Sam Schulhofer-Wohl on How to Improve Treasury Market Resiliency

Sam Schulhofer-Wohl is the Senior Vice President and the Senior Advisor to the President of the Dallas Fed, Lorie Logan. Sam is a longtime veteran of the Federal Reserve System and has also previously served at the Minneapolis and Chicago Federal Reserve banks. Sam joins David on Macro Musings to talk about Treasury market resiliency issues, the floor system, the Friedman Rule, bank deposits, the monetary policy implications of labor migration across the United States, and much more.

Check out our new AI chatbot: the Macro Musebot!

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: Sam, welcome to the show.

Sam Schulhofer-Wohl: Thanks, David. It's a real pleasure to be here. Thanks so much for having me on.

Beckworth: Well, it's great to have you on. Your name was first mentioned to me by Ellen Meade, another former Federal Reserve individual. She said, "You have to get Sam on the program. He's a great guy filled with many wonderful ideas." So, I tracked you down and here we are, and I'm excited to have you on the program today, because you do have a very interesting research agenda, [have] covered a lot of topics, and you have an amazing journey, too. I'm just looking at all of the places you've been. In fact, you've been a journalist at one point in your life, if I understand correctly.

Schulhofer-Wohl: Yes, correct.

Beckworth: So, walk us through that career path. [It’s] very fascinating.

Sam’s Wide-Ranging Career Path

Schulhofer-Wohl: Yes, well, thanks again. I hope I live up to what Ellen said about me. Ellen is one of my favorite Fed colleagues, so it's really nice to hear that. So, as you said, I've had a bit of a nontraditional career path. I started out as a newspaper journalist and then went back to grad school in economics and then came to the Fed. So, I'll tell you a bit about all of that. Before I get started, I do want to give the standard Fed disclaimer that anything I say about my career or economics or anything else are just my views and not those of the Federal Reserve Bank of Dallas, or the Federal Reserve System, or any of my colleagues.

Schulhofer-Wohl: So, that might be my first piece of career advice for anyone who's aspiring to be a Fed economist, is always remember to give the disclaimer. With that said, as you mentioned, I started out as a newspaper journalist. That's not a path that would be very easy to follow today. I just looked this up, [and] according to the Bureau of Labor Statistics, when I graduated college, there were 427,000 people, nationally, working for newspaper publishers. Today, there are only 88,000. The industry was really declining, and that was a big factor in my career choices.

Schulhofer-Wohl: So, I worked at a number of newspapers around the country, and in 2002, I was covering the education beat in Milwaukee at the Milwaukee Journal Sentinel. And there was a really intense debate going on among economists at that time about Milwaukee's school voucher program, which is one of the first in the country and was really a pioneering experiment in giving low-income parents vouchers that they could use to pay tuition for their children to attend private or parochial schools.

Schulhofer-Wohl: So, this is a policy approach that economists have debated for many years, going back to folks like Milton Friedman, as to whether this can both help the individual students and create competition that might foster a stronger public school system, and there are views on both sides. [There is] a lot of debate about a lot of papers being written by folks like Caroline Hoxby [and] Jesse Rothstein. It's an incredibly tough research problem. It's really hard to get the right control group for this experiment.

Schulhofer-Wohl: So, I was reading these papers. I was writing stories about these papers. You could see that none of the research was perfect, and I had the hubris to think that maybe I could do better. So, as I was thinking about other career choices, given what was going on in the newspaper industry, economics came to mind, and I'd also always been very interested in economics. I took not quite enough courses in economics at Swarthmore to have a double major.

Schulhofer-Wohl: I had some just incredibly inspirational professors, folks like Amanda Bayer, who was both my first professor for ECON 1, and then taught an economic theory seminar later in college. So, economics really came to mind in addition to what I was covering at work, and I had this ambition to do economics of education and off I went to the University of Chicago for a PhD. I very quickly learned just why economics of education is so hard, and I moved on to other topics, and I never wrote the first thing about it.

Schulhofer-Wohl: So, I greatly admire the folks who are trying to understand those very important issues in education, and I wish I had found a way to contribute to that, but I've worked more on macroeconomics, econometrics, and some other topics. So then, after graduate school, I started out, as many people do, as an assistant professor. Then, after a few years, I moved to the Minneapolis Fed, largely for the research environment. It was and remains just a phenomenal research environment.

