Philippe Aghion is the Robert C. Waggoner Professor of Economics at Harvard University. He received his PhD in Economics from Harvard in 1987. His research interests include contract theory, industrial organization, growth theory, economics of innovation and education. He is the author of 8 academic books among which “The Economics of Growth” (with Peter Howitt).
Interviewed by Georgios Petropoulos PhD candidate, TSE
Looking at your contributions and research activities I can tell that you are interested in many fields that at first glance do not seem to be closely related. On the one hand, mechanism design and contract theory; on the other hand, economic growth and economics of innovation. Of course I should not forget to mention political, experimental economics and economics of education. It is a very interesting mix. What is your motivation to work in so many different fields?
I was trained as an applied theorist. My thesis was on industrial organization and contract theory. I was particularly interested in the applications of incomplete contracts. But, basically what I tried to do was to incorporate growth theory with industrial organization and firms. I worked together with Peter Howitt on that. The idea was to develop a growth model where you look at growth from the point of view of firms and entrepreneurs. This did not exist before. With this model we could look at growth policy design: the effect of competition, education policy, finance, R&D policy, environmental policy, macro policy which I am doing recently. So, the idea was to have a framework with which we could talk about growth policy design but always to look at it from the point of view of the effects it has on firms, on entry, on turnover, on the nature of innovations chosen by firms and to be able to test those models. Initially, my first phase was to do theory. I worked with Peter developing various extensions of our initial model. But then I moved to the empirical part. First, with Richard Blundell at University College London where we did the work on competition and growth which was based on firm level data. That was a very new type of econometrics because before, the econometrics of growth was cross country. In contrast, we followed a more microeconomic approach using firm level and sectoral level data. That is what I have been trying to do so far.
So, basically, your main contribution is that you developed a link between industrial organization and growth theory that did not exist before...
Exactly! There wasn’t any link between industrial organization and growth theory before.
This brings me to my second question. As the field of economics grows and expands with new developing areas, specialization is necessary. However, in the last years we observe that the borders between different areas are not so distinct. Do you think that in the effort for new substantial contributions we gradually move from the time of specialization to the time of generalization? Do you feel that economists should be specialized in a particular field of study or they should have broader horizons in order to be able to identify links between different fields in economics?
I think that it is a little bit of both. You cannot be too broad. If you are too broad, you are too thin. So, you need to spend some time on a particular topic. To be trained as a researcher you have to write papers that are publishable in the best journals. So, you need to go intensive! You cannot just go extensive. But, I think that it is good to have a PhD training that acquaints you with some other secondary fields. In my case, it was useful that I learnt contract theory, industrial organization and macroeconomics. But, it is true that I did my PhD more in industrial organization and theory of contracts at that time. I think that specialization, especially in the PhD phase, is necessary.
Let’s begin covering your research areas and let’s start with economics of education. In a recent study you showed how the governance structure of the EU and the US universities affects their research output. Why do you think that the top European universities cannot compete with the top US universities in terms of research and job market placements? Is it a matter of funding, governance...?
I think it is actually both. In fact, part of my growth program has been devoted to education and its interaction with growth. The first thing we realized was that when you are an economy closer to the technological frontier, where growth relies more on innovation, especially on frontier innovation, not just on imitation, you need to have really good graduate schools. This is very important. If you are a catching up economy, it is less important. For a catching up economy, primary, secondary and good undergraduate programs are enough. You do not need to have necessarily very good graduate schools. But, when you become a frontier innovative economy, you need to have good graduate schools. That was one of the first conclusions of our studies. So, education is very much linked to the growth policy, to the growth work I have done.
Then, the idea was to say that you should not only invest more in higher and graduate education but also govern differently. And this is true not only for education but also for industrial and macro policy. The way you govern the allocation of funding is very important. Further, we saw that it was important to invest more not only in graduate education but also in more autonomous universities. So, autonomy is also important. There is a complementarity between autonomy and funding and the fact that you compete for grants. The fact that universities compete for grants is very important. Afterwards, we looked at other governance aspects and we saw that universities with boards of directors which have a significant fraction of external members seem to work well! So, we carried out a cross-country analysis and we examined different structures of governance. The board of directors may vary; you can have universities which are public, others which are private, universities with tuition fees or without tuition fees. But usually a governance structure with two main bodies, an academic senate composed of professors and a board of directors with a significant fraction of external members, works well. This is something we have been trying to push in France for example. I know that in Germany there are similar efforts towards such a governance structure in universities.
Next stop, economics of innovation. Your work showing that the relationship between product market competition and innovation is inverted-U is quite influential, popular and criticized to some extent. What do you think are the implications of your work for the possible ways we could use competition to promote innovation? What do you think are the main reasons for the fact that Lisbon strategy does not appear to be as successful as it was expected? Do instruments such as liberalization and privatization really work?
