A conversation with Robert E. Lucas, Jr., Nobel Laureate in Economics, 1995.
What do you consider to be the most important economic question facing us today?
I think the question of economic development and the incorporation of what we used to call the "third world" into the prosperous economy of Europe, English speaking countries--and Japan--is the leading question. It's going to be the issue that occupies us in the 21st century.
Your range of topics over the years has been vast: investment, labour supply, unemployment, money, econometrics, growth, and so on. Has the sequence of topics been driven by a systematic exploration of economics, or has it been driven more by particular ideas and interests that happened to have come to you at the time?
Mostly the second--certainly at the start of my career. One of the beautiful things about economics is that it's a unified field in the sense that the same theoretical apparatus is applied everywhere, so any economist (and there are a lot of people doing this besides me) can switch around from fields that look very different but use, essentially, the same kind of methods.
When I started out I was taking problems as they came my way. When I moved to Chicago in the middle 1970's I was responsible for teaching macroeconomic courses in a major department. I started thinking about macroeconomics as a whole and what were the important questions, what should be taught to graduate students--to give a more systematic approach. I got into monetary theory in a serious way, at that point in my career, because of the lectures I wanted to give. I feel free to switch gears at any point.
Is there any area in macroeconomics that you feel you haven't yet studied but you feel that you might?
I teach trade now. My students are always asking for the nominal side of trade theory--exchange rates, devaluations, and so on. I tell them I don't know anything about that so I'm not going to teach you that. If I had 20 more years left as a teacher I'd probably make it my business to learn something about it.
Of all of your contributions to economics so far, if you could only pick one, what would you most like to be remembered for?
Look, 20 years from now nobody is going to read anything I've written. I'm not doing it to obtain immortality.
Of all your contributions so far what are you most proud of?
The biggest step I ever made was I guess, the "Expectations and the Neutrality of Money" paper in 1972 (1), in the sense that this paper was for me and, (I think) for the profession, a big step from what other people had done. The economic logic of it was very much built on Phelps--the parable of an economy of islands--which wasn't my idea at all. But what Phelps got me thinking about was a general equilibrium model in which monetary shocks could induce people to work harder because of lack of information. It was a technically cool paper, it was hard to write, and it put me on the map as a theoretical economist.
On the other hand, I don't feel it's in anybody's toolkit for modelling business cycles today so, in that sense, it's already forgotten. It's an extinct creature.
It seems to me it was apart of the evolution of thinking about those issues. It played a role in releasing people from existing Keynesian views of macroeconomic modelling.
It was a step in that direction.
Academically, you started as an historian and then switched to economics. Now you're conducting research that could be classified as economic history. What do you consider to be the key methodological differences between these 2 disciplines, and what, if anything, do you think they can learn from each other?
I was never a professional historian, I started graduate school at Berkeley and, as soon as I found out what professional historians do, I dropped out. I'm an undergraduate student of history.
The lecture I'm talking about today (2) makes reference to Marx and Engels' "Communist Manifesto". I read that as an undergraduate student and I liked Marx's sense that economic theory could give us a unified way about thinking about all known societies. I thought that ambition was a noble one, and an accurate one, and I bought into it. Economics is an extremely powerful way of looking at the forces that shape any society. In that sense, all of us economists are Marxists.
Arguably, Keynesian economics has, at its heart, a policy orientation. What do you feel is the appropriate balance between positive and normative questions in macroeconomics.
Keynes' work was both positive and normative. The macroeconomics that Keynes knew was inadequate equipment for thinking about the 1930's so he was trying to improve the power of economics as an explanatory discipline, and I certainly signed on. I don't think he got it right, but I think he had his heart in the right place. The same place my heart is in.
On the balance between normative and positive, I go back and forth a bit. I love the Ramsey tax theory and I like to start out thinking about the resource allocation problem: what would be the best way resources could be allocated in this world? You make up this fictional world and you start with this--with a normative question--then think about the positive economics of it and if they differ, what stops the equilibrium allocation from attaining this first-best. I think it's a good way of thinking about economic questions. I'm always going back and forth between normative and positive economics, trying to give recommendations on how to make things work better.
