Discover more from The Capitalism and Freedom in the 21st Century Podcast
Episode 34. Greg Mankiw (Harvard Economics Professor) on New Keynesian Macroeconomics, Growth and Economic Policy
Podcast Interview Transcript
Greg Mankiw (Harvard Economics Professor) joins the podcast to discuss his career as a economist, New Keynesian models of the economy and their usefulness to central banks, economic growth, Greg’s time running the White House Council of Economic Advisors under George W. Bush, as well as Greg’s advocacy for carbon taxes. Listen to or watch the full Capitalism and Freedom in the 21st Century Podcast episode with Greg, which hosted at the Hoover Institution Economic Policy Working Group.
Jon Hartley: This is the Capitalism and Freedom in the 21st Century Podcast, an official podcast of the Hoover Economic Policy Working Group, where we talk about economics, markets, and public policy. I'm Jon Hartley, your host, today, my guest is Greg Mankiw, who is the Robert M. Beren Professor of Economics at Harvard University.
Greg is one of the leading macroeconomists of our time and one of the founders of New Keynesian economic modeling. Greg also served as the Chair of the Council of Economic Advisors under President George W. Bush, and is the author of the best selling textbook, Principles of Economics. Thanks so much for joining us today, Greg.
Gregory Mankiw: Thank you, Jon, it's nice to be with you.
Jon Hartley: Greg, I wanna start with talking a bit about your upbringing. You grew up in New Jersey, where you went to The Pingry School for high school. Then you attended Princeton for undergrad, where one of your classmates was David Romer, one of your longtime co-authors. And then you went on to do your PhD in economics at MIT. How did you first get interested in economics?
Gregory Mankiw: Well, when I started at Princeton, I thought I was gonna be a math major because I kind of liked math, I thought I was good at it.
And what I learned pretty quickly was that I was good in the sense of being the best math student in my high school. Once you sort of go to a top math department like Princeton, the level of talent there is far in math is far beyond what I had.
And math became increasingly abstract. And even though it was my first declared major, I didn't stick with it and eventually switched economics. The thing that got me to economics was actually a friend in my freshman year. She was taking an economics course her first semester, freshman year. And I didn't really know very much what economics was, I knew sort of vaguely, but not very specifically.
And she would come back and tell me what she was learning in her class was the principles of economics, microeconomics with Harvey Rosen, who's a great teacher. And I thought it was more interesting than anything I was learning. I sort of looked at her textbook, which was the Lipsey-Steiner book, and I started reading it, it seemed really interesting.
And so the next semester, I took both principles of macro and principles of micro, the principles of macro with Burton Malkiel, principles of micro with Harvey Rosen.
Jon Hartley: Wow.
Gregory Mankiw: And they were both great teachers. And I approached Harvey to see if I could get a summer job for it with him.
I knew also economics, I'd only taken the principals course, but I did know some Fortran programming, and he needed somebody who could do Fortran programming for him. And so I basically spent a summer working for him, and he had to teach me all the economics I needed to know for the job.
And so I learned a tremendous amount that summer. And he's really the one who sort of got me fundamentally interested in economics. It was a great pleasure that years later, I got to work with him on the Council of Economic Advisors, where we were both members. But it really was Harvey who got me interested at the Princeton.
I basically drifted toward macroeconomics as an interest, and I did my senior thesis with Alan Blinder. And he was the other sort of professor at Princeton that was extremely influential in my career development. You mentioned David Romer, it was actually an interesting group of future economics professors in my class.
David Romer was one of them, Danny Quah was another, Jim Rauch, who's now UC San Diego, is another. Larry Ossebel was a friend of mine, we were the same eating club. He wasn't an economics major then, he was a math major, but I knew Larry. So all of us ended up going into economics, and so going to get a PhD in economics seemed like kind of a normal thing to do because I had a bunch of friends who were doing the same thing.
I don't think it's as normal today among my Harvard students, I think it's much rarer. But at the time, it seemed like a very normal thing to do, and the great teachers I had at Princeton were a large part of that.
Jon Hartley: That's Fascinating. And Princeton is known for being this really big hotspot for macro, certainly in the 90s, but I'm sure many of the precursors that were there were forming in the 1980s as well.
A lot of those, like Paul Krugman, Mike Woodford, and others, who sort of started building a lot of the New Keynesian framework that you yourself have been a huge contributor, well, at Harvard. And I'm sure there's many interlinkages there.
Gregory Mankiw: Those people were at Princeton later, actually, they weren't there when I was a student.
I was a student there in the 70s, I graduated 1980. But the one person who was there was a grad student named Julio Rotemberg. And so Julio was sort of doing his early work, his dissertation of some of the early New Keynesian stuff.
Jon Hartley: Fascinating, fascinating. So what was graduate school at MIT, Econ, like in the 1980s?
