking tonight. I would also like to pay my respects to the elders both past and present, and extend that respect to any other Aboriginal and Torres Strait Islanders. I'd now like to welcome from France my friend and former PhD advisor Philippe Aghion, who's the professor at the coalition of France and INSEAD, one of the world's leading and largest graduate business schools. He's also a visiting professor at the London School of Economics. And among numerous honours, He is a fellow of the econometrics society, and of the American Academy of Arts and Sciences. His most recent book, The Power of Creative Destruction with co authors, Céline Antonin and Simon Bunel, is the one that we will be talking about tonight. And I encourage you all to get a copy, digital or otherwise, or both, as I have. And we really look forward to this conversation this evening, Phillippe, welcome and thank you.

Phillippe Aghion: Good morning, or good afternoon.

Richard Holden: Phillippe you have made many contributions to economics in contract theory, financial economics, many areas and obviously the topic of tonight's conversation will be with endogenous growth theory and growth theory in general. I’m sure it will be a wide ranging discussion, because growth theory has so many implications for economic policy generally, for macro economic policy, tax, the environment, many things, we'll talk about many things. But I'm interested to start with, I'm always reminded of a quote that I think is due to the great economist and Nobel laureate Bob Lucas, who said something along the lines of, “once you start thinking about economic growth, it's hard to think about anything else”, I wonder what first got you thinking in economic growth?

Phillippe Aghion: So, I came to economics because I was a political militant, in a left wing movement in France, in the early 70s. And until the early 80s, and, it was all about changing the world and making the world prosperous, and fair. So it was really about how to achieve shared prosperity. And you know, various people with different things on how to do it. And it was very hard to know who is right and who is wrong. So I started questioning myself. And that questioning never ended, to this day. And that's what got me into economics and to economic research. But it was economic research to change the world, to make the world better, to make the world prosperous. And in a way, which is fair, to get people out of poverty, to allow anybody from any social origin to get somewhere and to realise it is his or her potential. And that's why I went into economics. So it was very natural that I would go into growth economics.

Richard Holden: I think that's probably the best recruiting pitch I've heard for why young students should do economics. And so we'll borrow that, and we'll replay this clip to try and get more people to come and do economics, preferably at UNSW. So I guess, when you came to grow theory, there was a dominant theory of growth, beginning in the 1950s. And before you develop so called endogenous growth theory, that was the neoclassical growth model. Can you tell us a little bit about that?

Phillippe Aghion: Yes, absolutely. So the leading model was the neoclassical model of Robert Solow. Solow’s paper was published in 1956, in the Quarterly Journal of Economics, and it was a benchmark model, because it's a very elegant and parsimonious model. It has only two equations. The first equation tells you how you produce final goods, or consumption goods, using a stock of capital and stock of labour. And the second equation tells you how you accumulate capital. So the idea is that you accumulate capital, because part of the production is being saved, and the savings turn into investment and that's what generates capital. But what Solow showed is that under reasonable assumptions on the aggregate production function, you cannot get sustained growth just with capital accumulation. That doesn't work. Why, because you have decreasing returns to capital accumulation. When you increase the number of machines you have from zero to one, you increase a lot of production. But when you increase the number of machines from nine to 10, you increase production by very little, you will run into decreasing returns. And because of these decreasing returns, if you just count on capital accumulation to grow, you run out of steam at some moment, you cannot go forever just through capital accumulation. And so Solow would say when you realise that you need something called technical progress in order to have sustained rules, in order to have long run rules. But he would not tell you where technical progress would come from. They have to understand where it comes from, you have to go back to Schumpeter, because Schumpeter had this view that, you know, it's innovation at the heart of the growth process. So in fact, technical progress comes from innovation. But Schumpeter did not have a model and didn't have empirical analysis. So what we did with Peter Howitt, was develop a paradigm based on the notion of creative destruction, we operationalise the notion of creative destruction. And we put it at the heart of a new growth paradigm, where we developed a model and we confronted the model to micro data. And in fact, now there's been a whole generation of economists who have further developed the paradigm and further confronted it with micro data.

Richard Holden: Fantastic. So what are the key ingredients of your model, of the Schumpeterian model of creative destruction and growth?

Phillippe Aghion: In fact, there are three main ingredients. The first ingredient is that long run growth comes from cumulative process of innovation, each innovator builds upon the giant shoulders of her of his predecessors. So that's the first ingredient, you need a system of cumulative knowledge. The second thing is that innovation that does not come from heaven, it results from entrepreneurial activities motivated by innovation rents. I innovate, because I know that if I innovate, I become a monopoly for a while. You see? I am able to produce something that nobody else produces, or I am able to reduce the production costs of a good, and therefore monopolise the market for a while. Okay? So I get an innovation rent from my relations. So that's very important. And the third idea is creative disruption, new innovations make old technologies become obsolete. New innovations replace old technologies. But you see, by the way, before I move on, that at the heart of Schumpeterian growth, you have a contradiction. On the one hand, you need innovation rents to motivate innovation, but on the other hand, there is a temptation for an innovator to use her or his innovation rights to prevent subsequent innovation, because we want to avoid creative destruction. I don't want my set to be creatively destroyed by someone else. So I'm very tempted to use my innovation rents, to put entry barriers, to bribe the government to do all kinds of things to prevent entry and subsequent entry and subsequent innovation. And regulating capitalism is all about how to manage this contradiction. And the whole book is about this contradiction. It has very… for the debate on inequality, for the debate on competition policy, for the debate on secular stagnation, for the debate on the environment, everywhere you go… for the middle income trap we talked about, it's for all the big dilemma, the big, you know, enigmas of economic growth. At the heart of it, you have this contradiction that I just mentioned.