On policy making it seems to be that a central message from your papers has been that serious policy analysis is quite difficult to do. For example, trying to take proper account of the Lucas critique when computing optimal policies in any non-trivial dynamic model can be quite difficult particularly in light of other difficulties such as time inconsistency. Would you agree with that, essentially that good policy making is extremely difficult? And if you agree with that what would you recommend policy makers do, considering that doing nothing is actually a policy choice?
Your question is too abstract for my taste. The Lucas critique had to do with monetary policies, mainly. The reason was that monetary policy in my opinion, (and I think its still true), a lot of its effects depend on information, on fooling people, on tricking people, on inducing them to spend more or do something different when nothing has really changed in the real situation. Those things are very hard to analyse and predict because the economics of information is a hard subject to work with and we don't know very much about it.
Now, if you just ask a question about would you rather fund the school system with government run schools, or a voucher plan, or would you rather raise revenue from capital taxation or value-added taxation those are policy questions I think we can answer them very well--I'd be happy to give advice on those. It just depends on the question and whether the tools we have match up to those questions.
Questions of economic dynamics, where expectations play a role, are just fundamentally different from questions like: Should we have a voucher plan? Once and for all changes in the rules of the game are very different from time-varying policies in a dynamic setting--and that was not appreciated in the 1960's. I think my work and Sargent's work, and so on, clarified those issues a lot. It's not a question of policy in general being impossible, or of government in general not being useful. There are a lot of policies that have been very well analysed, and we know how to do it. Fine-tuning monetary policy, in my opinion, is not one of those issues: we don't know how to do it--and we should quit trying to do it.
In your work so far there has been a preponderance of models with competitive market structures: competitive equilibrium rather than other market structures that you can imagine. There are clear analytical advantages to following this approach (it's simpler to analyse), but also clear implications about the efficiency of the equilibrium allocations considered.
Those are quite different. I think externalities of various kinds are important. You can't talk about air pollution or environmental questions without talking about differences between the market equilibrium and socially efficient allocation. But you can talk about both those issues without market power or have any large players--just a lot of guys with dirty cars and spilling crud into the air. We can analyse that with a competitive economic model, but there is no presumption that individual choices people make, and how dirty their exhaust should be, are going to coincide with what we want for our society.
But do you think that a competitive equilibrium is, roughly speaking, a reasonable description of the macroeconomies that we see?
I think that it's the working model of applied economics and I think that obviously there are some issues where market power that matters, but I think that it's the most overrated and overplayed problem. I think that if we completely eliminated all market monopoly anti-trust actions, and just permitted free entry, that would be free.
Now, if you want to discuss the question of should the government break up Microsoft or not, you can't treat Microsoft as a price-taking atom. If you want to talk about that question, you have to have a model with imperfect competition.
In terms of quantitative analysis of a macroeconomy, say dynamic general equilibrium models, you feel then, that a competitive equilibrium is a reasonable equilibrium to use for analysing the responses to shocks, and so on?
Yes I do, very much so.
An issue that some people think is important in macroeconomics, and it well may be, has to do with price setting behaviour of firms, or maybe labour unions. Now, in competitive equilibrium, individual firms and workers are price takers, not price setters. So if you want to talk about price setting behaviour, you can't do it in a competitive context.
You need something else--and here you think about Blanchard and Kiyotaki (3) applying the Dixit-Stiglitz model (4), and you get to the imperfect competitive model not so much because they're interested in the imperfect competition or market power, but because you want to have a model where some agent is actually choosing a price, that's his decision variable. To get that, we have to get outside of perfect competition. When Golosov and I studied price setting and fixed costs in a recent study (5), we used Blanchard and Kiyotaki's framework for the same reason those guys did. You have to use a model that lets you get a start on the question you're asking!
In your interview with Arjo Klamer in the early 80's (6) you mentioned (and this was a passing reference) that, "Economics" only criterion really is efficiency ...". One interpretation of this statement would be that you don't consider equity as a criterion that economists should use. Is that the interpretation you intended?
If we think about an Edgeworth box, we have a lot of efficient allocations and, by redistributing the initial endowments, we can retain efficiency and change the distribution of utilities any way we like, so it looks like there's a complete separability between economic efficiency and questions of distribution.
