Greg Mankiw on Macroeconomists as Scientists and Engineers

Greg Mankiw is a professor of economics at Harvard University and served as the chair of the Council on Economic Advisers under President George W. Bush. He joins Macro Musings to discuss the history of macroeconomics and how macroeconomists function as both scientists, who formulate and test theories, and as engineers, who set out to solve real world problems. Greg also shares his thoughts on the debate between the New Keynesian School and New Classical School and how that debate has shaped how we think about economics.

Read the full episode transcript:

Note: While transcripts are lightly edited, they are not rigorously proofed for accuracy. If you notice an error, please reach out to [email protected].

David Beckworth: Welcome to the show.

Greg Mankiw: Nice to be with you.

Beckworth: Well, it's a real treat to have you on. I know many of our listeners have used your textbook at some point in their past. So we're dying to hear from you. Many of us also read many of your articles as grad students, and as researchers once we got out. It's a real pleasure to have you on the show.

Mankiw: Thank you.

Beckworth: Let's begin as I do with all my guests and ask, how did you become a macroeconomist?

Mankiw: Really two steps. The first step was becoming an economist, and then how I got into macro. When I came to college in the fall of 1976, I had no idea that I would end up an economist. I think I was thinking about math or science. My first declared major was mathematics at Princeton, where I was.

Mankiw: Then I started dating a young woman my freshman year of college. She was taking an economics course from Harvey Rosen. She would come back and tell me what she was learning. It just seemed more interesting than anything that I was learning in any of my courses. I started actually reading her textbook, and thinking, that's kind of interesting.

Mankiw: I started taking economics courses the next semester, also I took micro from Harvey Rosen, and macroeconomics from Brent McGill. I just really liked the field. That was what got me into economics. In fact, I got a job as a research assistant for Harvey Rosen that summer. I got a taste of what economic research was like.

Mankiw: Now, getting to macroeconomics was also kind of random. During the summer of my sophomore year, I wanted to get a job in Washington, and see what Washington jobs were like. I got a job working as an intern at the Congressional Budget Office.

Mankiw: They just happened to put me in the macroeconomics decision, which I think was then called the fiscal analysis division. I worked there, and they had me working on some topics involving the Phillips curve, which is the topic that actually has dominated my career in many ways.

Mankiw: This got me interested in macroeconomics, and macroeconomic modeling. Then I came back to Princeton, and I took the graduate level macroeconomics from Alan Blinder, whose then also served as my senior thesis advisor at Princeton. That was really what set me on my road to becoming a macroeconomist.

Beckworth: So we can thank a sweetheart and an internship for all the great work you've done?

Mankiw: Exactly. I went to grad school. I had other great mentors. I went to MIT in 1980. There I met Stanley Fischer, and Larry Summers. They were I think the two figures who were the most important to me during my graduate training. Really those four economists, and the sweetheart early on in the freshman year, but also the four economists, Harvey Rosen, Alan Blinder at Princeton, and then Larry Summers and Stan Fischer at MIT served as role models for me.

Beckworth: Absolutely, great teachers make all the difference in the world, and you had some great ones. Now, from there you went on to become a professor, you've written widely, you've written several textbooks. What was it like to do the textbooks?

Mankiw: Textbooks are a lot of work, but are fun if you enjoy that kind of activity. I really enjoy writing. And I like teaching. Textbooks are just a way of sort of teaching to a very broad audience, and it's been a fun activity for me. I started off doing the intermediate macro-book. I got tenure at Harvard in I guess it was 87. The next year I knew they were going to have me teach intermediate macros, or they needed somebody to teach that every year, and I was a natural person.

Mankiw: I knew that was on my agenda for the future. Indeed, for the next 15 years I did teach it very regularly. I said, "Well, I want to teach the courses. So regularly I might as well write a textbook that really reflects how I want to teach the course. I wrote that in the late 80s, and it came out I think in 91. Then when that came out and started selling, well, publishers started approaching me about writing principles book.

Mankiw: Having written lots of textbooks I realized I enjoyed doing it. I signed up to write the principles book. I had been a section leader, what we called it back then at Harvard, my first teaching jobs at Harvard. I'd done principles before, but it wasn't part a major part of my teaching at the time. I wrote the principles book. That came out in the mid to late 90s. I guess I signed the contract in, when Bill Clinton had just gotten elected.

Mankiw: Then I spent a few years writing that. That came out, and that's in its eighth edition. And I'm starting to work on the 10th edition of the intermediate book.

Beckworth: So you must run into people all of the time who have said, "Hey, we've read your book. Hey, you've been an influence in my life." Is that right?

Mankiw: Yeah. I run across people all the time very randomly who have read my book. Then having a last name like Mankiw means that they quite quickly identify you with a book they've read, unlike a name like Campbell or Stein, it's kind of an unusual name. Recently ... I'm in Nantucket for the summer, and I was down at the beach, and the young man who was working on the beach was from Bulgaria. When I said my last name to him, he said, "Oh, yes, I've read your book." And that he was studying economics in Bulgaria, and he's here for a summer job.

Mankiw: It happens quite frequently. I sell now in the United States something like 200,000 books a year, and probably an equal number in translation around the world. There is just a lot of people every year who see the name Mankiw in some capacity.

Beckworth: So you're like the Samuelson of this generation then, right? You're the one-

Mankiw: I mean, I'm not the economist that Samuelson was, but to the textbook writing, I think we have a pretty good market share as when Samuelson did in his day.

