David Beckworth on the Safe Asset Theory of Inflation, Comparing Central Bank Frameworks, and a Year of Macro Musings in Review

In this special end-of-the-year episode of Macro Musings, David Beckworth joins guest host David Andolfatto of the St. Louis Fed to discuss a wide range of macroeconomic topics, including podcast highlights from the whole of 2021. More specifically, both Davids talk about the similarities and differences between average inflation targeting and NGDP targeting, the recent inflation puzzles that have plagued the macroeconomy, David’s safe asset theory of inflation, and more.

Want to support the show? Visit donate.mercatus.org/podcasts. Your gift helps create enhanced transcripts like this one.

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 Andolfatto: Hello, long time listeners of the Macro Musings Podcast. My name is David Andolfatto of the Federal Reserve Bank of St. Louis, and I'll be filling in for David Beckworth today. Our guest today is none other than the creator of this podcast, David Beckworth himself. Good morning, David and welcome to your show.

David Beckworth: Well, thank you, David. It's a real honor to have a Federal Reserve official interview me on this podcast.

Andolfatto: Well, it's really, really great to be here and thank you for giving me the opportunity to interview you David, after extending the favor to me, at least on a few occasions. So before we get into economics, I'd like to take a little bit of time just to talk about you and your background and in particular, what motivated you to become an economist and how you ended up at the Mercatus Center?

Beckworth: Well, I'll try to keep this short. In college, I went to a small institution, a small liberal arts school. They didn't even have an economics major, I was a business major. I took the prerequisites in macro and micro and I really, really loved macroeconomics. It just clicked. In fact, I remember sitting with some of my friends who were, we were considered the brainiacs in the group and I was doing as well or better than them on the exams. They were all shocked. I was shocked, but I didn't register. "Hey, maybe you could do this as a career,” until much later in life. In fact, I went on and was working on an MBA, this is in the late 90s and I took another macro course.

Beckworth: It just really clicked again and it started me thinking, I really enjoyed that as an undergrad and I'm really enjoying it now in the master's level program and at the same time, I was reading, believe it or not, Paul Krugman's work. He had a lot of interesting stuff out like ... I believe the book was called “Pop Internationalism,” but basically it was him going off in all these international trade cranks, it was really excellent reading. He wrote a lot for Foreign Affairs too, so I found that really fascinating. It seemed to make sense of the world in a systematic way and then, the other thing I remember very vividly is I bought a used copy of the book, “The Secrets of the Temple” by William Greider.

Beckworth: So it's a very populist book, but it's a well-written book. It's one of these books that a journalist puts together, inside accounts and what happened during the Volcker years at the Fed. Man, I was motivated by this. In fact, I remember one time riding in the subway, reading the book and a lady stopped and asked, "What are you reading?" It turns out she was a consultant for McKinsey and so she kind of knew what I was talking about and stuff, but just really all these little pieces of the puzzles coming together and I said, "Okay, I'm going to do this. I like this. I'm going to do it," and then I headed off to grad school and as it turns out, I ended up at a place where George Selgin was working and he became my major professor, like on my dissertation committee and such, but I had him ... he taught some classes as well that I took.

Beckworth: So it was really just serendipity that I ended up at a place where he was. So he kind of broadened my thinking even more and then, many years later, he and I ended up both working in a similar place in DC, in the think tank space. Now, how did I get to Mercatus? Well, I was working at state universities. I was at Western Kentucky University, before that at Texas State University. So, my career path was kind of low key regional state school. I was a man of the people, David. I was there with the proletariat, hanging with the masses. In fact, I don't know if you've ever heard this story, David. I once taught a very famous baseball player named Paul Goldschmidt, maybe you've heard this story.

Andolfatto: I have not.

Beckworth: Yeah. He was in my money and banking class and he was extremely bright. The baseball fans out there know him. He's one of the best first basemen in major league baseball. I had a lot of student athletes come through institutions like this, big state schools. Anyway, he was really bright in money and banking. He was a finance major. They had to take the class and we had actually another class that we offered and it was like a portfolio management class. He actually managed real money as an undergrad, but it was a competitive class to get into it. I said, "Look, Paul, I will recommend you to the professors teaching this class, because you were like the best student in here." He was really torn and he thought about it and came back to me. He goes, "Dr. Beckworth, I really want to do this, but I got baseball." I said, "Baseball, are you serious?" I gave him a little spiel about, what are the chances of you becoming a professional baseball player versus becoming someone on Wall Street?

Beckworth: I tried to tell him how this class was amazing. All the students, who went through it, got good internships, went to work for Goldman Sachs, went to Wall Street. I said, "This is the path you need to take." Later, word got out and of course, the faculty laughed that I made this recommendation to him. In fact, when I was at Western Kentucky University, before I came to Mercatus, I got an email from Paul Goldschmidt and he says, "Dr. Beckworth, baseball turned out okay." Anyways, he burned me, but he's really a good sport. We chatted and caught up and stuff. So yeah, so that's where I've been working at before I came to Mercatus. What I also was doing was a lot of blogging and writing op-eds, popular stuff.

Beckworth: That's where we first met was in the blogosphere, David, back in the kind the aftermath of the Great Recession and that just really opened up doors for me, and I think for all of us who participated in that, that blogging, I got to know and meet people like you, it opened up opportunities for interviews. The internet was this great leveling field. I didn't have to be at a top 10 department to be a part of the conversation. So, the blogging and the op-eds, which eventually led to this opportunity at Mercatus where I am now.

Andolfatto: That's fantastic. I love that baseball story. I've been doing a little bit of research here and according to my notes, this is the 314th episode of your Macro Musings podcast. Is that correct?