Richard Holden: That's fantastic. And it does seem like a paradox or a contradiction, as you say, because on the one hand, we need these powerful innovators, and we need to give them some monopoly for a while, to give them the right incentives to innovate. But if they can use that power, and that could be political power, through money, and so on, to corrupt the process and protect themselves, then, you know, we get the other end, the other edge of that sword, and it comes back to hurt us. So at the heart, it seems to be kind of a political problem as well as an economic problem of incentives.

Phillippe Aghion: Absolutely.

Richard Holden: So I see why you're the perfect person, the left wing agitator and the formal economist, you have to put those two ingredients together to get to the heart of this problem. We're going to come to a lot of those things you just mentioned, because there are so many implications of it. The first one I thought we might start with, you already mentioned it, because it's been very topical, since maybe 2013. Is this idea of secular stagnation, and the way, you know, correct me if I'm wrong, the way I think of it is, you know, this was your colleague, former colleague, Larry Summers, sort of re-popularising the term from Alvin Hansen, from a while ago, and sort of suggesting that the, you know, the speed limit is… the way I think about it is the speed limit on economic growth is consistent with, you know, stable capital accumulation, is lower than we thought it was in the past. And we're gonna have to learn to live with, you know, maybe 2% economic growth in the United States, or Australia, or France, and not the kind of three and a half percent or 4% that we've seen in the past. What are the implications of your theory for secular stagnation? And can we get out of this world where we've been for nearly a decade, where even before COVID, and even, you know, absent a big financial crisis, we kind of had pretty low growth in advanced economies compared to historically.

Phillippe Aghion: So first, you have to understand first, that developing countries, countries that are catching up mode, they tend to grow faster. China grew at 10% a year or 7%. But that's because China is catching up with more advanced technology. When you start from a very low level, and you catch up with the most technological frontier, you are bound to grow faster. So first, you have to understand there is a normal reason why you should slow down as you catch up, okay, we cannot grow at the same rate as China, because we are developed economies, Australia and France, we are already at the technological frontier. Okay? So first, you have to understand that. We cannot grow at the same rate as China. But still, we could grow more than we do. The big enigma was right… how come in the US in particular, with the IT revolution, and with the artificial intelligence revolution, the US experiencing a decline in productivity growth since the early 2000s. And in chapter six of the book, we explained that the main reason has to do very much with the  Schumpeter contraction. During the IT revolution, the IT revolution made it possible for some firms to become superstar firms, hegemonic firms, Amazon, Google, Walmart, Facebook, these firms grew very big, and they invaded all sectors of the economy. And at first, it's perfect productivity growth. If you look at productivity growth in the US between 95 and 2005, it was much higher than before. But then once they invaded the other sectors of the economy, they stifled innovation by other firms, you see, they discourage innovation by other firms. And that's why you have the decline. And so now the problem in the US is a problem of competition policy. You saw that Biden last week made a big move on competition policy, it was very much needed, because you want somehow to limit the power of these superstar firms. They could do merger and acquisition without any limits. They could, you know, explore the advantage they have on data to put entry barriers to other firms, they could take advantage of tax, you know, advantages, they get a road to have a competitive edge on other firms. We know that big firms in the US have a huge power to finance political campaigns, to lobby. So you need in the US to really deal with competition, to regulate merger and acquisition, to make sure that it's not preventing subsequent innovation and entry, you need to fight lobbying, which is stupid there, you need to regulate the financing of political campaigns, you may have to break up some of these. But you need big action to limit the power of incumbent firms, and that goes back to the contradiction I was mentioning before. In the US the problem is that large incumbents, particularly these superstar IT firms, they can really stifle innovation by other firms. And you have to really reform competition policy, to adapt competition policy, to the digital era in the US, in order for growth resume, because it's really a competitive barrier, a competition barrier, to more sustained growth in the US.

Richard Holden: That's so interesting, because I think that lens of innovation gives one a different perspective in the new economy, on competition policy, than in the old economy. You know, the old economy or the pre digital economy, we would think if there are four firms in the market that have 85%, or some number, you know, we have these indices in economics HHI, and things like this. We say that's a very concentrated market. But maybe that's not quite the right way to think about it in this sort of new economy, which is, if we have one firm, has 80% market share, well, that might have great network externalities, and might have great benefits to be on one platform. It probably depends more, what are they doing for innovation? 