Like your advisor there was Stan Fisher. You had, I mean, over the course of your career, you've had a many famous students. Stan Fisher has had many, many famous students over his career as well. You've also taught act ten for many years. I'm curious, what was that experience at MIT econ like, and you're sort of interested in advising and mentoring students.
Was that something that you kind of something that grew on you while you were being advised in grad school?
Gregory Mankiw: Well, there were two people at MIT who were particularly influential for me and my cohort. One was Larry Summers, who was there before he is an assistant professor before he joined the Harvard faculty.
Olivier Blanchard was not there at that point. Stan Fisher was the other person who was very influential in my cohort. My cohort had a lot of people became macroeconomists. Myself, there was Mike Woodford, Matthew Shapiro, Bob Barsky, Larry Ball, Steve Zeldes, and I'm sure I'm missing a couple others.
So there's a lot of people went on to quite prominent careers in macroeconomics, and it was a natural to be interested in macroeconomics because the faculty was obviously fantastic. I mean, Stan was really a role model for a lot of us, but also the economy really demanded interest in macroeconomics.
So this is like the heyday of high inflation, and Paul Volcker was just getting appointed and the whole Volcker experiment was going on. And so there was natural interest among my cohort and the questions of macroeconomics. And Stan was a great person to be academically rigorous, but also very focused on policy and current events as well.
And I think that combination was really extremely compelling. And in that group, I developed co authorial relationships with a bunch of my colleagues have written papers with Matthew Shapiro and Bob Barsky and Larry Ball and Steve Zeldes. Oddly, I’ve never written a paper with Mike Woodford, even though he has a tremendous number of common interests.
It's just never kind of, it's never worked out. But I remember spending many hours talking to Mike, and he's a close friend. And it was clear, by the way, that he was one of the stars of our class, even from, even from the beginning. He won his MacArthur Fellowship while he was a grad student, which is kind of extraordinary.
Jon Hartley: That's unbelievable.
Gregory Mankiw: It is, this is early in the MacArthur. I don't think the MacArthur has given the award to people quite that young now, I think. But early on, they were trying to get people who were at the early stages of their career. And Mike quickly established himself as a star among the faculty.
Jon Hartley: Well, that's amazing. Well, I want to talk to you a little bit about how your ideas developed while maybe you're a grad student and as a professor at Harvard. I really just want to really get into the history of New Keynesian DSGE modeling, which I think some of your most well known work and major contributions lie.
How did a lot of those early New Keynesian papers come about, the three-equation model? I know there was a lot of early work that was done on sticky prices and sticky wages, people like John Taylor and Ned Phelps, and you're a co-author with Ned Phelps on some of those early papers.
You also wrote a number of the early papers as well, on menu costs and sticky prices. I'm just curious how that thinking came about, because I've talked to certain people in some of those early contributors. And some of them kind of say, well, we had this RBC revolution in the 1970s and 80s, rational expectations revolution, full information, rational expectations.
That was sort of a response to the static IS-LM framework to incorporate micro foundations, micro-founded rational behavior, to also think about intertemporal decision making, that culminated in this RBC model. Think Sargent, think Prescott and others, but my understanding is that a lot of people weren't happy with that framework.
It was also, some people argue, in response to the great inflation of the 70s and ISLM not being able to account for things like stagflation, inflation going up while unemployment going up at the same time. My understanding is that there were a lot of people who weren't happy with the fact that Real Business Cycle models didn't have room for policy.
And so my understanding is that some people just said, well, what's great about some nominal rigidities is that, well, now you have room for policy, and that's why we threw it in there. I'm curious, was there an empirical basis for including sticky wages and sticky prices, and Bewley later sort of, I think, tried to do some empirical work.
But I'm curious what the history was there with why people felt it was important to include sticky wages or sticky prices, or sticky information, which is, I think, an amazing idea, and perhaps one that's not appreciated enough. But why was it important then to include those in our macroeconomic models?
And did that lead to a better description of how business cycles work? And do you think that's still a relevant description?
Gregory Mankiw: I remember at some point in graduate school Stan Fisher giving a lecture, and he said, why do we believe that money matters, that it has real effects?
And he said, two reasons, Paul Volcker and Friedman and Schwartz. So, I always took it as established fact that monetary policy had real effects, I think very few economists today would doubt that. Very few economists say, money is completely neutral, what the Fed does doesn't have any impact on the real economy, even though we don't fully understand always the impact it's going to have.
So I always took some degree of stickiness as necessary to understand how the business cycle works. And I thought Real Business Cycle theory was really a turn in the wrong direction. My own personal interest in this literature dates back even before Real Business Cycle theory was around. But really it dates back to me writing my undergraduate thesis under Alan Blinder.
I remember taking a graduate macro course from Alan, and he was explaining the standard Keynesian mechanism at the time. And at the time, the big emphasis was on sticky wages. The nominal wages were sticky. And that's the reason why monetary policy is not neutral. And the mechanism was basically, when aggregate demand increases, prices increase.