Beckworth: That is correct.

Andolfatto: And you broadcast your first episode on March 31st, 2016. Your guest at that time was Scott Sumner. God, I think a lot of people might be curious, myself included, I mean, how you came up with the idea for this podcast and I've also been a little bit curious to know how you came up with its name as well.

The Beginnings and Background of Macro Musings

Beckworth: Well, the name actually followed my blog. My blog was called Macro Musings. So I just said, "Let's run with that. We'll be musing about macro stuff." As listeners know, it's macro, it's finance. We get to the operational details of central banking and such. The reason, I mean, I wanted to come up with this podcast, because at the time, it's 2016, as you mentioned, there were a lot of good podcasts out there and ones that I listened to. In fact, there were even economics podcasts but none of them, at least the ones that I was aware of, focus kind of singling on just macro and financial issues. So I thought this will be kind of a nice gap to fill, a niche to fill. No one else is doing it, so I'm not competing against anyone. I can be the first one out the door.

There were a lot of good podcasts out there and ones that I listened to...but none of them, at least the ones that I was aware of, focus kind of singling on just macro and financial issues. So I thought this will be kind of a nice gap to fill, a niche to fill. No one else is doing it, so I'm not competing against anyone. I can be the first one out the door.

Beckworth: Now, there are other ones out there now, but at the time, there wasn't at least that I was aware of. So, I got into it and it worked out great. Now, interestingly, when I was doing this, this is in 2016, so this is right when I started at Mercatus and I have two bosses. Well I have multiple bosses but two at the very top, Tyler Cowen, which many listeners will know. He's of course, a big name from GMU, but he's also the head of Mercatus. Then, I have another boss, Dan Rothschild, kind of the operational boss, but they took me out to lunch. Actually, I was doing a sabbatical there and then they were going to offer me a full-time job, but in the process of this lunch, I brought up the idea of doing a podcast on macroeconomics and they both kind of looked at each other, looked at me and like, "I don't know, David." Maybe seven, eight episodes we could see you squeezing out of that.

Beckworth: I think maybe what they were thinking was Dave is going to do seven episodes on nominal GDP level targeting and that's it. I was like, "No, no, no. What I want to do is something that kind of keeps up with current events, looks at literature, looks at all the unsolved questions, the history." There's so much as you know in macro that's unresolved and new things occur all the time. I mean, we have not had a boring moment these past two years of the pandemic. So, there's constantly great material for the show. So, I just feel privileged and delighted that I get to do it.

Andolfatto: Yeah. Well, we certainly derive a great benefit from it, but I mean, what particular motivates you to produce the podcast? Because it must be a lot of work. You produce a lot of episodes per year.

Beckworth: Yeah. It is a lot of work and it does take away from me doing research. So before coming here and doing research, like I just got tenure at Western Kentucky University before I came and I still do research but the podcast does seem to take up a lot of my time. There's a tradeoff involved and I'll come back to that maybe when I give you my last answer. So let me answer your question, your question, what motivates me to do it? I'd say three things. First is simply I enjoy it. It's fun. It's actually invigorating and I'm looking forward to conversations and kind of, you wake up, "Oh, I'm going to talk to David Andolfatto today." And you kind of get going. Any of these guests, I feel that way, I have to read their work, read their articles.

Beckworth: Sometimes it's a big book. I'll confess, some big books I've learned to skim very effectively, because there's a lot of work that goes into this but I just really enjoyed it and it challenges my thinking. I've had, for example, some post-Keynesians on the show and we've debated, what's the best way to deal with inflation. So for example, we debated what Paul Volcker did in the early 80s, was that necessary and of course, they would suggest ... the traditional approach, increased interest rates, tighten fiscal policy, tighten monetary policy. They would go more towards like doing price and wage controls, which I find to be very troubling, but I've had interesting conversations. They push back against what I think. So, I just find it challenging, invigorating, that'd be the first thing and I enjoy it.

Beckworth: Again, I get to talk to interesting people like you and the others on the show. The other thing I derive from this is that it gives me research ideas. It helps me think of new ideas, new things and I know we'll talk later about a working paper I've just finished and that came from conversations with you specifically, Ricardo Caballero, Alp Simsek, Adam Posen and others talking about safe assets and that wouldn't have happened if I hadn't done the podcast. Finally, my third thing, I'm going to come back to what I was saying earlier is I like doing the podcast, David, because it's a way around the gatekeepers of the profession. I don't have a great pedigree. As you know, I come from a state school but this podcast has allowed me to be a part of a national conversation.

I like doing the podcast, David, because it's a way around the gatekeepers of the profession. I don't have a great pedigree. As you know, I come from a state school but this podcast has allowed me to be a part of a national conversation.

Beckworth: I get invited to conferences, I probably wouldn't have otherwise been invited to. I get to speak to people I probably wouldn't have otherwise spoken to. I like to think I'm a part of a broader conversation, whereas, the typical traditional path would be go to a top school, get published in the top journals and then, people will start to take you seriously and maybe, you can write an op-ed and do all that. For me, I have found this alternative route has worked out much better for me. So, it's been a real opportunity and a privilege for me to be a part of this conversation via the podcast.

Andolfatto: I mean, I think you're probably going to motivate a lot of people to start their own podcast, including myself. I mean, so if this has been like quite a success I would say and I hear that you have fans of your show from all over the world and I'm sure you get stopped in the street and asked for autographs, isn't that right?