It is not enough to liberalize to have innovation happening. It is true that competition is a driving force of innovation because you innovate to escape competition. At first this was not obvious. Initially, there was a view that, since you innovate to get some monopoly rents, competition is a bad thing because it reduces these rents and therefore discourages innovation. This is the Schumpeterian effect. The downward sloping part of the inverted-U relationship is driven by this Schumpeterian effect. But, you have another effect which is called the “escape from competition” effect. It is that you innovate in order to escape competition with other firms. This effect drives the upward sloping part of the inverted-U curve. In fact, in most sectors that are frontier, it is the upward sloping part that matters. The backward sectors are more subject to the Schumpeterian effect. But, the more frontier sectors or frontier countries are more subject to the “escape from competition” effect. So, it’s true that liberalizing helps, but it is not enough. You also need to have knowledge as well as a good higher educational system – this is very important. This is something that I have looked at in particular studies in various countries. Liberalization is good for the sectors that are frontier, but it is very bad for the sectors that are behind. So, for liberalization to work well, it is required to have complementary policies that train workers and help them to move from lagging to frontier sectors as well as policies that promote research and good functioning and organization of universities. So, it is a set of policies that accompany liberalization that determines whether liberalization works well or not. Thus, in countries where liberalization did not work so well, one has to look at whether such complementary policies were set in place or not, and in most cases they were not!
There are situations where liberalization does not always do the right thing. For example, Spain and other countries in the Southern Europe liberalized and invested massively in real estate and non- tradable goods. And this is not good! So, you may need the government to play a role to induce innovation and to direct it to sectors that are related to tradable goods. You need what I call an industrial policy, but it has to be a new kind of industrial policy, the one that is competition friendly. So, the role of the government is also very important. In the instances where there is too much investment in real estate, non-tradable goods, polluting innovations, there is a role for a smart state, a smart government policy, probably monitored by a European institution to make sure that competition is preserved. This is because the big danger is that industrial or sectoral policy will kill competition policy, which we do not want.
So, from what you say, I understand that there is a long way for the realization of the goals of the Lisbon strategy.
Yes, because, in fact, initially, the Lisbon strategy adopted the view that you should give R&D subsidies and it works. Then, we realized that you needed also liberalization. And now people realize that you need on the top of that a smart state. Many countries that liberalized did not have a smart state. They followed the Washington consensus which says: liberalize, privatize and stabilize; but a smart state is more than that.
Influenced by the experience and the political intensity of the recent French presidential election, I would like to ask you the following question. What do you think about the role of the mass media in democracies? Do you identify any conflict of interest in their function? Do you think that stricter regulation needs to be imposed in the framework they operate?
Yes, absolutely. The big problem for a country like France is that you need to modernize the state. For example, in the Northern Europe or in Germany, the government is transparent. They have very high standards in terms of fighting corruption and nepotism. For example, in Sweden, a minister had to resign because they found out that she had bought a Toblerone chocolate with the credit card of the ministry. Just for a Toblerone she had to resign!! In France, we are very far from that. When I talked about a smart state, I said that the state has to select priorities where to invest in favor of innovation. It can be horizontal targeting: R&D, education, smart industrial policy or small business act, and also vertical targeting: the state favors sectors that are growth enhancing. But, if you do not have transparency, targeting may turn into corruption because then you will give to your friends, you will not have objective criteria to allocate investments in a growth enhancing way... So, it is very important to have independent media. In France, the big problem is that the most of the big media, the big newspapers or television, are in the hands either of the state (the head of the television is named by the president of the Republic himself now; this was not the case in the past) or of groups that benefit from public procurement contracts. So, these groups make sure that they do not displease the presidents because otherwise they stop having these public procurement contracts. We should find ways to make sure that such groups do not influence political actions. The other thing is about justice. For example, the executive power intervenes too much in the choice of judges. That should be much more independent. Moreover, while most of the Northern European countries have developed very good institutions to evaluate public policies and the effects of laws, while in the US there exists the Congressional Budget Office, in France there are no commissions or institutions for evaluating public policies and laws. So, all that part, you see, is missing in France very much. France needs to upgrade itself to become immune to nepotism.
My final question is motivated by the current Eurozone crisis. Nowadays, in Europe we experience a very sharp recession which led to fiscal imbalances and stagnation. What do you think is the impact of the imposed austerity measures on economic growth? Is there really a trade-off between austerity and growth? Should we go for austerity policies or policies that promote economic growth? How can we recover faster from this stagnation?
I think that the choice is not between austerity and growth. I think one actually needs both. One needs to combine fiscal and budgetary discipline with growth enhancing policies. Both things are needed, not the one or the other. So, you can no longer implement Keynesian policies where you stimulate public spending everywhere to stimulate demand and then stimulate growth. Those days are gone! There are areas where the state can save money, but there are areas in which the state should invest. Then, there are sectors where you can save money. For example, with the progress in the ICT revolution, monitoring costs have gone down. Thanks to the ICT and computers, the information revolution, there are a number of services the operating cost of which can be reduced. So, it is important to have good governments to make sure that the money spent is well used. That’s what a smart state is. It is not either austerity or growth. Also, it is important austerity to be well shared. In some countries people refuse austerity because they think some people get away with it. For example, in countries where people do not pay taxes, how can you implement an austerity program when you know that some people get away with everything? It is very hard! For this reason, it is very important to have a transparent and fair fiscal system. You need some austerity somewhere, but you need to invest in growth somewhere else. And you need to combine both. In fact, I very much believe in the triangle of budgetary discipline, growth and social justice. I very much believe that these three things work together. I think this is what, for example, Mario Monti is trying to do in Italy with a very narrow political margin that he has. He knows that he has to ask everyone for a sacrifice so that they feel that everyone contributes in a fair way. He knows that fiscal and budgetary discipline is needed not only to get good ratings but also to be able to conduct countercyclical macroeconomic policy. Otherwise, it is very hard to conduct such policies as it is very hard to borrow in recessions. Growth is obviously also important because if you do not have growth, you cannot have fiscal discipline in the long run. So, those three sides of the triangle are really complementary.