Phillippe Aghion: Exactly.

Richard Holden: So maybe if it's, if they're buying up competitors, if they're killing innovation of small companies, you know, some people have talked about this idea of killer acquisitions that they'll buy a small firm to shut off the innovation and things like this. And so it seems that innovation lens is very important for thinking about competition policy going forward as well. So that's interesting. And it brings in, seems to me, it also brings in, you know, international tax policy, as well, as we'll get back to it later. But with Janet Yellen as Treasury Secretary working on trying to dial down some of Europe's digital tax, but also maybe cut off some of the Irish tax havens and things like that, it seems all connected these things to, to innovation ultimately.

Phillippe Aghion: Yeah, absolutely. So for example, it is very important on competition policy very much the way to daunt, I trust, up to now in the US, but also in Europe is very much based on the notion of market depletion and market share. If you have a huge market share, I would go against you, you see what I mean? But you may have a very big market share and be a contestable market, when you are contestable is you have high potential entry. So I may have the whole market share, for example, the high speed trains in Europe, you know, there was Alstom and Siemens, and they wanted to merge, and the Competition Commission in Brussels to the no, because the two of you, you have the full market share. But that was a very bad argument to prevent the merger because the market for high speed trade is highly contestable, because the Chinese produce high speed trains. So anybody in Europe would go and buy trade from China. So you see, it's a very limited notion, market share and market definition. And it's much better to say, before I forbid a merger and acquisition, I look at the effect it would have on subsequent entry and subsequent innovation. It's a much more, you know, this dynamic way of doing choices, I think, much more relevant and particularly to the digital era, and to deal with the problem I was mentioning before. So that's one aspect, you have to change the way the practice of antitrust, putting much more innovation and entry at the heart, and much less the static market share market definition, notion at the heart. And the other thing, as you mentioned, is that, of course, if the launch gap firms, you know, the Facebook et al, they could take advantage of, you know, tax havens to get a competitive advantage over other firms in the US. And I think whatever can be done to limit the tax havens worldwide will also increase competition, because it will make the market a level playing field much more. And that I think is also the tax level is as important as the direct redesigning of competition policy. But to make the environment more competitive.

Richard Holden: Yes, that's clearly very important. So one of the things that he's touched on earlier on is obviously we need to provide incentives for people to innovate. And you also talked about the inequality. So how do we think about inequality in a world where we're trying to foster innovation? So you mentioned some superstar firms like Amazon, they are also, I don't know if they're superstar owners or people, but they're certainly very wealthy. But they've also created very big important things. Does it really matter, if you know, Steve Jobs makes 10 millions of dollars when he creates Apple? I mean, you know, I don't know how rich Thomas Edison never got, but he created some pretty great things. Do we really care if Thomas Edison's worth $100 billion or $200 billion? What’s the right way to think about the link between innovation and inequality?

Phillippe Aghion: So my view, my view, and there I depart from other other people working on inequality is that other people working on inequality, they will never distinguish where the rents come from. It is not the same thing if the rent comes from successful innovation or if the rent comes from, you know, entry barriers, lobbying, etc. You see what I mean? So that's why I distinguish between Steve Jobs and Carlos Slim. Steve Jobs became rich because he created Apple. Carlos Slim is a rich Mexican industrialist. Became rich a lot because he's at the head of telecom, the Mexico Telmex, which is a non regulated monopoly. And those are two very different things. And you cannot ignore that distinction. It's very important, when you look at the rents to know if the rents come from innovation or from pure entry barriers. You should not treat that the same way. So now my problem with inequality is not that you have rich people. My problem is that you have poor people. I am upset by poverty. I don't want people to be poor. I don't want people because they are born out of poor families to not have the same possibilities as people born from other families. So I discussed that a lot in chapter five and chapter ten of the book, where we look at the lost Einsteins. You have many people who couldn't be innovators, and they are not innovators because they come from poor families. I would like everybody to get access to the same education with the same aspirations, so that they could all become innovators. You see what I mean? So my view is that the problem is not that you have rich people, the problem is that you have poor people. And by virtue of poor people, the poor people do not have the same access to education, the possibility to become inventors and to realise life. And that's what I'm obsessed by. So my policy is really to deal with social mobility. And it's very important, social mobility. You can reach social mobility with education. Education is very important for social mobility. But innovation also, because because of creative destruction, innovation generates social mobility. So it's very interesting with innovation, because innovation is true, that it's a source of inequality, is a source of inequality at the top, because when you innovate, you get grants, okay? But it's not the only source of top income inequality. Lobbying is also a source of inequality. But the big difference between innovation and lobbying, is that innovation is a source of top income inequality, that raises social mobility because of creative destruction, and as a result, innovation does not increase global inequality. That's what we show in the book. So it's a very interesting source of top income inequality, innovation. It raises the trop income inequality, the inequality at the top, but it also increases social mobility. Therefore, it does not increase global inequality. And it raises rules. So innovation is a good source of income inequality. Whereas lobbying is a bad source of top income inequality, lobbying raises top income inequality, but because it reduces entry of new or new firms, it reduces creative destruction, it reduces social mobility, and therefore it decreases global inequality. But also lobbying reduces growth because it reduces innovation by new entrants. So you see, it's very different. Lobby and innovation are very different sources of top income inequality, because innovation is a source of top income that doesn't increase global inequality and increases growth. Whereas lobbying, it's a source of top income inequality, which reduces growth and increases global inequality. So you should not deal with them the same way. Now, having said that, should you worry about the rich? In Sweden, you have rich people. In Sweden, you know, Mr. Sky became very rich because he invented Sky. You have a wealthy family. But the thing is that they cannot, they are no problem because they don't prevent new firms from entering. And that's where it’s Schumpeterian. You want to make sure the rich do not use their wealth, to buy the competition, to prevent entry, to bribe governments. And in Sweden, you can't bribe government. A minister had to resign in Sweden, because she purchased a big piece of chocolate with a credit card of the ministry, a piece of chocolate. You cannot buy a politician if you're wealthy in Sweden, you cannot buy this competition. So I don't mind having rich people. But you need to put a no fly zone above that. That means you can be rich but don't use your wealth to prevent entry. Don't use your wealth to lobby. Don't use the wealth to bribe governments. Don't use the wealth to finance political campaigns. And that's my way of looking at equality and innovation.