Beckworth: No, I have not been stopped in the street for an autograph. I have been asked for autographs, but not like that. I think I've signed one autograph just for the record, but I'm nowhere near that level. I do get stopped and approached by lots of people at conferences and other big gatherings for economists, which we haven't had in a while. So for example, back at the last AEA in-person in San Diego, January, 2020, I went to a party that the Bank of England was hosting. I believe it was Saturday night. All these central banks are hosting parties and I went to the Bank of England and I'll tell you why, because it was open to anyone. It's like, "Well, I'm going to go crash that party. I'm going to go see who I can find there." I found some friends there but lo and behold, I had multiple people from the Bank of England, come up to me and say, "Hey David, we listen to your show."

Beckworth: One of them actually said, "My wife is a big fan. She's not here and if I don't get my picture taken with you, she'll be very upset." As it turned out, his name was Arthur Turrell. He ended up coming on the podcast this year. He's nuclear fusion scientist. I'll encourage listeners to go check that out, but stories like that are pretty fun, pretty amazing. I also have been approached at conferences by Fed staffers and just other people who listen to the show. I've had several prominent members of the Fed leadership. I'll say at least two governors tell me they listen to the show and give them some feedback on it. I'd say probably where I see most of the reception and engagement is from people on Twitter, as well as graduate students. I get a lot of graduate students reaching out to me. So I think part of what I'm doing and what we're doing at Mercatus with the podcast is reaching the next generation of young macroeconomists and hopefully, pointing them in a good direction.

I think part of what I'm doing and what we're doing at Mercatus with the podcast is reaching the next generation of young macroeconomists and hopefully, pointing them in a good direction.

Andolfatto: Right. Fantastic. I think by any measure, that's success in my books. Well, in light of the guests, let's say the guests that you've had just this past year, 2021, what would you say were some of the best insights or ideas that you learned from your guests this year?

Ideas and Insights from 2021 Macro Musings Guests

Beckworth: Well, I have a few but I could probably draw lessons from all the shows. So this is not going to do justice to what we covered, but some things that really resonate, maybe this is again recency bias, things that just at least stuck out in my mind because of my own research agenda, but I'll start with Ricardo Reis. I had him on the show and he's always high energy, interesting to listen to. He's an amazing researcher, that seems to produce a lot of output but we're talking about the global dollar system and how the Fed had bailed it out, essentially or maybe that's a strong word, had backstopped the global dollar funding markets in March, 2020. There's a lot of people who have of concerns about that. That the Fed is always going to have to step in whenever there's a panic, it did in 2008, It did in 2020 and probably will have to do so in the future.

Beckworth: Are we not just incentivizing more risky behavior? Is there a way to maybe get those shadow banks, those shadow dollar markets into the formal system. So, there is that real concern and I brought it up with him, but he said, David, you got to do the right counterfactual. He referenced the paper by Matteo Maggiori, which we had later on in the year on the podcast. A paper by him and the late Emmanuel Farhi where they take a model and they show, if you don't have a dominant currency, like we have now with a dollar, you could have multiple currency regimes around the world, which could actually lead to greater financial instability. So he said, that's not necessarily the outcome. That is one outcome that comes from their model. So he goes, it's not clear that an alternative world would be better. He goes, "It might even be more financially unstable."

Beckworth: So the global dollar system we have with all its bugs and issues, maybe the best thing we could hope for which I found a very profound insight that I hadn't considered, so that would be my first one. The other one ... and this is tied to some of our conversations too David, but I had Markus Brunnermeier and Hanno Lustig, and they kind of reshaped my thinking of the fiscal theory of the price level. Again, you have too ... but Markus Brunnermeier in particular has this paper, "Fiscal Theory of the Price Level with a Bubble." I know there's controversy about the bubble and is it really something meaningful or not? What he shows is, the fiscal theory of the price level doesn't do a good job explaining places like Japan, at least on the surface. Now, I know there's ways to resolve it, and if you're very clever.

Beckworth: The way he resolved it is to say, "Look, there's a bubble term that we're leaving out. It's a more general approach and once you do that, it makes more sense,” but the thing I find fascinating about ... including that other term, whether you call it a bubble or a wedge in that equation, the fiscal theory of the price level ... and let me step back and explain that equation. I know David knows what I'm talking about, but for listeners, fiscal theory of the price level has this key equation. It's basically a government intertemporal budget constraint, that's been iterated forward through infinity. What it says is look, the real value of government debt today is equal to expected future primary surpluses discounted to the present, and the idea is that if the government can't finance itself, can't pay for this stuff in the future, inflation is going to occur and it caused the real value of those securities to adjust and therefore you can have inflation or deflation depending on the outcome.

Beckworth: What Markus Brunnermeier and what Hanno Lustig pointed to was that can't be all of it. There's got to be more and that's where that wedge term, and that bubble term comes in and the way I've come to appreciate it ... In fact, David, you pointed me to a paper by your former colleague and now governor Chris Waller. That term, that bubble term can be viewed as the demand for these securities, the real demand for it. That's an important consideration. To me, it kind of brings back kind of a quantity theory. I know you've written on that, so anyways, that'd be the second thing. Then, if I had something else to add, I don't want to go on too long here. Let me add maybe more of a casual anecdote, not so much a powerful insight, but I had Agustin Carstens on the show and he's now, I guess the leader, the president of BIS, but he also has been at the IMF and he was the head of the Bank of Mexico for many years.