Beckworth: It’s legendary, right?

Schulhofer-Wohl: Yes, it truly is. People there are so excited about economics, so engaged, [and] so committed to the power of economics as a tool for understanding the world. So, I knew that my research would take off if I went there, which it did. But I also really fell in love with the Fed's policy mission when I got there. In monetary policy, we've got a dual mandate, maximum employment and price stability, and if you think about the employment side of that, the U.S. labor force is 168 million people. So, that means that a tenth of a percentage point on the unemployment rate is 168,000 more Americans who have jobs. That is just incredibly impactful, the work that we do every day. And of course, the Fed is this huge institution, [and] there, each of us is a small part of this big machine working to achieve that mission, but what we do together just matters enormously, and that really inspired me and has driven my career since then.

Schulhofer-Wohl: So, over time, I've moved around to a lot of parts of the Fed's work. I've probably gone against a lot of career advice in doing that. The standard advice for early career PhD economists, whether you're in academia or in public policy, is to become a deep expert in some particular area. So, when people say, “Who is the best at X? They're the best at X,” I've done the opposite. I've been a generalist and tried to apply the tools of the ways of thinking of economics to whatever the problem of the day is. So, I've done research on labor markets and geographic mobility, as you mentioned.

Schulhofer-Wohl: I've done policy work on money markets and financial stability. For a time, I oversaw the public affairs and community development departments at the Chicago Fed. Right now, as you mentioned, I'm senior advisor to President Logan at the Dallas Fed. So, I get to advise on a full range of policy issues facing the bank, which is really fascinating. And I also lead a small team that works across the bank to make sure that we're integrating perspectives from all of our professionals and all of our business lines to get the best answer to our problems.

Schulhofer-Wohl: So, just as an example of that, we were thinking about, how is artificial intelligence going to affect the economy? Of course, there's a macroeconomic perspective on this, but you can also think about, what is this going to do in financial institutions? What is this going to do to low-income individuals and households and communities? What perspective can we bring to it from community development? What are the implications for the financial system as a whole and for financial stability beyond financial institutions? Also, really, how does this technology work at the nuts-and-bolts level? Do our folks in IT have something to say about this? So, in the team I've got now, we get to make all of those connections. It's just endlessly fascinating and a lot of fun, and also, of course, I think, very impactful for the public. So, I'm grateful to have had the opportunity to do this career.

Beckworth: Sounds like an amazing journey, lots of different directions you've taken. And I think the point you make about being a generalist probably is really useful in the role that you're in now. You're advising a policymaker. You're her senior advisor, and if you are too narrowly focused in an area, you don't have the breadth of knowledge to weigh in on these matters. So, I think that that general wide scope of a research agenda has really probably paid off for you in that role. So, tell us about that role. What do you do on a day-to-day basis with President Lorie Logan at the Dallas Fed?

Schulhofer-Wohl: Every day is different. There are days where I'm working on a long-term project of one sort or another. We may be thinking about an issue that we see out there, and we want to understand it more deeply, or we may have come to a deeper understanding of some issue and want to put out a view on it, for example, in a speech or in preparing for public events. Sometimes we're organizing conferences, sometimes we're doing policy analysis within the bank on shorter-term issues, and sometimes, events take over.

Schulhofer-Wohl: So, we had a banking crisis in March of 2023, and in a situation like that, you are not sitting and thinking, "What would be the best thing that we could do if we had five years to think about this problem?" Now, someone early in my Fed career said to me the difference between policy and academia. In academia, you can take as long as you want to get a really good answer. In policy, you have to get an acceptable answer on the time horizon that the decision has to be made. So, when there are bank runs going on that are the fastest bank runs that we've ever seen, you have to work really fast. That doesn't mean that you care any less about the quality of it, but part of quality, in that circumstance, is thinking about things quickly.

Beckworth: Yes, and we'll talk about a paper that came out of some of that work that you were doing on the bank runs of 2023, which is really interesting, and sometimes those crises reveal new insights to us. And the paper that we're going to talk about later is one such example of that. Now, one more thing on your career before we move into your research, and that is your time at the Minneapolis Fed. You were the research director, if I read correctly.

Schulhofer-Wohl: Yes. So, I started out as an economist and then eventually was research director for a few years, which— [it’s] just an incredible privilege to lead this amazing team of great economists. I thought, in that situation, that the best I can do is just create an environment for this amazing group of people to do their best work.