Now let's think about that same structure in the world of Arrow and Debreu (7) where all future goods and all present goods are part of a single space. There is one big Edgeworth box in this world that contains the entire future under all possible realisations of shocks. We can rearrange the individual endowments but then the dynamics of that system is going to run and generate an income distribution over time. If we look at that income distribution 100 years down the road and say, "there's way too much inequality", and then start rearranging the distributions all the time, to obtain equity, now we're moving away from the efficient allocations we started off with. A once and for all change in initial positions is, I agree, consistent. But the idea of an ongoing responsibility for maintaining an income distribution out of certain shape is in no way consistent with economic efficiency.
It raises the issue of a trade-off between the two: efficiency and equity. You've written some papers recently that have been talking about these two issues. Do you feel that there is a well-defined frontier between these two?
There is a well-defined frontier, and it pushes what you're calling equity-it's not a term I like. Pareto efficiency is a democratic term, it involves unanimity all of who would rather be on that side of the hill than on this side of the hill.
Equity is all of a sudden is going to be a story where I have a right to take something of yours and give it to somebody else. I'm nervous about that and don't believe that that kind of . . . I don't think equity is a fundamental value that I feel obligated to endorse, as I do Pareto optimality. (I don't want to overrate Pareto optimality in some issues, too --I'm not going to let every father choose to beat his children.)
I'm nervous about rationalising what I regard as paternalistic, or inappropriate, government policies in the name of equity. It looks easy from the Edgeworth box point of view, but I think if you reinterpret that in terms of modern dynamics it's not so easy. There are choices we have to make that can't be governed by Pareto optimality--we have to think about those issues.
It seems to me that insisting that we only recommend policy changes when they would bring about Pareto improvements is a very tough policy criterion.
Yes, me too.
But is that what you're recommending?
No not at all, but just don't pretend that there's some economic theorem that tells us that. I think that we should move away from capital taxation towards consumption taxation, and I'll argue that with you or anyone else, but as one citizen to another. A lot of things in life can't be settled by economic analysis, most public policy issues are in this category. That's why honest people can disagree. That's why we have a democracy.
Do you subscribe to the view that there is a clear Chicago school of economics?
No--absolutely not. Chicago has been the most important centre for economic research for a long time, but there's one kind of economics. There are vast differences among our faculty, and there always have been. There has never been an attempt to reinforce some party line or avoid hiring people that might violate it. It's just not a part of our ... we're proud of our achievements, of our department, of our colleagues ... but we don't say: "We don't want to not hire this guy, even though he's a great economist because he's not a 'Chicago Economist'". If you said something like that at a faculty meeting, people would just throw you out the door.
With the recent passing of Milton Friedman there have been some reflections on his legacy--not only on his academic writings, but also on his influence on policy making. What do you expect will be his most important and enduring contribution to social science?
Milton was such a protean figure. As a public figure, and a spokesman for libertarian for liberal views and liberal values, he's just played a huge role in the 20th century. It helped him that he was unquestionably a top economic scientist, but it was almost a separate role.
Within economics, Sherwin Rosen (my former colleague) used to say that Milton Freidman's important contribution was using economic theory to interpret data--it was the interpretation of how economic theory functions in empirical work. His work on consumption, money, human capital, all changed the way we do empirical work.
The way economics works, his influence is going to continue whether he gets the credit for it or somebody else does. More and more now people think of economic theory as not getting the mathematical logic of Adam Smith (and so on) clear, but as a tool for discovering empirical regularities and making sense out of them. If you like the idea of theory and empirical work coming together then you are follower of Friedman's in some sense. That's what I think.
The endogenous growth literature that burst onto the scenes in the mid 1980's was largely a Chicago product. Paul Romer was a graduate student there and you were a faculty member there. Could you describe the genesis of the idea and the link with the Chicago view of the world?
Yes, well it is, in a way--and I'll take some of the credit for that--but with people like me, and Prescott, Solow was the main guy. He was the one that wrote down models of economic growth that made some economic sense and applied them empirically. Prescott and I thought of the Solow model as the starting point for a certain way of looking at economic dynamics. So, I would call that, in some sense, an MIT contribution.
MIT was the place for growth theory when I was starting out in the 1950/60's. Once you're into that. Solow's papers showed us that you can't get sustained growth out of capital accumulation alone. If you want to get sustained growth out of this neoclassical model you're going to have to build it in. And he said: 'Look, you can do it with [e.sup.[alpha]t],--exogenous technical change.