Beckworth: Now you've also written widely in macroeconomics, and we'll get to that later. Some of your other responsibilities and accomplishments, you were the chair of Harvard's economics department. You were also I think, there are listeners who might care a little bit about, the director of the CEA under President Bush. Tell us a little bit about that experience, what was it like?

Mankiw: It was a great experience. I'd been at Harvard at that point for I guess 20 years or so, when I took the job. I needed a break to do something different. I've always been interested in policy. I actually spent a year ... When I was a grad student, I spent a year on the staff of the CEA when Marty Feldstein was chairman. I knew what CEA was like as an institution. When the opportunity came up to chair the CEA under bush, I was very excited to do it.

Mankiw: I basically did it from February 2003 to February 2005. Harvard let's people go away for two years for public service without having to resign. For me it was fun, because I got to apply economics in a different kind of way than you do in academia if things are more practical problems.

Mankiw: I had a fantastic staff of people, my two main deputies were Kristin Forbes from MIT, and Harvey Rosen who shows up again. I got to work closely with him at the CEA. It was great having him as a colleague. Having worked for him as a student assistant many, many years earlier. Then a wonderful chief of staff Phil Swagel, who is now in University of Maryland.

Mankiw: I just learned a lot from all the people that I was working with. Then I got to meet a lot of interesting folks who weren't economists. I mean, people in the west wing. Obviously from the president, the vice president, to the advisors like Karl Rove, and Steve Friedman who came from Goldman Sachs, to be head of the National Economic Council while I was there.

Mankiw: It was a broadening experience for me. And a tremendous amount of work by the way. I mean, I was happy after two years to relax, and go back to academia where the pace of life is much more manageable. I just met a lot of interesting people, learned a tremendous amount, and probably would encourage if they have the opportunity to take it.

Beckworth: Yeah. We previously had on the show Jason Furman, Lisa Cook, Jay Shambaugh, people who've all worked at the CEA. They've mentioned it's a great experience, but you're busy all the time.

Mankiw: Yeah.

Beckworth: Writing memos, doing briefs to the president. How regularly would you go see president bush?

Mankiw: I'd probably on average see him two, or three times a week. It varied, sometimes he would be out of the country, and I won't see him all week along. Some days I'd see him every single day. I'd say on average two or three times a week. Rarely was it small one on one meeting.

Mankiw: I think there is only twice, and that I count, where I was literally the only person in the room with the president. Once on air force one, and once in the oval office to discuss something. Most of the meetings would be a meeting of two dozen people sitting around the conference room, discussing whatever the particular policy issue at the time was.

Beckworth: You are considered a department head. Right, you are on the president's cabinet in that position?

Mankiw: It varies from administration to administration with the CEA's cabinet level position. When I was there it was not a cabinet level position. It was actually the staff position, along with let's say the NEC director. When there are cabinet meetings, I would attend the cabinet meetings, but I'd sit in the outside. I won't sit at the table, which is where the cabinet officers sit. I would sit around on the other side of the room.

Beckworth: All right. Interesting. Of course right now the position has been demoted. It was cabinet previously, now it's demoted under President Trump. I know our listeners will want to know too. You became well known for a comment you made during that time about outsourcing which any economist would find is reasonable, but apparently it blew up in the media, and even surprisingly the republican leader of the House of Representatives responded to it.

Mankiw: Nothing I said was controversial as a matter of economics. I think what I said was really, very plain vanilla textbook economics. I think it was portrayed in certain media circles as if I was saying it's great when people lose their jobs. I think the way it was portrayed. Particularly, there was a headline from LA Times that I think sort of misrepresented it in a way to make it look politically unpalatable.

Mankiw: Fortunately, I had economists from both right and left came to my defense. I think the people inside the administration recognize this was just a political firestorm, and nothing I said was really all that objectionable from an economic standpoint. It came and went. It was politically charged, because it was coming after a presidential election. Therefore, everything becomes more politically charged then.

Mankiw: I had no doubt, but I had the support of the president during this. In fact, I mentioned earlier, one time I was in the office with him, and I was the only person in the oval office, we actually discussed this. He told me not to worry about it. It'll all pass.

Beckworth: But back then, the context back then was outsourcing was a big deal. I mean, China just entered the WTOs early mid-2000s, right? Outsourcing was a big thing. You have a great paper you wrote on this, the politics, and the economics of offshore outsourcing with Phillip Swagel.

Beckworth: You go back and really neat, I encourage the listeners to take a look at it. You document how this was already a hot topic before that sentence came out in your report. So, it's kind of a context of the time that made this really explosive, and all you really said was what you would teach in the econ 101 class, right? Specialization.

Mankiw: That's exactly right. I think trade is a very difficult topic, because economists are pretty clear when we think about it. The general public is not convinced. The general public has a view that's closer to the mercantilist that Smith argued against this idea that trade is good, because you can export stuff and create jobs, and imports are that burden we have to bear to export stuff.

Mankiw: Imports are good, because they're going to increase our consumption basket, and exports, so we have to bear because people want to get paid for what they're selling us. Trying to convince the public of the virtues of trade is hard. I mean, President Trump has I think already taken advantage of the anti-trade bias in the public through a lot of his public statements.

Mankiw: He hasn't done a whole lot lately with that. We'll see how much he ... I mean, my perspective is, it was not a great thing that he backed away from the Trans-Pacific Partnership, but he hasn't followed through yet, at least on renegotiating all the other trade agreements.