Beckworth: So I asked him, what's the difference between being a central banker in the advanced economies and emerging markets, places like Mexico. He goes, "David, we're like street fighters." Advanced central bankers, they're very proper and do things in the right. We're street fighters. We get out and we have a knock down mixed martial art match with the markets. That's probably the way he described it. I found that to be a very powerful illustration. There are many other things I could go into, but I'll stop there for the sake of time and I apologize to all the other shows I didn't get to, but again, a lot of interesting stuff we covered and those topics are ones that maybe speak to my heart near and dear, because I'm doing research on them.

Andolfatto: Yeah. Sure. That was a bit of an unfair question. I know that every one of your guests in podcasts are interested in their own right. I was just wondering if anything stood out. Now, you were interviewed actually once before on this podcast by Cardiff Garcia back in 2019, and you guys were talking about your, then, new paper on nominal GDP targeting. Now, I recall that was a fun interview and we learned a lot about why you've been such a big proponent of nominal GDP targeting for many years now. I went back and I read the transcript of that show there and I see that even then, you could see that the Fed was not likely to adopt the proposal and indeed as we know, they went with the average inflation targeting framework instead. So can you tell us just briefly how the two frameworks are similar and how they differ? Also how you think the Fed's average inflation targeting framework performed over this last year and whether your own preferred proposal, a nominal GDP targeting framework would have resulted in anything materially different.

Comparing and Contrasting the Theory and Practice of AIT and NGDPLT

Beckworth: Well, I will begin to answer that three part question by saying, I think they are in practice, very similar, although on the surface, they're very different. One is an average inflation target. The other one is explicitly targeting total spending or total dollar incomes in the economy. Let me explain how I think they're very similar, and I think this is the way that the Fed is doing this average inflation. In fact, the Fed likes to call it the flexible average inflation targeting, that little qualifier there gives it some extra degrees of freedom. If you look at what I consider to be the intellectual forefathers of this, I would go back to Ben Bernanke. Ben Bernanke had his temporary price level target. Rich Clarida, the vice chair of the Board of Governors said in later speeches, this is effectively a temporary price level targeting approach.

Beckworth: If you look at Bernanke's early note on this, I think it was a blog post he did on this. He said that you want to do level targeting, effectively catch up or makeup policy when you're at the zero lower bound, but when you get back to the normal path or when you're outside the zero lower bound, it's back to regular inflation targeting. Also, he said, you want to ignore temporary inflation spikes or movements caused by supply shocks or energy changes that will fade, which is very interesting. Then later, Rich Clarida said the same thing, that it's an asymmetric approach. You can see through supply shocks. If you kind of put those pieces all together, what they're saying is you want to run demand really hot after real big demand shocks to get away from zero lower bound, and then you kind of go back to the normal level and you don't want to overreact to supply shock.

Beckworth: Well, that's exactly what a nominal GDP target does. In stabilizing demand, you're not worried about what's happening to inflation in the near term. Over the long term, of course, you hopefully keep it anchored. The way at least those two individuals described it, I think makes it very similar in spirit. In fact, your boss, David, when this was passed, I have a video clip, I've saved of him where he said, "Average inflation targeting is very much in the spirit of nominal GDP level targeting and price level targeting." So I think they're very similar in practice and in fact, if you look at nominal GDP right now, it looks like… if I didn't know any better, I didn't know anything, I just looked at the graph of nominal GDP targeting. If you told me that there'd been a recession…

Beckworth: I'd say, they must be doing some kind of level targeting, maybe even nominal GDP level targeting. So maybe to jump ahead to your second question, I think, nothing is perfect. There is ways things could have been done better, but overall I'd say, "Man, take the win. What an accomplishment." We had a quick recovery from a very deep recession. Now, it was both fiscal policy and monetary policy, but the Fed could have really messed things up. It could have tightened preemptively and it didn't and compare that to 2008, the recovery from that I'd say, wow, take the win. What an accomplishment. We did not have widespread bankruptcy. We didn't have another great financial crisis and I hope the big takeaway from this year as we move forward is not the inflation surge, but the victory we accomplished. The feat of returning the dollar size of the economy back to its trend path.

I hope the big takeaway from this year as we move forward is not the inflation surge, but the victory we accomplished. The feat of returning the dollar size of the economy back to its trend path.

Beckworth: I guess if I had to say how they're different and maybe how my approach would've been different, probably not a lot except for, I would like to think more clarity. I think if you explicitly target a nominal GDP level target, you're being a little more clear about it and I think it's fair, and you can't speak to this Dave, given your position. I think it's fair that there is a lot of ambiguity and uncertainty about what the Fed means. Now, for me, I think I can define it pretty clearly, they're going to respond like a nominal GDP targeter would, but there's so many people out there who didn't get that. I mean, smart people commenting on markets, completely, I think missed this level of nature to it, number one and then number two, that you see through supply side shocks.

Beckworth: I think clarity would be the big distinguishing thing and I know critics will say, "Well, nominal GDP targeting also has its communication challenges." So again, no perfect world, I guess to kind of wrap up this part of your question, again, take the win. This was an amazing accomplishment for policymakers. Hopefully, we learn from it.

Andolfatto: So you're happy, you're obviously, satisfied.

Beckworth: Yeah, especially if you consider where we could be right now.

Andolfatto: Where we could have been, well you're so pragmatic. Yeah, so as it turns out you have a new paper and that begins kind of with this startling question, the very first sentence is a question: what causes inflation? Now surely, economists like yourself and central bankers like myself know what causes inflation, or do we? I mean, can you tell us what motivated you to write this paper?