Once you've got to that point, it's a pretty obvious step that you can ask 'where is this technical change coming from?' My first paper was a model of technical change in the Review of Economic Studies, back in the 1960's. (8) Karl Shell was working on it [for example]. Once you state the model with exogenous technical change you almost immediately want to say "okay ... let's make it into a model of endogenous technical change".
I think Romer's paper (9) was a huge step forward. It influenced me and was new in my world. It was trying to think about a new ambition, which was: the same growth model should be able to account for both rich and poor countries. We ought to be able to account for the whole damned industrial revolution in this growth model if it's any good. We don't want to have a separate type of economics for each different kind of country.
When I wrote my mechanics paper (l0) (which I do think is a really cool paper) I started from Solow's growth theory (and from the application of that theory--what he and Denison, and other people, did), and then asked: can we take this model to the whole world, and interpret data? And then you start running into serious questions. You start thinking seriously about: how can we adapt a model, and modify it in such a way as to get it to work?
I've been thinking about that, ever since.
In that paper it seems to me, to some extent, that you were throwing down a challenge to macroeconomists to think about endogenous growth issues and long-run development issues. Do you feel that challenge has been met by the profession?
I think everyone is thinking about growth.
The growth area in the 60's became a laboratory for people who wanted to learn about Pontryagin's maximum principle, Euler equations and so on. You could think of some variation of the growth model, and work it out, and you'd get a paper out of it. People weren't seriously thinking about this theory as a way of dealing with data. Solow and Denison were: they were empirical from day one. But then that literature took a very theoretical, mathematical turn and a whole generation learned a lot of good math out of it. But when that vein was mined out, people moved on to other issues and the whole field of growth theory kind of faded away--because a purely theoretical, literature-driven field is going to die.
So what Paul did, and what I did, was to try to apply this theory with more explicitly empirical ambitions. It has, kind of, taken over--whether it's to my credit or not - and millions of guys are looking at data using what I regard as serious theoretical models of growth. So--yes--it's here to stay, I think.
Think about it: being a poor American puts you above the median in 80% of the countries in the world!
Thinking about equity, though, if you wait to go through the growth process, it takes generations before you get equity.
Yes it does take time--but not as much time as it took the West to do it. America and the UK have never grown much faster than 2% per-capita, and it took the UK a long time to get it to that level--and we were the leaders. That's peanuts compared to what happened in Japan, and South Korea, and to what China's doing now. It's a long haul, but you don't have to wait that long before you get some exciting results. Just look a China now--what an exciting place!
There has been a resurgence recently of new Keynesian macroeconomic modelling with dynamic general equilibrium analysis but with sticky pricing. Do you feel that this is a fruitful avenue of research? In particular, would you recommend that current grad students invest expertise in these models?
One of your students put this really well. If your goal is to work in, and participate in the economic discussion in, any one of the world's central banks, you'd better get on top of that theory because that is becoming the academic language of monetary policy.
Is it going to be that way forever? Well, I hope not. I like the ECB's idea of trying to keep one eye on money supply the other on interest rates, but that's fading away even in the ECB.
Those guys are addressing good questions. The whole question of interest rates as the control variable as opposed to monetary aggregates: what's the connection between those two? Those questions aren't too well understood. Monetary aggregates, especially the reserves, the narrowly defined aggregates, are tiny compared to the spending force. It's a very small tail wagging a very big dog, if you're going to pin everything on control.
There a lot of practical problems in monetary policy that only the new Keynesians are addressing at all. So they fill the vacuum--and I'm not going to fault them--vacuums need to be filled!
As economic theory, I think their standards are low. They are rather loosely related, or unrelated, to the whole rest of the economics profession --it's a little like the old Keynesian days--they're off in a world of their own.
It's an awkward situation, but I don't know what to do about it, and I'm too old to address it. Somebody else is going to have to do something about it. It's kind of shocking theory--that's a fact.
You're arguing that, in the absence of good theory, you need some theory?
Central banks have to make decisions--they have to have staff--I don't know. I don't think the bankers pay too much attention to it anyway.