Mankiw: Quite the best analysis of the public's view is a book Bryan Caplan, *The Myth of the Rational Voter*, where he talks about all the biases that the general public has about how the economy works. Two of them he talks about are the anti-market bias, skeptical market mechanisms in allocating resources. The other is anti-foreign bias.

Mankiw: That people are generally suspicious of foreigners. I think it's probably true, especially when foreigners are different from us. I think to be honest, I think there is a little bit of racism involved in how people think about trade. When people talk about with Canada, or England, they don't seem to get very upset, because they're a lot like us. They're much more concerned about trade with China or Japan, because those people seem very different from us.

Mankiw: That's extremely unfortunate, but you kind of understand where it's coming from. If you think of when the human species evolved, if you saw somebody that was very different from you coming over the horizon, they were probably coming over to kill you, and steal your women and food, than they were to engage in mutually advantageous trade.

Mankiw: I think it maybe the bread and the bone of human beings that you're suspicious of people who are different from you. I think in the money global economy that's not called for, because I think we understand, presumably our rational cells should be able to transcend those reactions, but I don't think it always does.

Beckworth: When I used to teach I'd often bring up trading between states. Do we put borders between the state of Michigan and Indiana if there is trade going down? We don't. We don't think twice about it, and my students would always say, "But they're American." Well, we're all human, we're all part of the human family. So this shouldn't be any different in principle.

Beckworth: A question I have I'd like to hear your perspective on is, when you made that comment, we were coming out of the 2008 recession, the recovery in terms of jobs was weak. I mean, the economy itself was growing, but the labor recovery was weak. If you look at the present situation coming out of the great recession, both in Europe and US a weak recovery.

Beckworth: My question is, do you think recessions, weak recoveries tend to foster, create that animosity to foreigners, to trade? Maybe the counterfactual question would be, what if we had a robust recovery from the 2008 recession, would we have that same amount of populism as we have today?

Do Recessions Foster Populist Sentiment?

Mankiw: I probably would not. I agree with you, tough economic times gets people concerned, and they start looking for a culprit. When things are bad there is a natural temptation to say, "Who can I blame for this?" Blaming foreigners is one of the culprits running around. Politicians tend to foster that, because they certainly don't want to take the blame themselves. So blaming other countries is a very tempting target for them to feed into.

Mankiw: I think you're absolutely right that tough economic times does not bring out the best in people. In fact, my colleague Ben Friedman wrote a book called *The Moral Consequences of The Economic Growth.* One of his big themes is that we tend to be more moral as a society, more open to other people that are different from us when growth is robust.

Mankiw: When growth is poor we tend to be more exclusionary, whether it's foreigners, or underprivileged people, or people from our own society who are different from us when we're struggling. This was his case for why we need to promote economic growth. Not only for reasons of material prosperity, but for reasons of, if we want to be a good people, we're more likely to be a good people when there is robust growth.

Beckworth: Yeah, I think this underscores why we need good macroeconomic policy, even at the business cycle horizon. We need to minimize business cycles as much as we can to avoid these happy developments.

Mankiw: Absolutely. In bad economic times, people are grasping for change, and sometimes they're grasping in the wrong direction.

Beckworth: Let's move onto your 2006 paper in the Journal of Economic Perspectives, *The Macroeconomist as Scientist and Engineer.* That's really fascinating reading, I encourage my listeners to get it. By the way, we'll post these articles on the webpage on SoundCloud for listeners, so if people want to take a look at it they can google it themselves. I want to use it to motivate a review of the history of the discipline of macroeconomics. We'll journey through ... Because you do that in the paper, really fascinating.

Beckworth: Then also take the framework that you use, and the scientist versus engineer we'll talk about in a minute. And kind of maybe try to make sense of some to the developments since the crisis, the past decade or so. I want to begin ... This is part of the discussion by reading a quote from your paper. Let me start with this.

Beckworth: “This essay offers a brief history of macroeconomics, together with an evaluation of what we have learned. My premise is that the field has evolved through the efforts of two types of macroeconomists. Those who understand the fields as a type of engineering, and those who would like it to be more of a science. Engineers are first and foremost problem solvers. By contrast, the goal of sciences is to understand how the world works.”

Beckworth: “The research emphasis to macroeconomists has varied over time between these two motives.” Unpack that for us. How are macroeconomists fulfilling both the scientist role, and the engineer role?

Macroeconomists as Scientists and Engineers

Mankiw: I wrote that paper as I was coming back from the Council of Economic Advisers. I spent two years at CEA, and I came back. I spent most of my career as an academic, and that's primarily how I view myself as a professor. I did sort of go into Washington occasionally, and I probably spent in total just three and a half years of my life in policy jobs of some sort.

Mankiw: My sense is that a lot of academics don't that. Don't understand how economics is used in the policy process. The purpose, the large part of that paper was to try to explain to people, and over time the academics, how academia is different from the policy world. What I viewed is, the scientist is being the pure academic, the person who is spending their time writing articles for the American economic review, and who never dream to go to work in Washington.

Mankiw: I explained it, the other economists, the engineers who are going to Washington, they're not trying to solve practical problems. Both kinds of macroeconomists are useful and important, but they have a slightly different perspective on the activity that they're doing. My sense was that, sometimes there is not enough respect of each group for the other. And having been one of the two people, who go back and forth fairly frequently, I was trying to explain that dynamic.