The Inflation Puzzle

Beckworth: Yeah. Now, this paper has taken a long time to write in part because I've been doing too many podcasts. No, this paper actually has marinated, that's maybe a better word. It got lots of great input, including from people like you. You've corrected some really egregious directions I was going in that paper. The reason I ask that question is because if you recall, around 2012 to maybe 2019, there was this inflation mystery. Why was inflation below the Fed's target? In fact, I look at 2010 to 2019, but on average the PCE inflation was maybe 50 basis points or so below the Fed's 2% target. There was a lot of coverage of this. In fact, in the article, I go back and I count up all the news stories and you see this huge spike in 2013. It comes down a little bit, but it's still ... there's lots of coverage on this inflation mystery.

Beckworth: I mean, there were newspaper articles with titles like, "The Inflation Quagmire. The Inflation Mystery. The Inflation Puzzle. The Fed's greatest challenge," and even Fed officials begin to say these things as well, I have some quotes in the paper, that kind of is the big motivation, why was inflation low? In fact, I went to a conference at Brookings, I believe late 2019, one of the last conferences I've been to. They had a whole conference on this, why the low inflation and they went through a litany of things from flat Phillips curves, labor market issues, globalization, technology. I felt, at the end of the day, that was incomplete. There was something I think that was very obvious, it was screaming to me and I actually raised this question in the Q and A. I didn't really get a good answer.

Beckworth: For me, the missing piece of the puzzle was the safe asset shortage story and I'll come back to that idea in a minute, but I just wasn't satisfied with the answers that were being given at the conference and by Fed officials. So, I don't know if this is a fair interpretation but my sense of what the Fed officials kind of settled on, some of them, I don't think your boss did but many of them settled on ... I think they acknowledged it's not a complete answer but what a lot of them settled on was a flat Phillips curve. That's kind of the most common explanation I saw and then, along with that, the Fed itself is a victim of its success and anchoring inflation expectations. I just didn't find those stories very helpful in understanding what happened in 2010, 2019, for several reasons.

For me, the missing piece of the puzzle was the safe asset shortage story.

Beckworth: First, there's been a lot of recent work out, not a lot, but some important work out recently on the Phillips curve. So Emi Nakamura, Jon Steinsson, Jonathon Hazell and Juan Herreño had this paper showing that the US Phillips curve is flat, all the way back to the 1980s. The BIS has a nice graph of a Phillips Curve for advanced economies that shows it's been flat for several decades as well. So, I think you can make the case, Phillips Curves have been flat for a long time, so why suddenly in 2010… the timing doesn't line up. There's got to be more to the story.

Beckworth: The other thing is it's a global phenomenon. There's low inflation going on around the world, not just the US, in Europe, Japan. Finally, the other thing that some Fed officials were hanging their explanation on, we've anchored inflation expectations, that's why it's so low. If you look at inflation expectations around the world, they're also beginning to gradually drift down during this time. It wasn't just headline inflation or core inflation. It was inflation expectations. So inflation expectations weren't as firmly anchored as they thought. So putting those pieces together, to me, the safe asset story made the most sense of it and I don't know, do you want me to get into the theory or do you have something to ask about?

Andolfatto: Well, I was just going to say, so you've identified kind of the issues here.

Beckworth: Yeah.

Andolfatto: The phenomenon of this low inflation “puzzle,” the various hypotheses that have been brought to bear on the question and your general dissatisfaction with all of these explanations, at least to some degree. So, I was going to ask, what sort of theoretical framework did you bring to bear on this issue? I mean, you mentioned the safe assets shortage phenomenon, maybe take a little bit of time to explain what this framework is and kind of how it helps you interpret this low inflation episode.

If you look at inflation expectations around the world, they're also beginning to gradually drift down during this time. It wasn't just headline inflation or core inflation. It was inflation expectations. So inflation expectations weren't as firmly anchored as they thought. So putting those pieces together, to me, the safe asset story made the most sense of it.

The Safe Asset Shortage Theory of Inflation

Beckworth: Yeah, so before I get to maybe the theory, let me just describe the safe asset shortage phenomenon and I think many of our listeners will know since I've had many guests on talking about this. Around the world, there was effectively a dearth or a shortage of safe stores of value, securities that would maintain their value in a very large financial crisis, some kind of systemic event. Of course, US Treasuries fit that description so do German government bonds and Japanese government bonds…

Andolfatto: Their nominal value, right?

Beckworth: Excuse me.

Andolfatto: They're nominal.

Beckworth: Their nominal value. Yes. Thank you, not the real value. The nominal value. Yeah. So it's a nominal value that's going to be expected to be preserved in a severe crisis and treasuries have been the biggest portion of that and there's been people doing a lot of work on this Ricardo Caballero and his colleagues have kind of traced this. In fact, Ricardo traces this shortage all the way back to Japan in the 90s when they had their big bubble burst. He said, that was the first incident of, there being a relative shortage of safe assets. They never fully recovered in terms of the supply. Then, he goes on to tell other stories, which is also common in the literature. I'll list a few of them. Demographics. Demographics is another reason, and what that does, it increases the demand for safe assets.

Andolfatto: Can I just back up a second there?

Beckworth: Sure.

Andolfatto: You mentioned Caballero and Japan as being the first episode of this so-called safe assets shortage, where the demand outstripped the supply. Now, surely you've seen the debt to GDP ratios in Japan. I mean, how can you possibly say or how can one possibly say that there's a shortage of these securities in light of what appears to be a massive increase in their supply?