If nominal wages are sticky, real wages have to go down, and it's lower real wages that increase employment. And that's pretty explicit in Keynes’s General Theory, the countercyclical real wage. And I remember talking to Blinder about that, and he sort of admitted to me that, that's, yeah, I know that's true in the model, but that's not really true of the data.
The real wage is not countercyclical. And I said, wow, that's a pretty damning observation to me. And that's what I wrote a lot of my senior thesis about. Actually, that senior thesis became a, part of it became an article that I co-authored with Alan Blinder in the Journal of Monetary Economics.
It's one of my earliest articles, not one of my most important, but one of the first things I published. But it was that idea that real wages weren't countercyclical, that said, you have to start thinking about not only sticky wages, I have to start thinking about sticky prices.
And if I'm gonna start thinking about sticky prices, you have to have firms that are not competitive, that are price setters, not price takers. Because if you're going to think about the incentives that firms have to adjust prices, you can't have them being price takers. And it was that that got me to write my small menu cost paper, which I think was my first very successful paper in the sense of garnering a significant number of citations and getting some notice.
I wrote that actually after, I think after my first year of grad school. And so that was what did think. But then I think the motivation was largely conservative, not conservative in a political sense, but conservative in intellectual sense. Lucas and Prescott were all about, this Keynesian stuff's a failure, let's throw it out, let's start from scratch.
They were very derogatory to much of the literature that had come before them. And I remember always thinking, that seems a little too radical for me. My temperament was always like, okay, not everything's right with these models, but there's some truth to the models too. And just throwing everything out and starting from scratch may seem like fun from the standpoint of intellectual revolution, but it seems a little too immersed.
And what the New Keynesian literature tried to do is say, okay, there's some truth in this stuff that Tobin and Modigliani and that generation was saying. But how can we take Lucas seriously without throwing that previous generation's ideas out the window? And I think that was a large part of the motivation of the New Keynesian literature.
Jon Hartley: That's fascinating, any thoughts on, I feel like now there's, you're absolutely right in a great observation that Keynes’s General Theory was more focused on sicky wages than, say, sticky prices. And I feel I feel like now we're slowly maybe shifting back, or at least in terms of what papers are using, I think slightly less focus on sticky prices.
When you have companies like Amazon that can adjust prices basically in matters of minutes or less or in seconds. Where you have dynamic surge of pricing in Uber and elsewhere, I guess prices are becoming less sticky. And there's work from folks like Alberto Cavallo and others that suggest that with a bigger online economy, prices, prices are becoming less sticky.
So maybe wages matter more or wage stickiness matters more. Maybe in the housing market, rental contracts, those are renewable only once a year, and there's a lot of frictions there as well. I'm curious, your sticky information paper with Ricardo Reis is also one of your most highly cited papers.
I'm curious what your thoughts are on sort of the legacy of sticky information. I think it's such a great theory and informational frictions are so important. I mean, they're one of the reasons why, for example, the work done by Joe Stiglitz and others arguing why the first welfare theorem doesn't work so well.
And similarly, you sort of apply that idea within business cycles. But I guess there's the problem, which is like, how do you measure information? And that's a very difficult thing to do, so I think it's a great theory, but a difficult empirical kind of thing. And Ricardo Reis has done more follow up work.
I'm curious what you think about the legacy of that paper, I feel like is very early also behavioral economics, too. And a lot of people are sort of interested in behavioral macroeconomics. There was interest in behavioral macroeconomics, I think, 20 years ago. I think there even was a behavioral macroeconomics NBER group of some form 20 years ago.
And now there sort of is like a NBER Behavioral Macro summer camp now and a working group again, but I'm curious. I also feel like now there's a bit of retreat from behavioral economics, like we had sort of an apex in the mid-2010s. But there's since been questions about Dan Ariely’s work, and I guess some questions about replication and things like that.
But I'm curious what you think about the legacy of that paper and informational frictions.
Gregory Mankiw: Well, that particular paper was motivated by what I thought was a flaw in the prevailing paradigm. I mean, if you look at most of the work in New Keynesian economics. The model of price stickiness in the Phillips curve that people turn to is usually some version of what's called the New Keynesian Phillips curve or the Calvo model.
I mean, Julio Rotemberg had a version of this, too, slightly different. But the essence of the Calvo model is that prices are set in this forward looking way. And as a result, inflation depends on what people expect inflation in the future to be. And that was introduced one interpretation of Friedman's natural hypothesis.
The problem with that is that in that model, inflation can jump in response to monetary policy. So because inflation depends on expected inflation, and that expectation can jump immediately. When there's a monetary shock, inflation can adjust literally instantaneously to the monetary shock. So the price level is sticky in that model, but the inflation rate is not the inflation rate is a jump variable.
And it seemed to me that in the data, people didn't seem to think that was the case inflation was deemed as very inertial. So the question is, how can you get inertial inflation and have some sort of model of price adjustment the way people often do it? If you look at a lot of the applied work, they will often take the Calvo model.