Beckworth: So, there's a lot of interaction between that research department, the Minneapolis Fed, and then the University of Minnesota's Econ Department, correct?

Schulhofer-Wohl: Yes, really great engagement. Many faculty from the University of Minnesota, at least when I was there— I don't want to speak for how it works today, though, [and] I think that my sense is that not much has changed— But many faculty from the university had offices at the bank, spent a lot of time at the bank, and many economists at the bank taught courses at the university. Most of the research assistants were graduate students from the university. So, it was not quite completely integrated departments, but really an enormous amount of interaction and a real force multiplier, I think, both for the bank and the university.

Beckworth: I just think of some of the great papers and research that's come out of the Minneapolis Fed Research Department, like Sargent and Wallace in the early '80s. Chris Simms was there, and just a lot of influential game changers for macroeconomics, and now that you've been a part of that tradition, that's just really fascinating.

Schulhofer-Wohl: Yes, a great history there and great work to this day, there. I still love reading the papers that come out of that department.

Beckworth: Alright, well, let's turn to your research, and you, again, have a number of areas that you've worked on, but the first one that I want to go to are issues related to the Treasury market and the resiliency, or making it more resilient. You started your work on this, as I understand it, based out of Chicago. You're at the Chicago Fed, so, you saw things there from a unique perspective, and then you built upon that. And you have a paper in the Journal of Financial Market Infrastructures in 2022, titled, *The Customer Settlement Risk Externality at U.S. Securities Central Counterparties,* and that's where we want to end up. Maybe walk us through the journey in getting to that paper.

The Customer Settlement Risk Externality

Schulhofer-Wohl: It was quite a roundabout journey to the research I did on the Treasury market. So, if you remember, in early 2021, there was this whole GameStop event, where people were talking about GameStop stock going to the moon, and everyone, it seemed, on social media wanted to buy it. And as it went up and up, suddenly, some stockbrokers wouldn't let people buy the stock even if they had money in their account. And people started saying on social media how upset they were about this, that this doesn't make any sense.

Schulhofer-Wohl: "My money is there. Why won't they let me buy? What risk can the stock brokerage be worried about in letting me buy it? It's not like I'm trying to sell something that I don't have. It's not like I'm at risk of losing anyone's money other than my own." And so, it turned out that if you read carefully what the stockbrokers were saying, they were saying, "If we let people buy, we're going to have to send some money as margin to something called the National Securities Clearing Corporation.”

Schulhofer-Wohl: “We're going to have to send our own money, even though it's the customers [that are] buying and the customer's money that is in the account, and that's getting really expensive to do, because the stock is going up and up and getting more and more volatile. And so, we don't want to be exposed to that, and so, that's why we had to stop them from buying.” Sitting in Chicago, I was, among other things, overseeing our financial markets team in Chicago, which monitors the derivatives markets and the infrastructure of the derivatives markets, the central clearing houses, in Chicago.

Schulhofer-Wohl: That was just incomprehensible to me, because in the derivatives markets, when someone trades futures or options, there are also these clearing corporations, and I'm going to come back to what these things are. And their brokerage does have to send money to protect against the risks involved in the trade, but they send the customer's money. And so, you just wouldn't have a situation, really, in derivatives where they're saying "The customer's got money in their account and I won't let them spend it."

Schulhofer-Wohl: It just didn't make any sense from the training that I had and the knowledge that I had of how the markets worked in Chicago, and as a trained PhD researcher, what do I do? I go off and read the rule books of all of these clearing corporations in gory detail. These things are long and detailed and they specify exactly what happens in lots of situations. And I came to understand that the rules of the game in the securities markets, both the equity market and the Treasury market, were just fundamentally different from how this works in derivatives and in the treatment, in particular, of customer money and customer trades.

Schulhofer-Wohl: And so, I'll come back to this, maybe, after I walk through a bit more of what's going on in the Treasury market, where my research got to, but what I understood was there was an externality created by the rules of the game in the securities markets that didn't exist in the derivatives markets. If you grew up in derivatives markets, you'd think, "This doesn't make any sense," but if you grew up in securities markets, it was the most natural thing in the world, because this externality was just completely baked in the rules of the game.