Beckworth: That is a fascinating puzzle and one I attempt to answer in the paper, but I think you could ask that same question about the US today, right? We added five trillion of debt. So now our nominal stock of debt, the one that matters, marketable proportion is around 21, 22 trillion dollars, around 100% debt to GDP. Yet, the 10 year treasuries cannot seem to get above one and a half percent and it gets stuck. It's currently a little below that. Same thing in Japan. So, in the case of Japan, I mean, just to briefly answer your question, my interpretation is that the real demand for those securities still outstrips or is relatively high compared to the supply. So, I think the challenge for many people is that when you look at debt to GDP, you're looking at it from a supply perspective.

Beckworth: My goodness, it's 200% plus in Japan. It's 100% in the US and you forget. There's a demand side, a real demand for these securities and that's something I got from you, Dave, is you got to look at both sides of the market here. It's too easy to get hung up on one side and you mentioned this in your work. I mean, one of the key things you need to remember is some of this buildup is an endogenous response. People are desiring to hold those securities and you can think of like the Japanese government or Congress in the US, kind of maybe unconsciously, unwittingly, but they're providing those securities. So the world demands these safe assets, drives down interest rates. It kind of enables Congress to do more spending. Maybe it's not the best spending but it does more. So you can kind of use this growth and debt to GDP as an endogenous response to the safe asset shortage.

Andolfatto: That's an excellent answer, Dave.

Beckworth: I'd pass your macro prelim, huh?

Andolfatto: Well, yeah, I mean, I think that focusing on supply is a natural tendency, people implicitly have in mind kind of a stable demand, but economics tells us there's supply and demand and these things are revealed in the prices, ultimately, inflation and interest rates too. How did you go about using this theoretical apparatus to kind interpret the low inflation?

Beckworth: Well, let me lay out the theories to how safe assets map onto inflation, because again, one of the things that motivated me in doing this, wasn't just the fact that there's low inflation, but that there wasn't a good story and although there's been a lot of work on safe assets, there hasn't been a lot of like explicit work, trying to tie the low inflation to the safe assets. I mean, the implications are in the research, I mean, Ricardo Caballero and again, the late Emmanuel Farhi, [inaudible], they had a paper where they explicitly show the safe asset shortage has implications for inflation, but they're all theoretical that there's no like empirical side to it and it's definitely not like used as a way to explain what happened in the decade before the pandemic. So what I attempted to do was kind of map that literature onto what we observed in the world.

Beckworth: Let me just give you several ways you could do that and I'll tell you the one I ultimately end up doing. The first one, which is the Ricardo Caballero approach, kind of a new-Keynesian story. You have this elevated demand for safe assets. It drives down safe asset yields. Eventually you get zero lower bound and you still haven't reached the equilibrium rate maybe lower than zero and as a result, you get an output gap and that output gap in turn works through a Phillips curve to create low inflation. I know David, we're not big fans of Phillips curves explanations for low inflation, but that is one theoretical channel that's in the literature. Another way we could think about this is that the increased demand for safe assets will increase future seigniorage flows.

Beckworth: So, even though increased safe asset demand today is maybe more of a spot price or a current observation, it can be used to forecast the future and that forecast may lead investors to think, man, there's going to be a lot more seigniorage flows coming into the government, and that's going to make government debt worth more, lower the velocity of that debt and lead to lower inflation. That's kind of a fiscal theory of the price level story. I think there's some merit to that. Then, the angle I take, which is complementary to the one I just described is to view safe asset demand as part of a broader liquidity demand or money demand. So I tell a real money demand shock story in my paper. I model this in the sense that safe assets as we mentioned, they maintain their nominal value in a crisis like money. They provide liquidity services like money. They are effectively a form of money.

Beckworth: So, if you view a broad measure of money, everything from cash to treasury securities and in between, and even some private debt securities, if they're relatively safe then I think you can understand the safe asset shortage from that perspective and that's how I perceive in the paper.

The angle I take, which is complementary to the one I just described is to view safe asset demand as part of a broader liquidity demand or money demand. So I tell a real money demand shock story in my paper. I model this in the sense that safe assets as we mentioned, they maintain their nominal value in a crisis like money. They provide liquidity services like money. They are effectively a form of money.

Andolfatto: Right. So, I really like your interpretation of this, just in kind of general terms, it's kind of money demand or money is defined kind of much more broadly than it's commonly done. Just very quickly, what are some things that you think kind of have been driving this. It's a global phenomenon, right? So, I mean, to put it in ... basically we're talking about shifts in the money demand function. And then there are forces that are driving the money demand broadly defined to include government securities and other perhaps private label objects. Do you think there are any particular forces at work that are driving this? Is it regulatory perhaps or technological?

Beckworth: That's a good question. So I provide a list in the paper, although that's not the main objective of the paper, but I'll run through, I think the most common ones, demographics. The aging of the advanced economies. It's also the aging of China. We see China ... we can think of Japan at the forefront, avant garde of the aging crisis. Europe is behind them. We're behind Europe and behind us is coming China, and that's a big chunk of the real population. I know there's some debate about the implications of this for demand for securities. I know Charles Goodhart thinks demographics will lead to higher inflation, but everything we've seen. I mean, Japan to me is a good case study of this. Japan has had demographic problems for a long time now and all it has done is increased demand for these securities and led to lower inflation.

Beckworth: So I think demographics will be with us. It will continue to be a factor waning down and two ways I think of it affecting the demand for safe assets, one is you got to save more because you live longer. You're going to be in retirement longer, so there's increased demand for safe stores of a value, but secondly, it affects the distribution of your portfolio. You're going to be substituting out of equities or other securities into safer ones. So, I think demographics is a part of that. Another part of it ... and this story was told by, I think Ben Bernanke and others early on is just emerging markets. Now, he called it a global savings glut story but the basic idea is that parts of the world have grown rapidly because of globalization. China, India, they've gained a lot of income, but they haven't had the similar proportionate growth in their institutions that can support that growth.