And then pend it with some ad hoc assumption, like indexation of some prices to past prices. So they basically built in an ad hoc way, inflation inertia. But that's nothing really in the model, that's just something they'd added on. So I was trying to think, okay, well, what might be generating this inflation inertia, other than just sort of putting it in an ad hoc way?
And that was what motivated me to start thinking about sticky information. You're absolutely right about the connection, by the way, to the behavioral economics literature. This idea that information may be slow come in. I saw a paper by David Laibson and Xavier Gabaix that was applying that idea to the consumption problem.
And I remember after seeing their present, their paper, I said, look, let's try to apply this to the price setting problem. That's what led to the sticky information model. In this model, expectations matter, but it's different expectations. In the Calvo model, what matters is current expectations of future inflation.
And the sticky information model, what matters is past expectations of current inflation. And because past expectations are already given, that's going to generate a certain degree of inertia that you're not going to get in the capital model. It was that motivated the model. How influential has it been? It's widely cited, but I don't think it's become the go to workhorse for price adjustment in the literature in that sense, I failed.
But I think I've sort of, I still believe basically the basic critique of the Calvo model.
Jon Hartley: Right, I'm going to talk just a bit about the legacy of New Keynesian DSGE models, and I guess just macro as a field more broadly. So while I think today New Keynesian DSGE are still being extended in some forms, like in the HANK literature, the heterogeneous agent New Keynesian literature.
There's several economists out there, Ben Moll, Adrien Auclert, their co-authors, and many others. And DSGE models are being simulated by central banks or maintained by central banks. But I think there's a lot of questions out there about the validity of things like the Phillips Curve. And really just whether these models are actually that useful, especially when they become more complicated.
We have certain things like, I think the Taylor rule, which empirically fits the US data, for example, fairly well. And I think a lot of central banks around the world use Taylor rules in some way as a guideline. I think that's actually one of the more successful components, if you will, of the sort of New Keynesian framework.
But in general, I think most central bankers kind of have something more simple like an ISLM model in the back of their head, rather than paying attention to every read off a DSGE model. Having talked to a lot of central bankers, this seems to be the case, while the staff does continue to crank out DSGE models.
They're producing forecasts from DSGE, but they're also producing forecasts from VARs, vector auto regressions. I think often the vector auto regressions (VARs) generally produce better forecasts, and then central bankers use that to form their analyses. So I'm curious what you think about that. I mean, there's also this sort of general decline of macroeconomics in economics departments giving way to applied microeconomists who argue that macro doesn't really pass a good identification smell test.
I'm curious, what do you think about the legacy of new DSGE modeling today in academia and elsewhere.
Gregory Mankiw: I mean, on the issue of micro versus macro, I have long thought that microeconomics has the better answers, but macroeconomists have the better questions. And so I don't think macroeconomics is ever gonna die because the questions that macroeconomists address are just too important, even if we're not gonna have the clean, natural experiments that micro people love.
On the general shortcomings of these big DSGE models. Let's flash back to when I was entering the economics profession at the time. There were these big models like the NPS model, the DRI model, these huge macro econometric models that were run by policy institutions and academics and some private firms.
And a lot of macroeconomic research was of that sort. And at the time I remember being, at the time I remember being skeptical about a lot of that. And if you go back to the rise of Lucas and the Lucas critique of these models, part of the reception to Lucas critique was all the stagflation and the events of the 70s.
Part of it was, I think people were getting a little tired of these big models because they were large, non intuitive. They seemed very black boxy, so you didn't really know what was happening in them. And so I think that people, I think they started losing credibility, in these big models.
I think that a lot of the DSGE models are suffering from the same fate now, they're getting large and complicated and lots of equations and you don't know exactly what's driving what result. And I do think at some point people are going to get tired of them for that reason.
I agree with you completely that if you're an actual practical central banker, you listen to your staff, present the results, but you don't take it as God's truth. And a good central banker takes a healthy dose of skepticism. When I was in Washington, I watched Alan Greenspan up close, and I think Alan, more than most, had a deep, healthy skepticism of the macro econometric models.
I should note, by the way, in my own research I tend to focus not on big models that purport to be realistic. But rather smaller models that are more illustrating points than trying to say, this is a real replication of the whole economy. I never really want to go back to those huge models.
We'll get better at these, but I think macroeconomics is so complicated that actually saying I have a model that really replicates the data, we can take it seriously for policy alternatives. I think that's, I think it's a very hard ask.
Jon Hartley: Yeah, no, I'm totally with you. I think simpler macro models that illustrate some sort of point, maybe with some empirics.
And I do see a lot more identification being brought in in different ways into macro papers. Ricardo Rice has done some of this, say, for example, his swap lines papers and thinking about those natural experiments and then adding in a model, or think about the work of Nakamura and Steinsson.