Another approach that has gained ground recently is search theory. You were one of the early contributors in this field, in your work with Ed Prescott. (11) Recent contributions have blended labour market search models with money market search models--I'm referring to Randy Wright's work with Berentsen and Menzio (12) in particular--where search money and search labour and generating equilibrium relationships between unemployment and inflation. Could you give me your view of the progress being made there?
I have doubts that that approach is getting anywhere. It's literature-driven, it's remote from empirical work, and the empirical work they're doing isn't very good.
I think that the original Kiyotaki-Wright (13) work, thinking about the fundamentals of money, was really cool. I think the search models of employment are really cool, but to put it all together and to put in some cash-in-advance-type model, with these night markets and day markets--what are we doing?
If Randy Wright were here, my guess is that he would respond by saying that the project is to build a model with frictions--to recognise frictions at the start and build a model of macroeconomics with that as a foundation.
But they quit on that job! Now they have some kind of utility function and some market where people meet and clear everything, Arrow-Debreustyle, every now and then. So, they're not addressing the issue--it's not a head-on address.
It's too bad because one of the problems with these models of monetary non-neutralities, especially the expectations-based models like mine, is that they don't last very long. They don't have any persistent effects. So, maybe, in a Kiyotaki-Wright type of world, where people are never seeing the whole picture you're forever bouncing around making these monetary trades, a monetary shock might last for a long time--that would be exciting! But Randy says: "that's too hard, so let's get together and clear at the end of every day". You're punting. The model has the potential for letting us think about persistent effects of monetary shocks, and then you're giving up on that potential.
So you think the real benefit from the search approach is going after the persistence of monetary shocks?
That's a potential that has been in the air since that start of these models.
Neil Wallace had a paper where he sort of looked at that issue. It involves studying the transition dynamics in these models, which is really hard, but it might be interesting too. You can't do it by just looking at steady states. It's a very, very hard problem to think about those transition dynamics.
There are people that have models where the distribution takes a while to settle down to a steady state, so that's just done numerically.
Okay--that's cool with me. Who's done that?
Miquel Molico. (14)
Okay, well, if he's doing that, then I'm interested.
The year is 2008 and 50 years has passed since the publication of Bill Phillips' 'Phillips curve' paper (15). Arguably the most important result of your 1972 paper was to demonstrate that Phillips curves may exist as statistical relationships but not as useable trade-offs for policymaking. Still, though, Phillips curves persist at being at the heart of macroeconomic policy discussions. Would you care to comment on the legacy of Bill Phillips 'paper?
He would be on my side of that issue. Somebody found a quotation (I think it was Chris Pissarides) where he anticipated the Lucas critique in a very clear way! Phillips was a really good, pioneering, theorist in thinking about dynamic systems.
But we see Phillips curves discussed everywhere--particularly at central banks. Do you feel that as a theoretical construct, or a mathematical construct, that it's a useful device or, rather, that it's more of a distraction ?
Well it's not really a device. Here's what I think: the issue, and this is where a Phillips curve comes in, is whether monetary shocks have some kind of important effect on the real economy. The monetary models that are the easiest to write down, even quite sophisticated ones, just say: no, they're not going to have any systematic effects. Yet a lot of people, and I would include myself here too, think we can see instances where monetary shocks do have real effects.
It's hard to tell--these are issues where honest people disagree. Nancy [Stokey] was arguing this morning in this way--which I think was a good point--if you have a monetary glitch (like the crisis in the late 1990s in the Asian countries) this can trigger all kinds of other effects, freezing capital flows or nationalising, doing something with bank regulation, and so on. So, then, if something follows from that, is it because of pure monetary effects or because of real actions that were taken? It's hard to sort that out.
If you think that money has non-neutralities then you're in the market for theories that let you think about that, and the theory that's on the shelf doesn't. So now Randy Wright and those guys have an inflation tax. Well, we've known about the inflation tax for longer than we've had Phillips!
But that's not what we're talking about. We're talking about short run shocks in spending and demand and that's what we think happened in Indonesia in 1997--that's not an inflation tax issue it's a contraction, some kind of monetary contraction. So we need to be able to think about that and, in some sense, the Phillips curve is a label for non-neutrality in money. So, people at the central banks are talking about something important. I'd like to be talking about too, if I knew how.