Beckworth: So they don't have the institutional capacity for safe stores of value, so what do they do? They go looking elsewhere in the world and they come to the US. I think that's been an important story. Probably still will be, maybe not as much going forward as these countries get richer. Demographics is probably more important. Then, a few others might be the regulatory angle that you mentioned, out of Dodd-Frank, banks have to hold more safe assets, other institutions as well. There's been some other stories. I mean, income inequality. There's the argument there that maybe that's driving part of it. Then, maybe, the last thing I'd mention is financial innovation. I know you've touched on this in your work. And maybe financial innovation really is just a way of restating what we've covered, but financial innovation. So stablecoins, money market funds, all these things need some kind of safe asset behind them. Maybe this is just kind of the technological manifestation of the existing safe assets of man, but those are the things I would throw out there. Did I leave anything out, David, that you would add or ...

Andolfatto: I think that was quite a comprehensive list. So yeah, this is very intuitive. I think most people can grasp this. If we hypothesize that despite the large apparent supply of these global safe assets in the form of JGBs or US Treasury securities that's not withstanding this large supply, evidently, the global demand for the product has been outstripping even that supply. That has resulted in downward pressure of the interest rates, and to the extent that interest rates can't adjust very much more close to the lower bound, we see the manifestation of this excess demand so to speak, to manifest itself in lower output, output gaps, low inflation, you can have Phillips curve interpretations, if you want or standard monetarist theories, will tell you the same thing as well.

Andolfatto: So conceptually, I think we have an idea here of how this safe asset shortage might contribute to this low inflation episode that has been puzzling Fed policymakers and other people for a long time. Now, the question is, this is not just what you do in your paper, right? I mean, there's a question of quantitatively, how important might this affect me assuming it even exists. So you also take some time in your paper to do some empirical investigation to address the quantitative question. Can you explain how you go about doing this?

Empirically Investigating the Safe Asset Theory

Beckworth: Well, I motivate the empirical part with a theoretical part, not real complicated, but I do have a money in the utility function model and I just kind of show it for the household kind of optimization side of it. I don't do a full DSGE, but I do optimize the household part of the problem and I show, if you include a monetary aggregate that includes government bonds as a form of money, and you get these nice results, you get results that show convenience yield on treasury bonds and you find a lot of interesting results that kind of feed into it. In particular, you can show that there's a common money demand shock that affects traditional money assets, as well as these institutional money assets like Treasury and Treasury yields. In fact, one of the results that falls out of the model is a money demand shock would cause both money velocity to go down as well as the convenience yield to go up or to put it more simply the yield on safe assets to go down.

Beckworth: So you get the results that we tend to see and if we actually look around, so if you look at ... for example, 10 year Treasury yields, they are very correlated with measures of money velocity. They're almost near opposite of each other, which is how do you explain that? And I think this is the way you explain it, there's a common money demand shock driving both or maybe another way of saying money demand shock driving all these safe asset yields and uses of safe assets. Okay. So empirically, I set up a vector auto regression for the US and then I do a panel VAR for 11 advanced economies just to see if these results carry through. What I find ... I'll just briefly describe the impulse response functions, that a money demand shock does lower Treasury yields, does lower velocity. It also lowers inflation and it also leads to central banks increasing their balance sheets in response and policy rates dropping. So you get all the theoretical outcomes you'd expect but how important, and that's the question you're asking, so why?

Beckworth: It's an interesting story, does it really matter? With VARs, it's a little tricky. It's not like a typical regression, you come out with an R squared of ... “I explain 50% of the inflation.” It's a little different, but I still have a way to kind of get a handle on this. Two things, first do a forecast of variance decomposition. So you look at the percent of the forecast error that the model generates, empirically that can be explained by these shocks. What I end up with is about a quarter or 25%, at most, about 25% of the forecast error and inflation. I look at headline core in the deflator. So, it's not like a majority, but it's still an important ... you imagine 25% of the inflation being driven by these money demand shocks, safe asset shocks, that's definitely a significant number. And that's over the whole sample. So it doesn't really maybe explain specifically 2010 through 2019.

Beckworth: So in order to do that, I do a counterfactual exercise with the estimated model and yes, we have the Lucas Critique here in the background, screaming at me when I do this, but what that counterfactual does is it allows us to get a sense of magnitude and that's again, answering your question. So what I do is I take the estimated model and I start from 2010 and I do a dynamic forecast forward using this model where I plug in values for the 10 year Treasury yield as if they would've returned to a more normal pre-2008 level. So the assumption here is that somehow the government satiates safe asset demand, issuing more debt or fiscal policy, monetary policy working away, so that we have solved this problem and we return, the 10 year Treasury ... and I do over different levels, four and a half, five, five and a half percent.

Beckworth: What happens, and what I find is on average, you'd have inflation about 1% higher over this period than we currently experienced. So, you think of the PCE being about 50 basis points below two. My results say, if you could get the 10-year Treasury yield up to five and a half percent, then you would have inflation about two and a half percent. So it is a sizeable number, again it's subject to the Lucas Critique, but it gives some sense of magnitudes. I also did a counterfactual for nominal GDP using this approach and it's several trillion dollars larger by 2019. So what I find is that both the yields on Treasury, the inflation rate and nominal GDP, all are affected by this safe asset demand as a part of a broader phenomenon of real money demand shocks.

What I find is that both the yields on Treasury, the inflation rate and nominal GDP, all are affected by this safe asset demand as a part of a broader phenomenon of real money demand shocks.