Beckworth: How was that article received? Did you find the scientists more sympathetic to the engineers after they read your article?

Mankiw: I don't know. I don't spend a lot of time thinking about how my articles are received. I write them. Then after they get published I move onto the next thing. People…

Beckworth: That's why you're so productive, just move on.

Mankiw: Yeah. I try to look forward rather than backward. I mean, I came into the macroeconomics profession at a time when I think the scientists were preeminent, and macroeconomists and engineers were who seemed to be in retreat. In particular, when I came with the grad school in 1980, what were the big ideas?

Mankiw: Well, Robert Lucas had just recently written his articles about anticipated money in the 70s. Robert Barrow had written the empirical article saying only United States matters. Sargent Wallace was writing about how systematic monetary policy was irrelevant, because only monetary surprises that mattered. Then there was Kydland and Prescott who said, "No, the real business cycle is driven by real shocks. But don't you think about money very much when think about the business cycle."

Mankiw: All those were tremendously influential academic articles. They seem to have really no impact at all on people who actually do macro-policy. That is, I don't think anybody who was running the Federal Reserve said, "Oh, yes, the business cycle is the optimal response to exogenous productivity shocks, and therefore it's perfectly fine. I think that vision was very prominent in academia, and absolutely no impact on the real world of policy.

Mankiw: Given the way we live today, or how the world works, that's a good thing. I don't think many people actually believe in all that sort of new classical macroeconomics today given the current perspective. At the time, to me as a student it was very disconcerting that there was a huge gap between what the scientists were doing, and what the engineers was doing.

Mankiw: I was in MIT at the time, the MIT crowd tended to be more applied. Stan Fischer, who was my PhD dissertation advisor. He's obviously gone to onto tremendously … policy maker. I think he was skeptical about a lot of that new classical stuff, but quite respectful of it.

Mankiw: Robert Solow was also one of my professors there at the time, was less respectful of it. He was quite dismissive. He thought the classical direct movement was a bit crazy. He saw both perspectives at MIT. There certainly was not who thought that yes, Ed Prescott is right, and what we were observing is a pretty efficient business cycle.

Mankiw: When I came to the profession, one of my questions was, okay, what is it about the world that the Lucas, Prescott view of the world doesn't apply? Why is it that that doesn't influence policy? How can we move forward this alliance of macroeconomics, so it becomes somewhat more relevant for the engineering side of macroeconomics?

Mankiw: That's what a lot of the new Keynesian research, that I and others, like [inaudible] and David Romer, and so on, worked on in the 80s, just tried to sort of make science and macroeconomics closer to the engineering of macroeconomics.

Beckworth: Yeah, your article is very great in chronologically listing the development of the discipline, and kind of the back and forth between the scientists and the engineers. You began with the Keynesian revolution, and the introduction of general theory. Kind of the early Keynesians in the 1950s and the 1960s.

Beckworth: There I believe you argue that was much more of an engineering approach, is that right? That they were there to solve the problems of the Great Depression, and preventing it from happening again.

Mankiw: Right.

Beckworth: Things like that. IS-ML model comes out. But also a lot of practical, large econometric models, which today we as academics don't put a lot of weight on them, but back then there was a lot of emphasis put on these large econometric models.

Mankiw: Yes. I mean, when I attended my first summer internship at CBO. This was summer of 1978. Robert Lucas stuff was starting to have an influence. What they had me doing was actually studying the parts of the big macro models. I remember sitting there with the equations. I believe it was the DRI model, trying to figure out how this model was put together, and how it worked.

Mankiw: They were still using these large macroeconometric models. Despite the fact that people like Robert Lucas were writing about how useless they were. What I was struggling with as a student, and I was trying to hit at, how do you square what the smart people at CBO were doing, and what the smart people in academia are doing? That question has haunted me really throughout my career.

Beckworth: So you had that initial stage. Macroeconomics is born out of the great depression. The first bit of history is the general theory, and the early Keynesians. Then you move on, and say, there were shortcomings in that. They made contributions but there definitely were shortcomings, and the new classicals come out of that. You mentioned there is three waves.

Beckworth: The first wave was Milton Friedman, his permanent income hypothesis, which raised questions about the marginal propensity to consume. Transitory income may not be that large, and therefore fiscal policy may be limited. His history works, so that money mattered. I guess the biggest thing, the policy one would be his Phillips curve critique. Then you mentioned as you did a few minutes ago the second wave was the rational expectation revolution, Robert Lucas, Sargent Wallace.

Beckworth: I mean, the third wave was their real business cycle. What was fascinating to me, and you touched on this, I just want to hear it a little bit more is, the new classicals were very disdainful of the old Keynesians. You had some great quotes in there.

Beckworth: I wanted to read just a few of them. I'm reading the next one here from your paper. You say, "One of their goals as the new classicals was to undermine the old Keynesian macroeconometric models, both as a matter of science, and as a matter of engineering. In their article after Keynesian macroeconomics, Sargent and Lucas wrote, "For policy, the central fact is that Keynesian policy recommendations have no sounder basis in the scientific sense than recommendations of non-Keynesian economists, or for that non-economist."