Schulhofer-Wohl: So then, I wrote a paper explaining what is the externality and how you can fix it. I came to it from the GameStop saga. As interesting as meme stocks are, the Treasury market is vastly more important than meme stocks for the financial system as a whole. And so, this issue also arises in the Treasury market and [is] connected to some bigger-picture risk and resilience issues that people were worried about at that time. So, I wrote this paper that was thinking about, how do you fix a problem that showed up in the stock market? It turns out that the biggest use of that research so far has been to work on making the Treasury market stronger. So, that's how I got to it, but maybe let me step back and just talk about, what is the Treasury market and what are the issues that people are thinking [about]?

Breaking Down the Treasury Market

Schulhofer-Wohl: As I said, it's a hugely important market. The saying is that it's the deepest and most liquid in the world. Why is it so important? Well, number one, the Treasury yield curve is the benchmark credit risk-free rate for the US financial system, and to a large extent, around the world as well. It's a very important source of financing for the US government. Then, on the flip side of that, for the people providing the financing— so, households, businesses, financial institutions— it's a really important source of safe and liquid assets.

Schulhofer-Wohl: Then, in addition, it's a big part of how the Federal Reserve implements monetary policies in this market. So, for all of those reasons, this is a market that we, as a country, need to work really well, and it does, by and large, work really well. But as the world evolves, as the financial system evolves, you've got to keep up with that so that it keeps working really well. It's just a really large market. There's currently about $27 trillion in Treasuries outstanding, what they say held by the public. That includes the Federal Reserve and foreign governments, things like that. But at about $27 trillion in Treasuries outstanding, it's up about $2 trillion over the past year, so, just a vast market, and it's got lots of different segments. There's the primary market, where the Treasury sells these securities, in the first place, to new buyers. There's the secondary market, where people trade existing securities with each other.

Schulhofer-Wohl: There's the repo market, where people borrow against Treasuries as collateral. There's the futures market, which I knew the most about as I started on this, which is at the Chicago Mercantile Exchange, where people trade derivatives that are pricing future interest rate risk based on Treasuries. Then, you can subdivide each of those markets. So, in the secondary market, you can think about the market for newer securities, what are called on-the-run versus older securities [which are] off-the-run, or you can think about trades between two dealers versus trades between a dealer and its client.

Schulhofer-Wohl: In the repo market, you can think about whether a third-party custodian is involved, whether there's a central clearing corporation involved in the trade or not. So, [there are] just lots of pieces of this huge diverse and important market, and as I said, most of the time it all functions super smoothly, which is what makes it the deepest and most liquid in the world. Also, it makes it very noticeable when there are any glitches. And so, there have been some disruptions.

Schulhofer-Wohl: In March 2020, at the start of the pandemic, we had the dash-for-cash. In September 2019, there were pressures in the repo market. There was a flash rally in 2014. There've been other smaller events. And so, observers have pointed to various vulnerabilities that these different events have exposed. They point to that there's potentially limited intermediation capacity in the market. So, if a buyer and seller have to find each other, typically that goes through a dealer that helps the buyer and seller.

Schulhofer-Wohl: First, finds the seller, then finds the buyer and matches this up, and there may be limited capacity to do that. People have pointed to the potential for very large one-way flows all of a sudden, and this stems, really, from the importance of Treasuries as a liquid asset. Because they're so valuable as a liquid asset, people sometimes might all want the liquidity from that all at once, as we saw in March 2020. And so, the market can be vulnerable to large one-way flows. People have pointed to various kinds of gaps in risk management, and also to a lack of visibility into, actually, just what's happening in the market, and that's— both those last two, the risk management and the visibility, are where my research really connected in. There's really substantial public and private efforts underway to ensure that the market continues to evolve to best serve these important roles.

Schulhofer-Wohl: And I've had the privilege to help out what's called the Inter-Agency Working Group for Treasury Market Surveillance, which is a partnership of the Treasury Department, the Board of Governors, the New York Fed, the Securities and Exchange Commission, and the Commodity Futures Trading Commission. I had the privilege to help them out with their recent reports on the vulnerabilities in the market and the work streams that they're thinking about to enhance the market strength. Then, as I said, I also did some research on these risk management and transparency aspects related to central clearing, and these changes in central clearing are the focus of a recently adopted rule by the SEC, which they adopted in December and will come into effect over the next couple of years.