And thinking about various natural experiments, things like unemployment trends or exchange rate devaluations or things like that, where there's clear exogenous policy shocks. Or thinking about what's a monetary policy shock, things like that. I definitely see that as being a step forward, but it's certainly on the empirical side.
Yeah, also, I definitely take a simple model, over 100 plus page model that I think only maybe the author can really claim to understand it, say the least. But interesting just in general, and I think you're seeing fewer, and fewer macroeconomists in economics departments. And I think one thing thats happened, though, is a lot of macroeconomists have moved to finance departments too.
And so I think thats an interesting shift. I totally agree with you that the big questions that people, the big macro questions that people in DC and elsewhere in policy circles and just broadly the public, when recessions happen, people are wanting answers. And I think really only macroeconomists or people who study macro questions are able to come up with some answers.
So I don't think macroeconomics is going anywhere, maybe it's shifting to finance departments, but I think it's still gonna be around forever as long as we have recessions. And not to mention, I think the other thing that I think is so important, in which you've contributed quite a bit to, is the literature on growth.
And there's a famous Robert Lucas saying, which is when you start thinking about growth, you can't stop thinking about it, you can't really think about much else. And so I want to talk a bit about growth. And Mankiw, Romer, Weil is a huge contribution to the empirics of growth literature, a very highly cited paper, one of your most cited papers.
And was really the first paper to do a comprehensive growth decomposition where you have output and you're trying to explain that with labor L, capital K and human capital is H. And thousands of researchers have followed this and repeated similar exercises. And in a sense, it's trying to come up with empirical coefficients for what something like a solo model would look like.
But then along came endogenous growth theory, and in the time period that you measured, I think it worked quite well. But then along came endogenous growth theory. And I think a lot of further criticism from the applied micro crowd and the applied crowds, criticizing aggregate production functions as being misspecified.
That is, maybe aggregate economic output doesn't follow something like a Cobb-Douglas production function where output Y is equal to alpha times K beta times L to the one minus beta. I'm curious, any thoughts on the evolution of the growth literature and where it is today. I think growth is so much more, so essential when you think about one recession.
Maybe a big recession is maybe four or 5% of GDP. But when we're talking about growth differentials, say, across countries, you've got countries, many countries in the world, say China's, say an 8th or a 6th of the GDP of the US. Increasingly, the US is starting to grow much faster than, say, Western Europe or Canada, a lot of advanced economies are starting to fall behind, and they've been falling behind for at least since the great Recession now.
And there's a huge divergence there. But there's been tons of research done over the years about unconditional convergence, which models predict should happen. That is why is it that poor countries aren't catching up with richer countries or poor countries growing much faster than rich countries to catch up to an advanced economy, economic level.
We think that ideas and technology are things that can spread easily. That's something that should be happening, but empirically, we didn't really observe that for a long time. The past couple of decades, 2000s, 2010s were maybe a time where we did see some convergence, but I think since actually, since COVID and the 2020s inflation, I think we've actually seen a reversion back to divergence.
So I'm curious what you think about sort of just growth in general growth theory, where it's at today, and what you think people should be thinking about on that topic.
Gregory Mankiw: Sure, that particular paper which you write is my most cited in some sense, is a little bit like the New Keynesian papers of being intellectually conservative.
Just as the New Keynesian stuff was sort of responding to the people like Lucas and Prescott who were saying throughout all this Keynesian stuff. This paper was in some sense responding to the early endogenous growth theory by Paul Romer and which had the tone of this Solow model and this neoclassical growth theory which said that's useless.
And our point was, no, no, there's some truth to the basic Solow model, especially if you extend it to include human capital. You can do a reasonable job of explaining some of the cross country variation in growth experiences that we see. So in some sense, that it was meant to bring Solow back from the dead in some sense.
That's why the first line is, this paper takes Robert Solow seriously, because it's like, no, we actually think this model is still very useful. I actually wrote that paper sort of the same time I was writing my intermediate macro textbook, the first 1st edition. And as I was writing the chapter on growth, I was writing the Solow model, and I was thinking to myself, am I wasting students time by teaching them the Solow model?
And I started thinking hard about the empirical implications of the Solow model. That's one of the things that led me in this direction. I think now we're in a situation where endogenous growth theory has made a big contribution, and I think it coexists happily with sort of more neoclassical models, and both provide insights for different questions, and I think they can coexist.
I don't think we need to sort of throw out one or the other. My most recent publication, which came out last year, and restart, was basically putting market power into neoclassical growth models. So I take that as some indication that you can still make progress and get some insights, even with neoclassical growth frameworks.
The sad thing is, I don't think either set of models give easy policy prescriptions for what poor countries can do. I mean, if you look at Sub Saharan Africa, you can point to a variety of elements they have that prevent them from growing. But saying, this model of endogenous growth, of this model of neoclassical growth, gives me an easy recipe for what Sub Saharan Africa can do to join the developed world.