You feel that, over the years, you've learned that the monetary shocks are not very important for recessions.
That's my view now. It's really close to Kydland and Prescott (16). But a lot of evidence has piled up, since 1982, that has pushed people--at least it's pushed me--away from the Friedman view.
If you were young again, say, 18, would you choose to become an economist again, now its 2008?
Well it's pretty hard to say isn't it? I like thinking about social policy, about history, about the workings of society. I always have. I got drawn to that before I even knew there was such a thing as economics really, in courses I had in school, and so on.
Like Marx and Engels, the questions those guys were asking, or even thinking about, that's what I wanted to do. I think, if was 18, I'd still have that view.
I was also drawn to the idea of mathematical models, so if I put those 2 things together, where else can I go? It seems inevitable, in that sense. Within economics there are a lot of different things to work on. I don't think I'd want to be an historian, a pure and simple literary historian. I'm not comfortable with that. I wouldn't be able to do it.
Do you feel that you don't understand anything until you have modelled it mathematically?
It's my way of getting into a problem. I'll just start writing down a model, and a lot of the time the model will be terrible, but you try and fix it up, and so on, and it feels like thinking about things in a serious way.
What else could I do when I was 18? I wasn't a good enough baseball player to be a starter with the Chicago White Sox!
Thanks for that.
(1) Lucas, R. E., (1972) "Expectations and the Neutrality of Money", Journal of Economic Theory, 4, 103-124.
(2) "The Industrial Revolution and the Macroeconomics of Ricardo and Marx", public lecture delivered at the University of Melbourne, March 26, 2008.
(3) Blanchard, O., and N. Kiyotaki, (1987), "Monopolistic Competition and the Effects of Aggregate Demand", American Economic Review, 77, 647-666.
(4) Dixit, A., and J. Stiglitz, (1977), "Monopolistic Competition and Optimum Product Diversity", American Economic Review, 67, 297-308.
(5) Golosov, M., and R. E. Lucas, (2007), "Menu Costs and Phillips Curves", Journal of Political Economy, 115, 171-199.
(6) Klamer, A., (1984), Conversations with Economists, Rowman & Allenheld Publishers, New Jersey.
(7) Arrow, K., (1953), "Le role des valeurs boursieres pour la repartition la meilleurs des risques", Econometric, Paris: Centre National de la Recherche Scientifique [Translated as "The Role of Securities in the Optimal Allocation of Risk-Bearing", Review of Economic Studies, 31, 91-96. Debreu, G., (1959), Theory of Value, Cowles Foundation monograph.
(8) Lucas, R. E., (1967) "Tests of a Capital-Theoretic Model of Technical Change", Review of Economic Studies, 34, 175-189.
(9) Romer, P., (1986) "Increasing Returns and Long-Run Growth", Journal of Political Economy, 94, 1002-1037.
(10) Lucas, R. E., (1988) "On the Mechanics of Economic Development", Journal of Monetary Economics, 22, 3-42.
(11) Lucas, R., and E. Prescott (1974) "Equilibrium Search and Unemployment", Journal of Economic Theory, 7, 188-209.
(12) Berentsen, A., G. Menzio, and R. Wright (2008), "Inflation and Unemployment in the Long-Run", NBER Working paper #13924.
(13) Kiyotaki, N., and R. Wright (1989) "On Money as a Medium of Exchange", Journal of Political Economy, 97, 927-954.
(14) Molico, M., (2006), "The Distribution of Money and Prices in Search Equilibrium", International Economic Review, 47, 701-722.
(15) Phillips, A. W., (1958), "The Relationship Between Unemployment and the Rate of Change of Money Wages in the United Kingdom", 1861-1957", Economica, 25, 283-299.
(16) Kydland, F. and E. Prescott, (1982), "Time to Build and Aggregate Fluctuations", Econometrica, 50, 1345-1370.
|Printer friendly Cite/link Email Feedback|
|Author:||King, Ian P.|
|Publication:||New Zealand Economic Papers|
|Date:||Jun 1, 2008|
|Previous Article:||Omitted productivity data: why haven't economic reforms increased productivity growth in New Zealand? (1).|
|Next Article:||The impact of recent changes in family assistance on partnering and women's employment in New Zealand.|