Beckworth: So, tell us more about central clearing. How will it ease some of the stresses that we've seen on the market? For example, would it have made much difference in March 2020 or [during] the repo [crisis of] 2019, or any of these other events? What contribution will it make and how?

The Importance and Effectiveness of Central Clearing

Schulhofer-Wohl: So, there's some analysis that suggests that it may make some difference if you had a similar event to some of the ones we've seen. It also can make a difference in a lot of other scenarios that we perhaps have not seen yet. I think that it's also important not to try to only solve the most recent problems or the problems we've already seen. You don't want to just be fighting the last war. So, that's a good angle to ask what difference would it have made, but it also addresses other issues. Just stepping back, most securities trades are not cash on the barrelhead. So, if people are buying and selling a security, they don't show up and agree to the trade and, at that very instant, trade money for the security. Usually what happens is that you agree to the trade and then exchange money and the security the next day, which is important, because it gives you time to organize, operationally. Get your hands on the security, get your hands on the cash, and make the trade.

Schulhofer-Wohl: But that delay— even though it's very helpful for not having to show up at the market already with your cash and security— it introduces settlement risk, because maybe one of the parties to the deal won't show up tomorrow to complete the trade. And if they don't show up there, their counterparty is going to have to find a replacement trade, and the price might have moved against them in the meantime, and then they'll be out some money. So, it creates a risk, and this is called settlement risk— the risk that your counterparty won't show up and finish the deal. Similarly, in a repo, which is a financing transaction— So, I borrow money against a Treasury security, similarly, overnight. If I borrow money, there's a risk that, tomorrow, I don't repay the money. Now, if I don't do that, my counterparty is still holding that Treasury security is collateral. So, they're not completely out, but they now have to sell that security, which is inconvenient to them. They may not get the price that they hoped for, and so, they face some risk.

Schulhofer-Wohl: So, in these transactions, there's this risk that someone doesn't show up tomorrow to finish the deal— settlement risk. With central clearing, an organization called a central counterparty, or CCP, steps into the middle of the deal and guarantees both sides. So, what we say is that they become buyer to every seller and seller to every buyer, right? So, if you and I agree that I'm going to buy a Treasury from you, and now we have to show up tomorrow, and you face the risk that I don't bring the money and I face the risk that you don't bring the security, [then] the minute we make that agreement, we hand it over to, currently, what's called the Fixed Income Clearing Corporation. It's currently the only central counterparty in this market. If that trade were centrally cleared, we'd hand it over to them. And now, tomorrow, I have to show up to FICC with the money. You've got to show up to FICC with the security. And if either of us doesn't show up, FICC just deals with that person, but makes the other person whole.

Schulhofer-Wohl: So, it makes it a whole lot easier to trade, because now I don't have to worry, and you don't have to worry, is David a trustworthy guy? Is Sam a trustworthy guy? We both just know that FICC, which is the central, very large, very well-run institution, is trustworthy. So, it's very helpful to a market, in promoting liquidity, to have that. As you can imagine, it also really strengthens the risk management, because if the central counterparty is going to guarantee the deal in this way, [then] they have to impose some risk management requirements so that, in fact, the deals will go through. They can't just let everything happen. Then, they wouldn't be able to give the guarantee. They impose uniform professional standard risk management rules on the market to make their guarantee, and that's really important.

Schulhofer-Wohl: And so, if you talk to market participants, they will all tell you that their firms have strong risk management and that it's maybe the other folks who are perhaps not so stringent about risk management, but I can see that there's the potential for a race to the bottom here and a central counterparty, a CCP, stops that race to the bottom. So, just as one example of this, on repos, a financing transaction where you borrow money against the Treasury, a standard risk management tool is what we call a haircut— saying that you can't borrow the full amount of that Treasury security that you've given as collateral. You can maybe borrow only 98% of the value, so a 2% haircut.

Schulhofer-Wohl: So, that's the standard way of managing the risk. The Office of Financial Research, last year, did a pilot data collection of a type of Treasury repo called non-centrally cleared bilateral repo. So, these are just private deals between two firms that don't involve any third custodian or central counterparty. They just cut the deal themselves, and it turns out that 74% of non-centrally cleared bilateral Treasury repos had zero haircut, no haircut at all. So, when you have a central counterparty, they stop things like that from happening, because they say that there are uniform rules here. We've got professional risk managers. We have to make sure that the risks are managed. We do it in a standardized way.