Richard Holden: I think that's very compelling. And it illustrates another good thing about thinking things through an economic lens, which is, you know, some people say, the fact that we have billionaires is bad. I think Alexandria Ocasio-Cortez famously said, every billionaire has a policy failure. And you know, one way to deal with that is to say, well, we'll have, you know, confiscatory tax rates so we don't have billionaires anymore. And try and deal with it that way. But as you say, if you distinguish between what people are doing with the money, rather than the fact that they might have a lot of money, you might not kill the innovation. Maybe it's more a question, say in the United States of money in politics, and supreme court decisions like Citizens United and structure of government. So yes, well, when maybe your next book is to be about how the US political market can become a little more Swedish or something like that, because that seems like something that's fairly far from the case. So that links up nicely with, you've got views, I think implications for innovation policy for tax policy and social mobility. So there's been some real debate with, actually quite a number of your compatriots, a very good French economist, with the view that, you know, we should be raising top marginal tax rates in countries like the United States, and Australia, and the UK a great deal. What's your take on social mobility and innovation, and tax policy in general?

Phillippe Aghion: So first, I want to say that you know, I am very much Scandinavian. I think what the Danish, or the Swedes are doing, I think is very good. So you need taxation first. Why do you need taxation? Because you want to have an education system free and accessible and of good quality for everybody. You need a good health system, free and accessible to everybody. You want also to finance active labour market policy, the flex security system in Denmark, maybe we will talk about it, you need to finance this. And you may want to have also a smart industrial policy, like the DARPA, BARDA, type of thing. So for that they need to raise tax. I mean, in Denmark, if Denmark was not raising tax they could not do what they do. So I think you need taxation. Okay? So I believe in taxation, I believe that taxation should be reasonably progressive. I believe that the tax rate should be lower for poor incomes  than for rich incomes. Okay? So that's why you wanna be in Denmark. So you have reasonably progressive taxation on labour income, and they raise tax, because if they don't raise tax, no education, no health, no labour market policy, no industrial policy and all that. So I think that's very important. So I believe in taxation, I’m not a Trumpist regarding taxation, not at all. I believe… I'm a Scandinavian, so I believe very much in taxation. 

This being said, Biden raised the tax, Biden was right because tax were way too low in the US. In the US, they have a huge problem that the social model is broken in the US, they did not properly insure against COVID. Many people lost their jobs because of COVID. And because they lost their job, they lost health coverage, at the moment when they needed health coverage. So they need to rework the  social model. And clearly from where they were, there was a need to increase taxes in the US. So Biden was right. Of course, the problem is Thomas Piketty, my friend and others, they want to raise tax in France, but in France they are already very high, and they want to raise the capital income tax, and which is already very high in France, they want to raise inheritance tax, which is already very high in France. I am for inheritance tax, I am not for lowering it in France, but I don't think it should be increased. Because if you have excessive… you used to have before, and that Piketty was okay with that. You used to have marginal tax rates on capital income of more than 100%. That's crazy. So what Macron did very much I pushed out that with him, to have a flat tax of capital income of 30%. You see, because it was crazy. It was discouraging innovation. And it's been warned by Stantcheva, Akcigit and others showing that indeed, excessive capital income taxation discourages innovation. And you don't want to discourage innovation, because if you discourage innovation, it is bad for growth, but it's also bad for social mobility. So you shoot yourself in the feet. So you need taxation, I believe in taxation, but you need to have a reasonable tax system, a reasonable tax system, it's a tax system, not over taxing capital income, and where you don't overtax. Because you don't want to discourage people. In Britain, in Sweden, they used to have also excessive taxation of capital income and wealth, people were leaving the country. And they realised if they wanted innovators to come back, they needed to put taxation at a reasonable level. Still, they have a very developed welfare state, in Sweden and in Denmark. So you need to strike the right balance, you need taxation to finance, education, health, labour market policies. So you need to raise the tax. You need to have reasonably progressive on labour income, but I think a tax… 30% of capital income, a corporate income tax at a reasonable 25%, is reasonable. And then you have a tax system that works. And I would be favouring the inheritance tax. In France, we have a reasonable inheritance tax, which is fine. And I would not raise tax in France. That's crazy. But I would have raised that tax, like Biden did in the US, it all depends where you start from.