It doesn't. It brought you maybe a framework for thinking about those issues, but there's no easy answers. Or similarly, why is productivity been slower since 1972 than it was before 72? We don't think we have any easy answers to that. You can say sort of general things about the importance of institutions, importance of savings and investment.
Jon Hartley: Running out of ideas.
Gregory Mankiw: Running out of ideas. Yes, right? So I think we can say, we can say things, but I'm not sure the mathematical models give tremendous additional insight beyond what is sort of obvious.
Jon Hartley: Do you, I guess, have any sort of stance on sort of the fundamental cause of growth?
This is, I think, been a big debate in the growth literature in the past, say, 30 years or so. Is it institutions, is it culture, is it geography? A lot of people having very strong opinions on that.
Gregory Mankiw: I don't think there's one thing, I think it's a combination of things.
It's always nice to say, everything's due to this, have a causal explanation, but I think it's multiple things in my principles text. I have a case study in recent editions about why is Sub Saharan Africa poor? And I go through a whole bunch of hypotheses and you read through these different hypotheses and you realize all these things could be true.
And maybe it's not one cause of ailments. It's not like if only you had a better health system or if only so many countries weren't landlocked. I mean, so it's not going to be. It's probably not just one thing, it's probably a combination of things.
Jon Hartley: I want to talk just shift gears a little bit to policy.
So you served as chair of the Council of Economic Advisors in the George W Bush administration. But I think some people may not know that you interned at the Congressional Budget Office while in grad school. You also worked in the Reagan CEA. There were a lot of very interesting economists that worked at the Reagan CEA along with yourself.
John Cochrane was there, Paul Krugman was there, Larry Summers was there, and many others. The list is actually pretty unbelievable. I'm curious, like, what were the highlights of your various experiences in Washington?
Gregory Mankiw: Sure, give you one small correction to your chronology. I was actually a summer intern at the Congressional Budget Office while I was an undergraduate, and I actually took off a year of grad school to work at the CEA.
So I worked at the CEA in the middle of grad school, and that's in the Reagan administration. And what happened there, this is 1982. Reagan had appointed Marty Feldstein to be chair of the Council of Economic advisors. Marty Felstein hired one of his star students, Larry Summers, to be on the senior staff, and Larry Summers, who knew me in grad school as a student, hired me to be on the junior staff.
So I easily came and spent a year, 82, 83, being Larry's assistant. And so I learned a lot. That was just a fantastic year because Larry is such a great economist. I was very generous with his time. And you're right, there were other people there at the time.
I mean, John Cochrane was there that year, Paul Krugman was there that year. But Larry was mainly the person who I worked closely with. I've myself primarily as an academic, but I have sort of spent probably three and a half years of my life in Washington, various policy jobs.
And one thing I'll say is they're fascinating jobs. They provide useful perspective. So I think academics benefit from taking some time off. And they view, when they come back, they probably view a lot of academic squabbles and a lot of intellectual battles and a lot of research lines is less interesting than they would after seeing economics from a more practical perspective.
One thing I'll say about being Chair of the Council, which I did from 2003 to 2005. And I worked harder those two years than any two years of my life, by far, because the days are long. In the Bush administration, every day started with the 7:30 AM staff meeting in the Roosevelt room, which is the conference room right next to the Oval office.
In all my years at Harvard, I've been in Harvard almost 40 years, nobody's ever called a 07:30 AM meeting. While I was at the White House, every day it was at 7:30 AM meeting. It's not like you take off early at the end of the day, you work long hours at the end of the day too.
So they're are very, very long days. I left my family behind in Boston, my wife was a saint and took care of my three small kids. And I basically moved into a hotel just a few blocks from the White House because I knew I wasn’t going to have much time to travel, basically, it was extremely long days.
And by the end of two years, when I was required to come back to Harvard or give up my chair, it was not a difficult choice, I was exhausted after two years. So I was happy to come back to academia, where life is much more relaxed. In terms of particular policies, let me just mention a couple of policies that I was involved with when I were there that people might not be aware of.
I'll mention one that was a small policy that we won that I think basically was completely under the radar screen. Sometime in the middle of my tenure there, the Financial Standards Accounting Board, FASB, was going to pass a regulation that required companies to deduct options as expenses for purposes of computing earnings.
Many Silicon Valley companies were unhappy about this, because what looked like profitable companies become unprofitable and just started subtracting out all the options they were giving to their executives. Their solution to this was to have Congress pass a law preventing FASB from making this accounting change. And so there were a variety of congressmen, mainly from Silicon Valley, who are pushing this legislation.
During the campaign, George Bush, when they asked about this, had actually expressed some sympathy for the Silicon Valley view. So they were hoping that George Bush would help this law get passed. The economic team got together, we said, this is a terrible idea for two reasons. One, we don't think Congress should get involved with accounting standards.
And secondly, we actually thought FASB was right, options were expenses, and not calling them expenses was just misleading. So we basically had to convince the president to change his mind on that issue. So I remember the meeting vividly, and at the end, I'm not sure we actually convinced him to change his mind, but we convinced him to basically not support the bill.