Schulhofer-Wohl: The other thing that a central counterparty can do is net down the positions. So, imagine that we have a chain of trades. Suppose that you buy a Treasury security from me, and then on the same day, you turn around and sell it to the person down the street. If there's no central counterparty involved, then when we settle that trade tomorrow, what has to happen? First, you have to come to me with money, and I've got to give you the Treasury. Then, your counterparty has to come to you with money, and you've got to give them the Treasury.

Schulhofer-Wohl: You can see that if one trade messes up, the other will also fail. So, if I don't show up with the Treasury, you won't be able to deliver it to your counterparty. If your counterparty doesn't bring you the money, you won't be able to pay me. In financial markets, these chains get a whole lot longer than two people. A central counterparty can see all of those trades and can say, "Wait a second, there's a chain. Bob is buying from David, and David is buying from Sam."

Schulhofer-Wohl: “Really, David has no position here at the end of the day. Really, we just have to get the Treasury from Sam to Bob.” And so, they can simplify it by netting that down and canceling out those offsetting positions that you have. This has a lot of benefits. It can reduce the potential for these chains of defaults. Michael Fleming and Frank Keane at the New York Fed have done research on how much this could have helped in certain scenarios. It also means that people don't have to use their balance sheet on trades that cancel out.

Schulhofer-Wohl: So, instead of you having to think about, “I've got these two positions with Sam and with Bob,” now you've just got nothing, because it netted down at the central counterparty. To some degree, that netting can also happen with accounting rules, even without a CCP. So, the difference is not night and day, but it does make a difference in netting these exposures down. And then, also, if someone defaults on a trade, it really simplifies the unwind of that situation. So, if there were no central counterparty and there's a default [and] now everyone's trying to sell their collateral, you could get a fire sale.

Schulhofer-Wohl: A central counterparty can see the whole thing. They cancel out a lot of the offsetting pieces, and then they're the only person or the only firm that's got to liquidate collateral, and they can think about how to do that best to not cause trouble in the market. So, there's a lot of benefits to having this well-run organization or, potentially, organization. So, there are now firms that are talking about, with the growth of central clearing, the Treasury market coming in and being additional central counterparties, in addition to FICC.

Schulhofer-Wohl: You could conceivably have more than one, but having these professional clearinghouses manage the risk, net down the risks, and so on. The other thing that central clearing does is increase visibility into the market, because, obviously, in order to guarantee a trade, they've got to have data on the trade. They have to know it exists. Then, that data can be reported to the market. It can be reported to the official sector. That helps everyone with understanding the market dynamics and making better decisions.

Schulhofer-Wohl: In the repo market in particular right now, there is a real challenge with visibility. So, researchers think that this non-centrally cleared bilateral market for repo is potentially about $2 trillion in size, but the reality is that it's hard to know at the moment, because we don't have uniform data collection on those deals. So, the Office of Financial Research has issued a rule requiring data collection on them that's just now beginning to come into force. So, they just launched the pilot period for firms to learn how to make the data sufficient. Then, we'll have the data. But you can have these giant market segments that, if they're not centrally cleared, there's not complete visibility into them, and that's a challenge of its own. And so, central clearing helps with the visibility as well.

Beckworth: Well, there are a lot of benefits to CCP or central clearing, and it's a good point that you made that this is not just about crises. It's about standard operating procedures, being more efficient, clean, saving costs, [and] also preventing crises— not dealing with crises, [but] preventing them. So, there's a lot there that you mentioned. I just want to go back on a few things. First, go over the central clearing party right now, FICC. What is FICC, and how did it get this role out of the gate?

The History and Role of FICC

Schulhofer-Wohl: So, it's developed over many years. I'm not a historian of FICC, so I'll try to stay super high level so that I don't get anything wrong here. FICC is part of a complex called the Depository Trust and Clearing Corporation, DTCC, that provides a lot of the infrastructure for the settlement of securities trades of all kinds. So, another piece of DTCC is what I mentioned before, the National Securities Clearing Corporation, that serves this role of guaranteeing settlement risk on an equity trade.