Beckworth: That's a pretty strong word there. Then you have a great quote from Lucas here in 1980 article titled, *The Death of Keynesian Economics.* He says, "One cannot find good under 40 economists who identify themselves or their work as Keynesian. I mean, people even take offense if referred to as Keynesian. At research seminars, people don't take Keynesian theorizing seriously anymore. The audience starts to whisper, and giggle to one another."

Beckworth: Here you are, you're coming out, and you're going to become the new Keynesian, what does it feel like to try to work in that environment?

Working in Economics as a New Keynesian

Mankiw: Well, I think there was antipathy between certain people in new classical school, and certain of the Keynesian. As I said, I think Robert Solow was very dismissive of the new classical. I think Robert Lucas took a lot of offense at not being taken seriously. As a result I think alto of antipathy for the Keynesians.

Mankiw: I actually don't think it's a particularly healthy way for an academic profession to operate. I'm glad to say that the younger generations to be a lot more civil to one another than I think the Lucas, Solow generations. I was coming up as a student, and things like meeting Robert Lucas, who was obviously a brilliant macroeconomist, being very dismissive of Keynesian economics, but also recognizing, you know what? These Keynesian economics, what they're doing at places like the congressional budget office, and the CEA.

Mankiw: How could that be? I mean, and trying to sort of square that. Trying to figure out ... Trying to figure out, there is obviously value in what Robert Lucas is doing, there is value in what the CBO is doing. How can you hold those two seemingly inconsistent ideas in your mind at the same time?" That was a large part of the motivation of …

Beckworth: You mentioned in the article that these new classicals didn't come to DC. You mentioned their work was very more theoretical. It definitely wasn't adopted in DC, but they themselves didn't go to DC, whereas new Keynesians did, so your generation did. You mentioned yourself John Taylor, Bernanke, and Larry Summers. You guys all went to DC, got immersed. You definitely took on the role more of an engineer while you were in DC.

Beckworth: One question I have that followed along this, and I know this term is dated, it doesn't really mean anything today, because this line is no longer clearly drawn. Bob Hall talked about freshwater versus saltwater economist.

Mankiw: Right.

Beckworth: Back when it did have meaning, could you map that onto the scientist and engineer labels pretty closely?

Mankiw: That's right. Bob Hall's labels, freshwater and saltwater was based on the coincidence that a lot of the new classicals were at Chicago and Minnesota, whereas a lot of the more Keynesian economists happened to be on coasts. A lot of Harvard, MIT, but also at Stanford, Princeton too.

Mankiw: I think he was noting this particular coincidence. But you're absolutely right, that the Minnesota, Chicago crowd tended not to go to take policy jobs as much, which ... There is different ways of viewing that. I mean, one could be that of more pure academics. They're really sticking to their knitting, and they're not going to be distracted by political concerns.

Mankiw: Another thing is that it means they're stuck in an Ivy Tower, and therefore not having to conform practical problems that motivate a lot of macroeconomics. I think one can argue that it's a strength or a weakness. I think it is probably a little bit of both. I know that by the way that of the new classicals, I think Milton Friedman is probably the most practical. I mean, he's the earliest ... He actually had done some work in Washington.

Mankiw: He did get involved in public advocacy to a large degree as opposed to just writing academic articles. So, I think he was in some sense much closer momentary to the Keynesians, even though his policy prescriptions are very different than the later generations, people like Sargent and Lucas, and Prescott who are more-

Beckworth: Brad DeLong wrote an article. You probably remember the monetarist revolution in 2001. He argued, new Keynesians really are a current manifestation of the old monetarists. I think it makes a lot of sense. In your article you go from the new classicals to the new Keynesians, and then you talk about the new classical synthesis. You mentioned [inaudible]] for his 2003 textbook, as an example.

Beckworth: My question is, what do you label today, when I think of mainstream macro, or the dominant view, I would call it new Keynesian. Is that the right term? When you think of central banks, and maybe cutting edge macro research, I would say it's new Keynesian, there's more bells and whistles added since the great recession in terms of modeling. How would you classify the current dominant paradigm in macro?

The Current Dominant Paradigm in Macroeconomics

Mankiw: Well, I think since the financial crisis there has been some rethinking of what the dominant paradigm is. Let me go back just like 10 years prior to the financial crisis. Then I think there had been a synthesis that was developed. The synthesis is summarized extremely well in…

Mankiw: That synthesis does include elements of both new classical and new Keynesian. I think of it as new Keynesian in many ways because the key price is playing an important, and if we're monitoring neutrality and systematic feedback roles playing important roles.

Mankiw: That to me seems very Keynesian. People from the classical traditional say, "Yes, in some ways it's Keynesian, but also gives a very strong role to expectations and forward looking behavior, which was something that the new classicals, particularly the rational expectations had emphasized. There is a lot more inter-temporal focus than it were in some of the traditional Keynesian models. I think calling it a synthesis is certainly a reasonable characterization. To my mind it seems more Keynesian than classical.

Beckworth: Back in your paper, you mention when you came to the CEA early on there was still ... You mentioned you were still using those models, those large macroeconometric models. You also mentioned Larry Myers book, and one of the points that you make is, a lot of the cutting edge, at least up until maybe a decade ago. A lot of the cutting edge macro still wasn't being widely adopted in the policy world.

Beckworth: I have this quote from you. The sad truth is that the macroeconomic research for the past three decades has had only minor impact in the critical analysis of monetary or fiscal policy. My question is, do you think that's still true? I guess, when I see policy being done today, I hear things that sound more like the synthesis, the new Keynesian view for example. You see a lot more of the top discussion of like the natural interest rate, which is very new Keynesian, very Wicksellian, or for guidance. Do you have a sense whether modern macro is permeating more in the policy role or not?