So basically, his silence on the bill was enough to kill it, and he didn't have to come out publicly against it, he just had to not support it, and that was enough to kill it. So that was one case under the radar screen, I don't think anybody was aware of this at the time, other than, of course, the few members of Congress who were trying to pass this bill.
That was a win for the economic team. Let me tell you another big issue that we faced, that we lost. This was before the financial crisis, it was probably four years before the financial crisis. So we didn't see the financial crisis coming, but we kinda knew that Fannie Mae and Freddie Mac were not ideal institutions, this sort of uncomfortable public private view of them.
They said they weren't backed by the government, but everybody knew they would be backed by the government. And so we thought, given that they were these basically publicly funded institutions in some ways, they should be better regulated and not take on as much risk as they did. And in that, we were joined with Alan Greenspan, who was very much in favor of better regulation of Fannie and Freddie.
So we put forward some proposed changes in legislation to regulate them better, and that we failed. The people who were lined up against this were basically the left, particular congressman was Barney Frank at the time, because I lived in Wellesley. Barney Frank, who was a chair of a House Financial Services Committee, I believe, he basically came out opposed to this and his view that, no, no, you guys just don't appreciate poor people getting houses.
So he was afraid that better regulation would mean poor people would have more trouble getting mortgages, and that'd be a bad thing. Fast forward four years, in the financial crisis, if some poor people hadn't gotten mortgages, they might have been better off if they had state renters. But that was one thing we had failed.
We didn't see the financial crisis coming, but it would have been a less severe financial crisis if we had succeeded in getting better regulation of Fannie and Freddie, and we didn't. Those are sort of the kind of things we worked on. And for me, it's a new perspective on economics, there's a lot of economists who work in Washington, it's nice as an academic to see that world.
And also got to meet lots of people who weren't economists. I mean, the White House is filled with people who are usually hardworking and successful, and they're normally not trained in economics. So I got to meet lots of absolutely brilliant and fascinating people when I was there. So, a person, for example, who I really came to love when I was there was Karl Rove, who was obviously in a very different job than I was, but he was very numbers-oriented.
Really wanted to hear about the numbers and was just fascinating to get to know and see his perspective on the world.
Jon Hartley: That's fascinating, I actually just saw Karl Rove a couple days ago.
Gregory Mankiw: Very nice guy.
Jon Hartley: Very nice guy. Very, very nice guy. I guess I'm curious if you have any just broad thoughts on CEA and, I guess economist influence in Washington in general.
I think what's interesting about how economists in Washington, or how, I feel like their influence has changed a little bit. If you look historically, most CEA chairs used to be all tenured professors like yourself, going back to James Tobin and many others, going back to largely, I think, the creation of CEA, the Full Employment Act.
And I think that was in maybe 1946. I guess what's interesting now is, I think there's been a massive change, which is the rise of think tanks. And think tanks employ a lot of economists or policy analysts who are very focused on just understanding policy rather than trying to maybe discover something about truth in the world, something related to economics.
And I think that's the general mandate of economics departments and finance departments as well. But I'm curious, do you think the CEA has kind of changed in terms of its role? When you're giving these stories of having policy influence, wow, those are amazing stories. I feel like CEA has become much more of a political machine now to build charts and graphs and things to sort of support already drawn up policy desires or policy ideas [by NEC or Treasury].
And this doesn't apply to CEA exclusively, but I think, and this is something that Steve Levitt mentioned on the podcast a while ago, which is: What often happens in DC is, or in these sort of policy settings, is that policymakers or congressmen or whoever have you will have a policy that they want.
And that they decide for whatever political reason they want it. Maybe without thinking too much about all the benefits and the costs, and then they will try and find academic research that just supports whatever that previously drawn up policy conclusion is. And so, I guess at some level, if you sort of believe that that's often, or, I don't want to say often, but when that's the case, that's not a real example of economists kind of having influence per se.
Because, I think you can kind of almost always find a paper somewhere, maybe not in the top journals, that will support almost any policy. I'm curious what you think about that, and what the influence of economists have been in Washington, and has it become more political cheerleader-type role than a real type of policy role?
Well, I know there's many exceptions to this, folks like Kevin Hassett have been very influential in things like the Tax Cuts and Jobs Act and I also, I think the CARES Act. But I'm just curious what your take is on the role of the economists in DC and how it's changed.
Gregory Mankiw: Well, there's always been some truth to what you're saying. I think Paul Samuelson many, many years ago used to tell the story that when he was raising his children in Belmont, there was two pediatricians. One, whenever they brought him there, he just said, recommended bed rest, the other that always prescribed antibiotics.
And so Paul Samuelson would decide what he thought his child needed and take them to the appropriate pediatrician. That's a little bit like what politicians do, is they choose the economist that are gonna give them the kind of advice they want, and therefore they're consistent with their views.