Schulhofer-Wohl: FICC does this for Treasury securities and mortgage-backed securities. They offer the service of guaranteeing the trade. As I mentioned, there are large segments where, currently, the trades are not centrally cleared. The SEC has adopted a rule that will require many more trades to go through FICC. FICC was set up by the financial industry to serve this purpose. As I understand it, a lot of it goes back to the paperwork crisis on Wall Street in the '60s, where, at that time, stock trades, bond trades, and so on were settled, actually, by carrying physical stock certificates somewhere, and the back offices just got overflowing with paperwork, and things got worse and worse, and people realized that you needed a central tool to manage this and make it more efficient.

Schulhofer-Wohl: And initially, in some market segments, there were multiple clearing corporations around the country. This, of course, was the era where you had stock exchanges in different cities, the Philadelphia Exchange, the Chicago Exchange, and so on. Then, as the national markets have unified, the same forces that were leading to unified markets also drove the centralization of all of that activity at DTCC. For example, these days, most stock is nominally owned through the Depository Trust Corporation, and it's just kept on record there. And when you're buying or selling stock, what's actually happening is DTC is making a note in its records that a different brokerage is the beneficial owner of a certain share, and that your brokerage is updating its records that you're the beneficial owner, but the actual share is not going anywhere. It's just immobilized, which is a much more efficient system than moving pieces of paper around in back offices.

Beckworth: Well, Sam, do you think that that's the best setup, having one corporation do everything? Is there some kind of natural monopoly here, economies of scale, or do you foresee competitors coming in at some point in the future?

Schulhofer-Wohl: So, certainly, in the Treasury market, a number of firms have talked about coming in and offering Treasury clearing services, and as with all questions of competition and monopoly, you can see benefits and costs to competition and monopoly. On the one hand, there are economies of scale, because you can net trades against each other if they're at the same central counterparty. It's harder to do that when they're not at the same central counterparty.

Schulhofer-Wohl: On the other hand, competition can also foster innovation and creativity and developing new approaches that might be better. And so, there are trade-offs, and they're the standard ones that you face in any market that's got very large economies of scale. So, I think that the SEC's approach to this in their rulemaking was to write it in a way that allows for the possibility that there could be multiple central counterparties, and we'll see how the market evolves given that they're allowing that.

Beckworth: Well, it seems like FICC has a first-mover advantage here, so I won't wait with bated breath for the competitors to rise. Let me ask one related question to that. So, I'm very sympathetic to this. We've talked about this on the show before, so, I like where this is going. However, there has been a criticism raised about increased use of central clearing, and that is, you're concentrating all of the risk in one node in the financial system. So, what if FICC itself collapsed? And I think there's good responses, but I want to hear what you have to say to that.

Schulhofer-Wohl: So, central clearing eliminates a lot of risks and then centralizes the risks that remain at one very critical node. And so, from my perspective, it's obviously crucial that that node— or potentially a few nodes if you have competing clearinghouses— but that that node or those nodes are well-run, well-risk-managed, well-regulated, well-supervised. No one is proposing that we centralize all of the risk at a badly-run, badly-regulated central counterparty. That's not what's on the table here.

Schulhofer-Wohl: I take the point that that node is critical. I think that it's already a really critical node, and FICC is a designated systemically important financial market utility by the FSOC, recognizing that. It will just remain imperative for both the folks at FICC and their regulators to continue to take that risk seriously. But we do have central counterparties in many markets, both in the U.S. and around the world. We know a lot about how to run them well, and I think that that's a challenge that people are well-prepared to take on. It's valuable for people to point it out and say, "Hey, don't forget this."

Beckworth: So, this is not our first rodeo with central clearings. And so, we know how to manage them, approach them. At the end of the day, the Federal Reserve is also there too. If push comes to shove, I'm sure it could step in. Alright, let me bring up a few other suggestions that have come up for reforming the Treasury market, and there's one in particular that I want to highlight, because I think it complements or goes hand-in-hand with central clearing.

Beckworth: But there's been a number of other proposals, such as tweaking the supplemental leverage ratio, doing more automation, broadened access to the standing repo facility. But one that, I think, again, complements what you're talking about is all-to-all trades, another proposal that's brought up to enhance the resilience of the Treasury market. Maybe walk us through how that would go hand-in-hand with broader central clearing.

All-to-all Trades as a Path to Reforming the Treasury Market

Schulhofer-Wohl: So, right now, if you or I wanted to go buy a Treasury security, we wouldn't go on a stock exchange and potentially buy it from whoever happens to want to sell a Treasury security that day. In the way that we would, if you wanted to buy a share of stock, you would submit your order through your broker, it would go on a stock exchange, it'd be matched with whoever is potentially willing to sell. But the Treasury market, by and large, doesn't work like that.