Richard Holden: It certainly does depend on the starting point, I always thought it was, you know, you can agree or disagree. But when, you know, Ronald Reagan wanted to cut taxes, because they were at a very high point, it's quite another thing to try and say, I'm always for tax cuts, some Republicans, I'm always for tax cuts, even if they're already, you know, 19%, you know, lower taxes, and so on. And always remember that President Kennedy, you know, cut taxes, marginal tax rates were 91%. In the United States, when he did that, that didn't make him a Trumpist either, by any stretch. So that starting point is obviously very important. I guess that also connects up with the idea of human capital mobility. And if you're going to have innovation and innovators as a country, you know, you can't be a country like Sweden, or Australia, or France, have some of your best minds and human capital, basically, you're innovators, take your education, grow up, get all that human capital, and then disappear overseas to pay lower tax rates. So you have to be internationally competitive in that area as well. I think one of the other things so important as an implication of your work is about, if you like, green innovation, and you talk about this in the book, and environmental innovation, the natural environment is one of the most pressing issues of our time and dealing with is something where, I think, many of us economists – you can tell me if I if I'm wrong on this – but many of us economists think we've understood a little bit what the solution has been to this for quite a long time, for decades and decades, but there's been a political problem in getting these things done. What's your view on how we need to make progress on dealing with the huge environmental problems the world has, but also in doing it in a way where we can foster innovation to actually, you know, find the new technologies that can reduce our reliance on fossil fuels and other things that are bad for the natural environment?

Phillippe Aghion: So on the environment, I think there was a very interesting natural experiment. And the natural experiment is that with the first lockdown in France, it was in March, April, May last year, GDP went down by 30%. But co2 emissions went down by only 8%. So that shows that negative growth is not the solution. The solution is green innovation. And the question is, how can you get green innovation? Then you have a problem, is what we call path dependence. And we did with that in chapter nine of the book, firms that have innovated in dirty technologies in the past, will spontaneously prefer to keep innovating in dirty technology today, because you want to keep doing what you're good at. That we call path dependence. So, first solution is to upgrade the disruption, encourage new firms to come because they don't have that problem. You see, the new firms they are not wedded to to the past. So they don't have the balance problem. So creating disruption is already a way to deal with Green Innovation, okay? But then you have to also redirect the innovation of the existing firms towards green technologies, then you have several instruments, you have the carbon tax, the carbon price, basically works very well, you have another instrument, which is subsidies to clean R and D, and in what we call smart industrial policy. In France, we have big nuclear plants and hydroelectric plants that help to reduce co2 emission, but we subsidize also all kinds of R and D activities, you know, to allow green technologies. So that's the second thing. So carbon tax R and D subsidies and industrial policies, the second one, and the third one is the consumers, when consumers want green competition is also a new way to make the economy greener, because if I don't innovate greener, you will. It is a combination between educating consumers and having competition, this combination is as important as the carbon price to induce greener innovation.

Richard Holden: Very good. So we'll see where the Biden administration goes, and other countries. Australia is… we're proud of many things, and we're world leaders on many things, tackling environmental problems is sadly not one of them. You know, we had a carbon tax and then got rid of it. And we have some issues with trying to deal with that. But I think the carbon border tax that Europe seems to be likely to put in place, and the US may copy, we may get a price on carbon in Australia, because everybody else would have a price on our carbon exports. So we'll see how we go on that front. I guess one of the big debates in the United States – and we'll see whether Biden administration comes out on this – has been between, you know, a carbon tax, price on carbon, maybe a carbon dividend, like the Climate Leadership Council have talked about, that’s had a lot of support, particularly from economists. And then something more like, you know, because you know, Cortez and others have championed, Bernie Sanders, or Elizabeth Warren, something more like a green New Deal. That's a bit more as I understand it, a bit more interventionist, and, kind of, smart industrial policy, to use your term, that's more spending trillions of dollars for the government to do a lot more stuff. How do you think that that debate should resolve itself? Where does innovation theory fit into that?