Mankiw: I think to some degree that is true. I think that is probably more true now than was 10 years ago when I wrote that article. I think some of those articles ... Some of the synthesis models have permeated. On the other hand, the financial crisis is cast from a shadow on that synthesis model, in the sense that, one of the things we learned in the last financial crisis was that, financial institutions play a pretty larger role in thinking about macro-policy.

Mankiw: Those financial institutions are completely absent in that synthesis model. If you read the Woodford book, you're not going to find very much on financial institutions. So, that's kind of obviously not thinking about financial institutions. I think what we learned in 2008, 2009, was that we need to spend a lot more time thinking about them. Even if that synthesis model is a nice benchmark, and I think it still is a nice benchmark, there is a lot that it leaves out.

Mankiw: That's the sense which I think ... in 2006, one might have said, we're reaching a consensus, and we kind of understand whatever the benchmark model is. Macroeconomists could feel proud of ourselves for having figured stuff out, and then all of a sudden that understanding became deficient, or saw a set of circumstances that were much more difficult to make sense of with that model than the previous…

Beckworth: There has been a lot of work, and a lot of commentary on the absence of financial markets intermediation in these models. Let me come to the defense of the new synthesis, just for a little bit, and you can push back. If I go back and read for example the work on Japan in the early 2000s by folks at Princeton? So, [inaudible] ‘98 paper, Woodford and Eggertsson.

Beckworth: A lot of them I think you would consider them part of the new Keynesian, new synthesis group. They argue, you can get ... They were worried about the zero lower bound, in the context of Japan. They weren't thinking it'd happen in the US. They had policy prescriptions like price level targeting, commitment to be ... Krugman's was irresponsibly credible.

Beckworth: They had policy prescriptions ... In my view, it never really tried maybe for political reasons, but level targeting. Woodford and Eggertsson argued that QE wouldn't be that effective in generating a robust recovery. You could look at for example the Great Recession as a case where the natural rate goes well below zero, and the Fed simply can't go down to it, which it's something that was understood before the crisis. Is there not any kind of vindication of that earlier, even though it didn't include the financial structure?

Mankiw: I think to some degree I agree with you. I think we still do not understand. I think the zero lower bound is going to get a lot of discussion. First the piece of Japan, and United States. I don't think we really at this point fully understand what the best tools are, to what extent is quantitative easing, to what extent is it for guidance. I think it's one of the big unanswered questions in my mind in monetary policy.

Mankiw: If Larry Summers is right about secular stagnation, then it's an issue that future fed chairmen and chairwomen are going to have to think more often about, because it's perhaps an issue that we're going to face more regularly.

Beckworth: Do you think the Great Recession has sharpened our thinking? Do you think it’s made our journey forward more focused as macroeconomists?

Mankiw: I think it has made us more engineers again. I think it's made us focus on practical problems, because when these problems confront you, doing pure theory seems somewhat inadequate. You want to do something that's actually more applicable. Sometimes that can involve developing new theories. I think there has been new theories of financial markets. Things like Geanakoplos's work on leverage cycles. I think it has focused our attention on a bunch of questions that we hadn't focused our attention on before.

Beckworth: Before we get into some of the discussions on the current Fed policy, I want to go back briefly and talk about some of your work. You're a big part of the new Keynesian tradition, you've done a lot of work there. One of the more recent things you've done is you've reintroduced I think, that's the right word, the idea of public information as being a rigidity, or a constraint that prevents markets from quickly clearing. Can you tell us about that line of research?

The Public Information Restraint

Mankiw: Sure. Throughout my career I've been thinking about why is money not neutral. The related question is, what's the right theoretical foundation of the Phillips curve. To me that's sort of a big question of macroeconomics, because I basically think micros works generally pretty well, but if this micro worked perfectly then the fed would really would be neutral, and we know pretty well that it's not.

Mankiw: I'm trying to figure out, why does it have real effects? Is a very fundamental question. The literature moves in the direction of developing a model as I was describing. Actually the standard model ... Even today is the standard model is something like the [inaudible] model. Someone has referred to it as the new Keynesian Phillips curve. That model though has a variety of features that seem to me deeply factual.

Mankiw: The first paper that made me really focus on that was a paper by Larry Ball in the American Economic Review, about how under teller-like Phillips curve you can get booms from credible disinflations. When we think of disinflation it's typically periods of recessions, like the Moca episode in the early 80s.

Mankiw: If you've had a credible disinflation, actually given the dynamics, and plus this model, that should cause a boom. I've been thinking about dynamics of this. I wrote, I guess I had a lecture that got published in the economic journal. About why the Phillips curve was an important thing to me to think about, but the model we had didn't work.

Mankiw: I was trying to think, what's the right model of that? That's why I teamed up with Ricardo Rice, who was a grad student at Harvard at the time. He and I started thinking about different ways of founding a Phillips curve that would provide dynamics that were closer to what we saw on the data.

Mankiw: That's when we got to this idea of speaking information. So it's an information they store, and in that way it's like Lucas, but unlike Lucas, we attribute a lot less rationality on the part of price setter, where we have a lot of people walking around with quite out of date information. The idea here is that people are setting prices, some have updated information, some have information that's a quarter old, some have information that's a year old.