Andolfatto: So that's a sizable number I would say. If you were ... I don't know, just based on this work and perhaps what else you might know or believe. I mean, if you were in charge, say Congress or the Fed together, let's say you're very powerful, let's consolidate the two, I'm not sure if your model speaks to this, but what would you kind of actually do as a practical policy matter to kind of alleviate or solve this safe asset shortage issue? I'm not even going to call it a problem. Is it a problem? What are the implications here exactly for policy. I mean, perhaps we know we might be able to get the inflation rate up higher, but would that have been a good policy and why? I mean, how do you think about this question?

Policy Implications of This Safe Asset Theory

Beckworth: It's a good, tough question. So maybe problem is the wrong word. I think there's several ways we could address this issue, this question. One of the reasons I think we can call it a problem is because markets aren't doing their work. The reason we have a safe asset shortage, because yields get down to the zero lower bound or the effective lower bound, maybe it's a little bit lower and rates aren't allowed to drop farther down, which another way of saying that is prices aren't allowed to clear the market. So, markets aren't working, so you could think one possible solution would be, well, let's remove that institutional constraint let's make negative interest rates as common as positive interest rates. Let's allow that flexibility into play. That could be one possible solution. The Fed starts doing negative rates, at least theoretically. I'm just not sure policy wise how practical that is.

One possible solution would be, well, let's remove that institutional constraint let's make negative interest rates as common as positive interest rates. Let's allow that flexibility into play.

Beckworth: So, we've seen the ECB do it and the ECB goes a little bit below zero, I guess Sweden and Switzerland, Denmark have had some negative interest rates, but I'm not sure there's the political will to really do what would be needed on negative interest rates, because politically very sensitive thing to sell to the public. We want to give you negative interest rates.

Andolfatto: Well why not ... I mean, if there's a shortage of safe assets, I mean, it seems like the obvious solution sounds like is just to make more of them.

Beckworth: Yeah, well, that's, that's another solution. Issue more than ... I was going to get to that as well, right? So instead of trying to play at the price, move that barrier, just let's prop it back up by issuing more securities. As you know, I mean, as we've talked about today, people look at debt to GDP and they freak out about that too, right? They freak out about negative interest rates. They freak out about the supply of nominal government debt. So the solution I've thought about ... in fact, I'm working on a paper, another paper with a colleague at Mercatus about this, is assuming this continues to still be a problem and all signs point to that again, 10-year treasuries, 20-year treasuries are at super low values, then it might behoove the US government to set up a sovereign wealth fund, where it takes some of that fiscal space, some of that capacity.

Beckworth: This sovereign wealth fund effectively issues its own treasuries, takes in the money, reinvest it in higher risk stuff, earns a dividend for the population. This could be kind of a form of the UBI for the masses or maybe if you don't want to hand it out, you save it for rainy day. I mean, Norway has the sovereign wealth fund they use with their oil resources. So one way to look at this, if this continues to be a problem, you can think of the US having the resource issue, where we're the main provider of safe assets. So we're not providing oil, we're providing safe assets, find a responsible way to use it. One thing I am concerned about debt ... well, there's a couple issues, I guess, on debt but one thing that does bother me is that the safe asset demand lowers yields. It enables Congress to spend, in my view, recklessly sometimes or without much thought is this really going to have a high return on the amount of spending than we're doing? I think a sovereign wealth fund would impose some discipline on that gift from the world to the US.

Andolfatto: Right. Very good. Well, we're butting up against the time here, Dave. Before we go, I just wanted to ask you quickly, what big lessons did that pandemic teach you about economic policy? You talked to a lot of people, you've given a lot of thought … I know you've been a student server of policy responses, both domestically and around the world. What big lessons did you learn from the pandemic in this regard?

Economic Lessons from the Pandemic

Beckworth: Well, I think acting quickly is probably a lesson we've learned. We mentioned on the macro policy front, the government did act relatively quickly, maybe not at the most efficient manner, but acted quickly. I've had Alex Tabarrok on the show and he's been a big critic of the FDA's ability to get these vaccines through quickly. He's also been, I think, very astutely, correct on better planning for tail events. So he's also a colleague of mine. He's at GMU, he's also at the Mercatus Center. He's advocated for more planning for tail events like an asteroid hitting earth or a solar flare knocking out our electric grid, things like that. We need more preparation. We don't think about them and then, when they do hit us, we're not prepared for them.

Beckworth: So I think it's important to be thinking about tail events to have relatively quick, macro responses and that may mean, you just set it up ahead of time. So maybe bigger automatic stabilizers, a better monetary policy rule and such. Again, the other thing I would again, encourage listeners, if any government official is listening to this, any central banker is listening to this, look at what was accomplished, the big takeaway from this pandemic. We closed that hole so quickly, so I encourage them to take the win as I mentioned earlier.

So I think it's important to be thinking about tail events to have relatively quick, macro responses and that may mean, you just set it up ahead of time.

Andolfatto: Fantastic. Thank you so much, David. It's time to wrap up here because I know your podcast only runs for an hour. I could do this for another hour, but I just want to end here, to say thank you so much on behalf of all of your listeners, all of your fans and myself included. I really enjoyed listening to your podcast and have had the honor of actually being a guest on your show on more than one occasion. Thank you so much and we all look forward to the next 314 podcasts.

Beckworth: Fantastic. Well, thank you David and I guess this means I get to get a nominal GDP targeting mug now that I'm a guest.

Andolfatto: That's right. Merry Christmas.

Beckworth: Merry Christmas. Happy new year, David.

Andolfatto: You too. Thank you.