So I think that's true. I think some contribution is really in the details of policy, because even if the, even if the president is setting the broad scope policy, there'll be lots of details that he or she's not focused on. And that's the stuff that the staff works on.
And the CEA and CEA staff work closely together with the rest of the White House staff, the Treasury Department staff, and Commerce Department, Labor Department or whatever, depending what the issue is. So I think you can get good economics in the details that are flying below the radar screen of the politicians.
The politicians probably come in with some set of ideas of what they want to do. And certainly, what we can do is point out unintended consequences and say, maybe you think this is a good idea, but have you thought about this? The CEA has no authority over anything other than giving advice.
That makes it very different from, say, the Treasury Department or the Commerce department, we don't run any programs, we just give advice. And so as an advice giver, you're only as influential as the person listening to the advice wants you to be. And that's going to vary over time.
And right now, we have someone who's not been an academic, Jared Bernstein, who's the head of the CEA. I know Jared reasonably well, and he's a very good guy, but he's a Washington economist, he's not an academic. Was that a good thing or a bad thing? I think it's a mixed bag.
He probably comes to it a little more politically because he's been very closely aligned with Joe Biden, he was Joe Biden's economist when he was Vice President. But on the other hand, I think because of that, Joe Biden has more trust in Jared than he would if he had some stranger coming flying in from Harvard.
So I see pros and cons of the changes, and it will vary over time as new presidents come in and decide to set their own relationship with the CEA.
Jon Hartley: And maybe there's something we said about gains to specialization, that maybe it's right that academics really should be focusing on doing research.
And you have think tankers to fill policy slots when their parties in power.
Gregory Mankiw: Absolutely, there's very good work in a lot of the think tanks. Not all of them, I mean, some of them are not very good, but I think some of the work of think tanks are very good.
And they're focused on the kind of questions that politicians want answered more than the kind of questions that the academic literature once answered.
Jon Hartley: Absolutely, one last quick question for you, Greg, you've been a long time defender of Pigouvian taxation, carbon taxes as a policy response to climate change.
How do you see the carbon tax policy battles going on today and the climate issue more broadly?
Gregory Mankiw: I think it's a sad state of play right now, because the Republicans seem to deny climate change. And I've been told by people who worked for Donald Trump, that Donald Trump does not believe in climate change.
And the Democrats who believe in climate change but don't believe in taxing anybody making less than 400,000 a year. So basically, there's no constituency for carbon taxes. You can make a carbon tax system progressive, but saying you're not gonna raise taxes on anybody making less than 400,000 a year, that's probably impossible.
So the carbon tax, which most economists think is the first best approach to climate change, is probably not on the table. And this is partly a reflection of a variety of dysfunction in Washington right now. Tariffs are popular in both parties right now, that's kind of sad. No party is really worried about long-term fiscal sustainability, that's kind of sad.
So we don't have, state of economic policy in Washington is not terrific in both parties right now. In my mind, that makes teaching economics all the more important, because I think in the long run, educating the electorate is the way to go to get better public policy.
Jon Hartley: Its interesting, too, just because internationally, carbon taxes, I think, have been a failure as well.
If you think about, in Canada, they're very unpopular. Justin Trudeau has introduced carbon taxes and it’s the top policy concern that's being highlighted by his opponent, Pierre Poilievre, who's very likely to come into office in about a year or so from now. And then in France, you think about the Yellow Vest protests.
And so it's, I think I agree, like the whole policy substitute idea, there's a great argument that, I think, you made on a National Geographic documentary that Leo DiCaprio, I think, was doing about climate and carbon tax policy substitutes. The idea: if you have a carbon tax, you can reduce people's income taxes.
And I think the challenge is, it's one of those things where it's an interesting economic idea. Of course, we want lower income taxes and it sounds like a great idea, who doesn't want less income taxes? There's this weird political economy problem of passing policy substitutes, which is just a very difficult thing to do.
Gregory Mankiw: Yes, yes, Milton Friedman used to talk about status quo bias, the people seem to like the status quo. So it's a hard sell, even though economists, both the right and the left, are in favor of carbon taxes.
Jon Hartley: Exactly, well, it's been a real honor to have you on, Greg, and hear about your amazing career. Thank you so much for joining us.
Gregory Mankiw: Thank you, Jon, it's great seeing you.
Jon Hartley: This is the Capitalism and Freedom in the 21st Century Podcast, an official podcast of the Hoover Economic Policy working group where we talk about economics, markets, and public policy. Our guest today was Gregory Mankiw, who is a professor of economics at Harvard University.
I'm Jon Hartley, your host, thanks so much for joining us.
Subscribe to The Capitalism and Freedom in the 21st Century Podcast
A podcast series of the Hoover Institution’s Economic Policy Working Group hosted by Jon Hartley. The podcast interviews economists, policy makers and practitioners to learn about their thinking featuring discussions on the wide range of economic topics.