Mankiw: The direction among the different price setters, with different amounts of information is what generates sluggishness of prices. We argued that this sort of alternative to the new Keynesian Phillips curve generates dynamics that were closer to what we think we observe in the data.

Beckworth: Is your sticky information that you developed with Ricardo Reis, is it similar, or it shares features with this idea of rational and intention?

Mankiw: Yes. I think it's very similar. It's a slightly different model of information processing. The dynamics aren't exactly the same. But yes, this came out ... Mark Woodford was writing a paper applying to the [inaudible] idea, about the same time we were. He generated dynamics, they're not exactly the same as ours, but I think more similar than different.

Beckworth: Yeah. Just to make this practical, and maybe concrete. May I ask this question, do you think how sticky information is, is it indigenous, it it dependent on the environment? For example, if I'm in Zimbabwe 2008, I'm probably very, very attentive to what's happening to prices, right? Because I'm living in a hyperinflation environment.

Beckworth: I'm very mindful of the government's doings, whereas maybe in the case of the US, a low inflation environment, I might be less mindful of it. Is there some endogeneity to this idea?

Mankiw: Yeah, absolutely. I don't doubt that that's the case. In our model we didn't have that. We had the updating of information being an exogenous parameter, but you're absolutely right, that in the real world, how much people focus on the macro economy, there is lots, depending on circumstances.

Beckworth: One of the critiques I've got ... I've been a big advocate of nominal GDP level targeting, or even just price level targeting, because I think it would do wonders in the environment, like zero lower bound, even preventing booms, but as well preventing bursts. One of the critiques I often get is, well, the average Joe is going to wake up at 5:00 am in the morning and go, "Woo hoo, nominal GDP target of 5%.

Beckworth: They don't know this. They don't care. I try to say, "Well, it kind of gets embedded in the world around them." Does your research speak to this? I mean, people are filled with so much information. We've got Twitter, Facebook, blogs, newspapers, TV, radio. In a world like we're living, maybe it is hard to get ... I'm making this really practical. It may be hard for the Fed to do certain things, like forward guidance, we talked about raising inflation, its inflation target. I mean, does your work shed any light on that?

The Fed’s Communication Problem

Mankiw: Like, I agree with your basic sentiment that, if you think about information, you have to acknowledge the fact that most people are not focusing on the macro economy. If you're ... I mean, if you go to your barber next time he changes his prices, let's say, why did you change your prices, and how did the fed policy impact that decision? My guess is he would say, "What do you mean the fed? What's the fed."

Beckworth: Right.

Mankiw: So, a lot of the people who are out there actually setting prices don't really have the kind of information that we might presume agents do inside our models. The paper with Ricardo was trying to find a crude way of doing that, of saying, some people are up to date, and some people have really outdated information, that was one way of modeling that. I think trying to get better ways to model that is going to be important.

Mankiw: We do obviously respond to their environment, and if the environment changes they're going to learn about the environment. Most people are busy doing things involved in their own personal lives, not thinking about the macro economy. This is one thing I think Robert Lucas got really right is sort of information storage, which is that what happens to most people most of the time is idiosyncratic, it's not about the macro economy.

Mankiw: There are macros being thrown out there, but it's not impacting your day to day life access under very extreme circumstances. I think trying to figure out that is important. I mean, if you think about the stuff I did with Ricardo, I was just trying to bring a little bit of behavioral economics into macro. I think that's actually a very promising growth area going forward. Sometimes it's an interesting information story along the lines of, look, we're willing to make people a lot less rational than Robert Lucas was.

Mankiw: As a result you get dynamics that are different. In fact, the dynamics from the model with Ricardo was actually more similar to the dynamics you got in the Stan Fischer's style of contracting models. We don't need contracts. But then that was coming from information dissemination rather than the contracts changing. The dynamics in terms of prices, and output were very similar to what you get in the Fischer's style model.

Beckworth: Very interesting. Let's move on in the last few minutes we have together to current developments. I want to first move on to Fed policy, and as you know the Fed has had a hard time hitting its inflation targets, since it was introduced in 2012. We could actually go before that. I mean, implicitly it was falling in something like 2% inflation target.

Beckworth: Ever since the bottom of the recession in 2009, it's really had a hard time consistently getting it to 2%. Do you have any thoughts on why the Fed has not been able to get back to 2%?

The Fed and Its Inflation Target

Mankiw: The Phillips curve has been very strange lately. I mean, if you told me back in 2006 that we were on the verge of the biggest recession, driven by a huge drop in aggregate demand since the great depression. I would have said, "I guess inflation would be coming way down." Inflation came down a little bit, but not a whole lot.

Mankiw: Then we've had quite a big recovery, and unemployment now was a bit below what most people it naturally was. Inflation doesn't seem to be doing very much. It's been one of the big puzzles to me, it's why inflation hasn't moved around more in response to macroeconomic events of the past decade. The standard explanation that people give is, the reason inflation didn't fall during the greatest recession more than it did was that expectations were well anchored.

Mankiw: Well, maybe but that seems to me like a bit of a cop out. I mean, it seems ... As long as ... If you say [inaudible] expectations, which is surely true, but then you allow expectations to be whatever they are to justify the data, sometimes you don't really have a model, you really just have an excuse quarter by quarter. I know Robert Lucas had said once, beware of if you're sparing free parameters. That I think is a good piece of advice.

Beckworth: Nice. Yeah.