Phillippe Aghion: Innovation has a lot to be bottom up. You see, I mean, grassroots you see, so I very, very much in horizontal policies, create the environment for innovation and creative destruction to take place. Okay? Make entry easy, so innovation first is always on top, education, taxation, competition, all these things, horizontal, they apply to all sectors. But it's true also that I believe in smart industrial policy. And in the US, they found the DARPA (Defense Advanced Research Project Agency.) It was created in the 50s for defence and space. That's a situation where the basic technology, the basic knowledge has been generated in basic research, but you need to translate it into applications quickly. And you have very well defined missions. So then you have to coordinate a dozen resources, and from that the data was very effective. I need to put the man in space. When GAGAN went in space, I need to put the man in space, in one year, I need to develop defence systems in one year or two year. And we saw that recently with the vaccine data, biologique advanced research agency, we are the ARN, messenger of technology, we need to produce within a year a mass production of vaccines based on this technology, so that you need this data. But what's very good with the data is that the money, the funding comes from the ministry, then they have team leaders, and the team leaders elicit competitive projects you see, so there, you have the bottom up path, it’s a mixture of top down and bottom up, and the team leaders elicit, you know, competitive biddings, many people come to them, and they encourage private/public partnerships. And that's competitive, you know, BARDA helped Pfizer, Moderna, many others that you don't know about, and that didn't succeed. And that I think is a good thing to do. So they did that for the vaccines. And I think they also have, in the US, ARPA energy. They have an agency for energy, you may want to do more, but they have the structure that allows to do it in the US, which is the ARPA energy, and we need to create the equivalent in Europe. And that would be great to do. I think that's the right tool. So it's important, you can put more money but the government is very important, because you don't want to have the pitfalls of bad industrial policy, which is to stifle competition, and to help incumbents at the expense of entrants. You want an industrial policy, which is pro competition and pro entry. And the DARPA, BARDA, well done can be that, you see, but it has to be well governed.

Richard Holden: I see. So, you know, this maybe draws a connection to some of your other work in contract theory. But it seems to me there's a little bit of an analogy here to the theory of firms versus markets. You know, I think we have learned over the years that, you know, one of the good things about firms is authority. But if you're the boss, and you want to make a car that's painted green, you tell people to paint it green. And if they don't like it, you fire them and you find somebody else who will paint the car green, you'll get a green car. But markets have competition markets also, you know, as Hyack famously pointed out, help aggregate and communicate information. It sounds like, a little bit what you're saying with DARPA, and with some of these, say developing vaccines, sometimes you have sort of, like a space race, wartime production environment, and there you really need coordination and authority. And other times you need to kind of let 1000 flowers bloom, let people come up, they try mRNA, it works, it doesn't work, they try something else. Is it that kind of thing where you want an ecosystem that can allow 1000 flowers sometimes to we're fighting a war, and we're going to do it this way.

Phillippe Aghion: Yeah. So you have to understand, for basic research, you don't have you don't need a DARPA for basic research, you need to have many people be free. Do you have areas, for example, Silicon Valley is not the DARPA, in software, you don't need that you see what I mean? So you need that in particular areas where the basic research has been done, and you need to go very quickly to industrial production, with clear missions. It's only those activities that are DARPA-ble, you see? I mean, not everything is DARPA-ble. And in those things there, you need coordination problem. There you have the coordination problem, and you need the structure to deal with the coordination problem, but you don’t want to kill the grassroot. And that's where DARPA is great, DARPA solves the coordination problem without removing the competition, the free entry, the bottom up part, you see, I think that's where they found something varied that works with a DARPA model. But you see you have to be aware of the different parts of the research. Basic research, for example, needs freedom and openness, very much actually, then you have the research within firms, but you have sometimes these areas where you need to coordinate actors and resources and on those segments, then DARPA can be helpful, but it's in particular instances. Otherwise, you want to go as much competitive as you want. You always have within the firm the tradeoff between initiative and coordination. And you need to find the right compromise between coordination and the need to raise initiative, too much top down kills initiative. But sometimes, too, if you're just bottom up, the incentives do not look good and that does not take place. That's exactly the same trade off. But I want to take advantage of this question that you asked, going back to the Schumpeter, was pessimistic about the future of capitalism, because he thought that innovators will turn into conglomerates that would stifle innovation. And what we say in the book is that the response to Schumpeter's pessimism is a triangle between firms, the state and civil society, and you need the three of them. If you have only firms then the big eat the small, or incumbents prevent the entry of the new innovators. So you need something else, it could be the state, the state could regulate the incumbents, but the state can be captured by common firms. So of course, for that you could say well with the state you have separation of powers, you will have noticed you, you would say when the judiciary would control and limit the extent to which the executive power colludes with government firms, but the problem is that constitutions are incomplete contracts. And we go back to incomplete contracts there, constitutions are incomplete contracts, and they may not be enforced. And that's where civil societies are involved in civil societies there, with the media, the union with the associations with the voters, to make sure that the subcontract will be enforced. And that's why you need the triangle between firms, state and civil society, that's there. We saw that in the US, it was very important that turn out for voting. Now if we have more competition in the US, Biden could be elected, because many more people would go voting. Now you can see that Republicans in various states in the US try to limit again, and you work on gerrymandering, you know very well about that. But they are trained by others, in addition to gerrymandering to lead the votes of the poor, because they don't want to lose again. So they want to limit civil society, limiting the vote means limited civic society expression, but civic society is what got Biden there in the US. And because you have Biden, you’ve got new competition laws. So you need the triangle between firms, state and civil society, you know, that Colombia has the same Constitution as France. But in Colombia, if you're a union leader, you're very likely to be killed. That's not the case in France, because you don't have the triangle as you have in France. France, you have decades and decades of fights, to secure the right of the media, to secure the right of the union leaders and all that which you don't have in Colombia. And you need to secure this third leg, which is civil society. Another example is the civil war in the US. The Civil War in the US led to the 15th Amendment of the US Constitution to give voting rights with anybody, including African Americans, but as you know, the southern states in the US for a hundred years didn’t implement it, because they would find poll tax, or they would put crazy literacy requirements, which will make it impossible for African Americans to vote. And it took the civil rights movement in the US to finally lead to the Voting Act of the Supreme Court in 1965. That made it impossible to have the literacy test, and all that, to do anything that would prevent the African Americans to vote or still, there are other tricks, which I just mentioned before, so again, you see how the civil rights movement is what allows you to enforce the constitutional contract. You may have separation of power, you may have a well defined constitution, but without civil society this constitution remains, *speaks in French*, as we say in French, remains, you know, abstract, it doesn't go, come into reality. And that's why the triangle between firms, state, with the separation of power, and civil society, this triangle is the way to avert Schumpeter’s specifism. And that's how you can have, you know, true competition policy. That's how you can have, you know, things that will prevent conglomerates, you know, to become a barrier to new innovation, that’s when we can all allow you to have green innovation. We discuss the green, I discuss the role of firms – the ones that do the green innovation, the state does the carbon tax and what we call corporate social responsibility, that also is very important in enforcing re-innovation.

Richard Holden: I think that's very important. We're seeing, kind of, a live experiment in the United States about whether it's incomplete constitutional contract can kind of adapt to changing circumstances. And, you know, the contract theorists in both of us, sort of, you know, thinks about one of the important things as being that, you know, you need to be able to adapt to changing circumstances, but also have certainty. And I guess we'll see if the United States can adapt to their changing political circumstances. Now we'll have a big impact on that, that limb, that civil society limb of getting around the Schumpeterian paradox, as you say. One of the big debates in the economics of innovation is around whether technology, you know whether really good innovations are still happening. So I guess Robert Gordon is maybe the most notable person who's maybe described as a bit of a techno pessimist and sort of says things along the lines of, you know, computers, but kind of a relatively minor innovation in the scheme of things compared to running water and toilets and refrigeration and things like that. That's led to an interesting, I think, debate about, sort of, you know, should we be techno optimists or techno pessimists? Where do you come out on that issue?

Phillippe Aghion: I think what Gordon underestimates is the fact that the IT revolution, for example, not only improved the technology to produce goods and services, but it also improved the technology to produce ideas, and that Gordon doesn't take into account, you see what I mean? So there is a colleague of mine, Salomé Baslandze, we did have a study on that and she realised that when you take into account the contribution of the IT revolution in the production of ideas, in fact, you get much more, you overturn the the assessment of Gordon, that the IT revolution was less important than steam engine or electricity. So that's my first. The second one is that I don't believe, I think on each particular technology, you eventually run up into decreasing returns, but then you have new general purpose technologies that come all the time. So the only issue is to make sure that the new general purpose technologies will produce the growth. What may prevent them to produce the growth, the competition, is more generally the fact that industry did not adapt to the general purpose technology. That's another thing, you see what I mean, so it could be that, you know, we saw with the digital, IT and AI, well, without proper competition policy in the US, instead of becoming a boost for growth, it becomes an impediment to growth, because thanks to the IT, these terms became hegemonic and prevent innovation by others. Growth is the combined result of technologies and institutions, if the institutions don’t adapt to the technology, you won't get as much growth from the technological revolution, as you would have hoped. And so that's why you need institutions and policies to adapt to the technological revolution. That's exactly what we experienced in the US. Without proper competition policy, US will not be able to take full advantage of the IT revolutions.

Richard Holden: Very good. You had a wonderful conference recently honouring, I guess, the 30th anniversary of the publication of your famous Schumpeterian paper with Peter. And it was written several years before that, as I recall, but I think it's the 30th anniversary, recently of the publication. And there were many great economists there speaking on all the implications of your work, as you alluded to earlier, in many, many fields. One person who's not an economist, but when I heard him speak, I thought he would have made a very fine economist, had he chosen to go into our field. And that's the president of France, Emmanuel Macron. And he said, I won't make you blush here on camera, but he says some very nice, accurate, but extremely nice things about your and your contribution. I thought that was wonderful to see. But there was one particular sentence or two that he said, he spoke for quite a while, four minutes, I think. But he said one thing that really stuck with me that I thought was very important, and I'm interested in your view on how we do this. He talked about the idea of the Washington Consensus, which I guess, you know, is something that has been around for a while and it's often associated with, say, the Clinton administration and people who, you know, really delivered on some important policy changes there. But there was, I think, a sense for many years now that maybe the Washington Consensus needs updating, that it's not quite fit for purpose for exactly the challenges that the world faces now. A lot of things have changed, technology among them since, you know, the early to mid 1990s. And what President Macron said, and I'm quoting here, he said, we must build a new consensus, and I think the Schumpeterian model must be a key part of this new consensus. So that's a pretty nice endorsement from the President of France. You know, what's your take on the Washington Consensus in the virtues or otherwise, or vices that it has brought us? And what do we need in that political sphere in terms of our international institutions, the IMF, the World Bank, maybe some of these international tax arrangements that Janet Yellen and Joe Biden are helping to put in place. What's your vision, if you like, of President Macron’s new consensus?