Mankiw: If you only have expectations to be a free parameter to explain the data, then you really are explaining very much. I think it's been a bit of a puzzle that I don't know the answer to. It's something that we need to work on.

Beckworth: What about the Fed's big balance sheets? The Fed itself has been talking of this whole year, signaling that it's going to reduce it. It's going to be a reduction, it'll be large than it was probably before, because of the growth more reserves. They want to reduce the balance sheet.

Beckworth: There are others including your coauthor Ricardo Reis, who want to see the Fed maintain at least a little bit more of a balance sheet to do what it has adopted during the crisis, some reserves. I wonder if there is any thoughts you have on what the Fed should do with its balance sheet.

The Fed’s Balance Sheet

Mankiw: I don't have strong opinions about that. I mean, I honestly think that both sides haven't come down from the heart of that. I presume what they're going to do is slowly let it diminish. As a lot of these assets pay off, they don't have to replace them with long lived assets. Sometimes stringing the balance sheet can happen gradually over time, almost automatically over time, as these debts they're holding are paid off.

Mankiw: One of big questions is, how far do we need to be away from the zero lower bound? I think the Fed would love to go back to more normal times, which means, the federal funds rate is further from zero, a balance sheet that's smaller. I think that's a great aspiration. I think whether that's a realistic aspiration depends on those circumstances.

Mankiw: My sense is that they've been predicting to be going back to normal operating procedure more quickly than it happened repeatedly, because the inflation hasn't come back very much. The economy has been growing slowly. They've been staying closer to those the zero lower bound than they wanted to.

Beckworth: Yup.

Mankiw: I don't know. I think it still in some senses could be data dependent going forward.

Beckworth: Along those lines. One way, back to more normal policy would be for financial rates to jump back up. One way that could happen is if we suddenly had a surge in productivity growth. All this talk about the new economies, smart machines, driverless cars. At some point if it manifested itself in the data, we saw productivity grow, that would lead to a higher natural rate. Maybe a deeper question is, are you hopeful about long run growth in the US that would lead back to that situation?

Mankiw: I've been following this debate, I think that it's Bob Gordon versus Erik Brynjolfsson.

Beckworth: Yup.

Mankiw: Bob Gordon is basically the pessimist that says we should get used low productivity goals, because that's the norm, and Erik Brynjolfsson says we're in the verge of this brand new economy with incredible innovations and driverless cars, et cetera. I have no idea what's right, but I have noticed historically they were terrible at predicting productivity growth. My sense is that ... This is one thing that I think, I agree with Prescott with this. Total productivity is pretty close to a random walk.

Mankiw: I don't really take any of these predictions all that seriously. If I have to predict going forward. The average historical productivity [inaudible], it's probably not a bad guess. I think anything that … slow going forward, or it's something very fast, is probably there is some more information that we have. I mean, we don't really have that ability to put its productivity. Therefore, historical average over the past 100 years is probably good or bad as any.

Beckworth: All right. I'd be remiss if I did not ask you about nominal GDP targeting, because listeners know I'm a big fan of it. Where I work here at the Mercatus Center, my colleague Scott Sumner is also a big fan of it.

Beckworth: You had a 1994 people with Robert Hall on it. You had a hybrid nominal GDP, or nominal income targeting rule. I know since then you had a more recent paper of exploration about [inaudible] paper, where you allude to something like a nominal GDP rule.

Beckworth: In other places I've read, like your work with Ricardo Reis on public information, and sticky and information. You've talked about a price level rule. Where do you come down today on nominal GDP level target?

Nominal GDP Level Targeting

Mankiw: I think that's an intriguing idea worth considering. I think all these rules are really benchmarks for central banks to think about as they're setting policy. I don't think any rules we have, whether it's the nominal GDP rule, or the price level rule, or Taylor rule. I don't think any of them are rules that we should feel confident enough in to want to legislate into a law, and get it fed into a computer and go home.

Mankiw: I think these are all sort of useful benchmarks for the Fed to look at as they're deciding. I think ultimately, given the world is vastly more complicated than our models, and given that we really don't know what kind of shocks are coming down the road. I think our situation now with the Fed is going to act with discretion. I think that's probably the best way to go.

Mankiw: I think it needs to also keep focus on these rules, because completely unconstrained discretion is ladder-less. I think these different rules provide a kind of guidance to exercise discretion more wisely than if you would if you were just going fully by the seat of your pants.

Beckworth: Okay. Well, last question, what would you recommend to a budding young macroeconomist today? Someone who just desires just going to grad school, what would you tell them to do?

Mankiw: I would tell them to get lots of different experiences. Try to be very broad about learning things, and then broaden to taking different kinds of courses. Taking courses in engineering and psychology, as well as the standard courses, just to sort of think of things from new angles with new tools. Then I'd also think about experiencing economics from different perspectives. That is working not only in academia, but working in government as I have, and maybe take a job in Wall Street.

Mankiw: One thing that I didn't do in my career is actually ever have a job at Wall Street. I kind of regret that, because I think if I'd worked in Wall Street for a summer, or a year, I would have learned about a different perspective on the economics profession. That would have been a good thing. I think to sort of bring a broad set of experiences, or a set of topics on your academic agenda. I think it's useful to be the most creative.

Beckworth: All right. Well, our guest today has been Greg Mankiw. Greg, thanks so much for being on the show.

Mankiw: It's been my pleasure.