Phillippe Aghion: So I think the Washington Consensus was very much, you know, on fashion. It's a term that John Williamston found because IMF, World Bank, and treasury, US Treasury, were thinking alike at the time. It was at the time of the transition of Eastern European countries and Soviet Union. So he was very much the idea, but also in Latin America, they were advocating stabilised, liberalised privatised. And the view was that you could go everywhere in the world, just advocate that and that would get growth going. So of course, for innovation you will need… of course, if you have an unstable economy with hyperinflation, you won't have innovation because, you know, nobody wants to innovate. If they believe that hyperinflation we eat up their profits. Liberalise, you want to liberalise entry and competition, of course. Privatise, you don't want the socialist economy, you want a capitalist economy,  where you have private sector. So that was right, I mean, in a sense those three things, of course are ingredients but are not enough. We realise that education is very important. Why is China, why did China do so well? Because they opened up their economy with education, if they have had an uneducated population, it would have led nowhere. So education is very important. We also realise that smart industrial policy is important. You see, in fact, the US did it. A lot of innovation came from the DARPA and the BARDA that I talked about. So that was not part of the Washington Consensus. Also, we realised when in Denmark, or Sweden, they put in place an active labour market policy that allows you to deal with people who lose their job and look for a new job, that helps creative destruction at the same time as it protects individuals. So the Washington Consensus didn't have anything about inequality, social mobility, education, you know, active labour market policy, they didn't have that at all. So it was very incomplete. They had three things which are important, but it needs to be completed. And we know that you can have a Washington Consensus without education, you go nowhere, without education heads and the basic public service and you go nowhere, without smart industrial policy is also very helpful. So I think it needed to be completed. And now what we learn from this COVID is two things, we learned that the social model in the US is broken, they need to completely reorder social model, there's been the work of Angus Deaton, showing that you have the sharply increasing mortality of the white non Hispanics middle age, because whenever you lose your job in the US, it destabilises your family, it creates drama, and because of the stress created by the fear of unemployment, you get opioids, you get obesity, you get people abusing sleeping pills, and mortality follows. Whereas in Denmark, as we explain in a chapter of the book, when you lose your job, the state gives you 90% of your salary for two years, helps you find a new job and retrains you. In Denmark, losing your job has no negative effect on your health, and has no effect on mortality. So, you want a system like this. You want capitalism, which is as innovative as the US capitalism, but as inclusive and productive Scandinavian capitalism. And we believe that you can get both. That's what we tell in the book. Some people believe that the lack of protection was the price to pay for innovation in the US, or that the lack of innovation was the price to pay for protection in Europe. I believe that we can have both. When you have a strict security system in Denmark, it helps creative destruction work better. And that involves growth and protection. When you have more education, you allow more people to become Einsteins, it helps the growth, but it helps growth be more inclusive. When you introduce more competition in the US, it will boost growth in the US because you will put an end to secular stagnation, but at the same time, because it allows new entrants to come in, it will make growth more inclusive. So it's not innovation or protection, you can have a model which does better on both grounds. That's the new agenda. And that goes well beyond the Washington consensus. It incorporates, of course, the Washington Consensus, but it goes way beyond the Washington consensus. 

Richard Holden: Well, that's a very important and all encompassing agenda, I can see the, sort of, you know, the perfect circle from your left wing agitating roots at the start, to your contributions throughout your career as an incredibly distinguished economist, and putting both of those things back together now. There's an agenda there for politicians and activists and policy makers. There's an agenda, rich agenda there, for I think, the economists of the future to work on people who work on empirical work, people work on theory, people who work at the intersection of political economy, it's been great to talk to you, Phillippe. You know, on a personal note, you've been very important to a number of people in the profession, a very important person in my graduate studies, and in my professional development, and in my life. And I've been very lucky in that regard. And I'll always be grateful. And you know, I think this book really translates a career of work, putting together all these incredibly important pieces of rigorous economics that are so connected to what will change the lives of people all around the world. Phillippe Aghion, thank you so much for your time.

Phillip Aghion: All the best. Thank you very much Richard. 

Ann Mossop: Thanks for listening. For more information visit centreforideas.com, and don't forget to subscribe wherever you get your podcasts.