Schulhofer-Wohl: The structure is instead that dealers trade mainly with each other with some high-frequency, largely algorithmic trading firms called principal trading firms. Then, ordinary customers trade only with a dealer where they've got an account. And so, if you or I wanted to buy a Treasury security, we'd go to a dealer where we have an account and say, "I want to buy this." They might have it in the inventory and sell it to us, or they might go out and source it. But either way, we're not putting that order onto the market for everyone to see.

Schulhofer-Wohl: Similarly, if we had one and wanted to sell, we'd go to our dealer and say, "I want to sell this," and we wouldn't be putting it on the market for everyone to see. So, the challenge here is that if lots and lots of people are trying to buy and sell at the same time, that the pipes through the dealers might not be able to handle all of the trades that people want to have happen. And there were reports that this was one of the many challenges that existed in March of 2020, given that there was this huge repricing, a huge change in the amount of risk in the world, and naturally, people wanted to change their positions.

Schulhofer-Wohl: Can you get all of that flow through the pipes that we've got? So, all-to-all trading is what you have in the stock market, what you're used to, which is that people want to buy or sell. You stick that order on an exchange, everyone else can see it, and whoever wants to match with it can match with it. In principle, that sounds like a larger pipe and one that could be helpful. Now, what's the connection to central clearing? It goes back to the settlement risk, because you're not going to do cash on the barrelhead, you're going to settle tomorrow.

Schulhofer-Wohl: You've got to trust that whoever you trade with will show up tomorrow. Now, if you're trading with your dealer who you have a longstanding relationship with— if it's a famous firm with a big name on Wall Street, you're going to trust them, and they've known you as their customer for a long time, [so] they're going to trust you. But if you stuck that order on an exchange, and who knows who it is, and it's totally anonymous, why would you trust that that person will show up?

Schulhofer-Wohl: So, at that point, you need some institution that guarantees that the deal will be completed, and that's what a central counterparty can achieve. And so, historically, a lot of why we have central counterparties in many markets— and they really started largely in derivatives markets more than 100 years ago, long before they came to securities markets— is to solve this problem. So, you had the Chicago derivatives markets, and you imagine those trading floors with all of those crowds of folks shouting at each other.

Schulhofer-Wohl: And you had the folks coming from out of town with the grain, and the locals trading the grain, and how are they all going to trust each other and know that these deals would actually go through? In derivatives, the problem is actually way worse than securities. In securities, you're going to show up tomorrow, or in the US, until recently, it was two days later for stocks. That's not very long for that risk to be outstanding.

Schulhofer-Wohl: A futures contract can last months or years, and so, you've got to worry about that person performing on the contract over months or years. So, they knew that they had this problem of, how does everyone trust? And the solution was [to] put a central counterparty in the middle, and everyone else in the market just has to trust the CCP. So, having a CCP can really support central clearing. Now, I think that the flip side of it is, what's the demand for central clearing? So, if you ask, why don't we have a lot of central clearing in the Treasury market right now? Or why do we not have universal central clearing? Because we do have a lot, but certainly far less than universal.

Schulhofer-Wohl: One reason is that a lot of the trades are these relationships where people are not so worried about trusting the counterparty because it's a longstanding relationship. If you had growth in all-to-all trading, that would increase the demand for central clearing. And so, there was a debate over, if we want more central clearing, which comes first, more all-to-all trading, or more central clearing? Marta Chaffee and I wrote a blog that said, "Hey, look, if we had more all-to-all trading, that might be a path toward more central clearing without mandating it. Ultimately, the policy approach was different, which is that the SEC has adopted a rule that requires any central counterparty to require its members to clear most of their trades. That will effectively get everything into central clearing. But the approach was taken to do central clearing first, and that will be the infrastructure that supports more all-to-all. It is also true that, historically, the growth of all-to-all trading has been a motivating factor that drove growth of central clearing.

Beckworth: One last question and [then] we'll move on from central clearing, but what is the timing of this central clearing? We have the new SEC rule. You've mentioned a number of developments. What do you see happening? What's the timeline?

The Future Timeline for Central Clearing

Schulhofer-Wohl: So, the SEC adopted their rule in late 2023. For the most part, at the moment, only Treasury trades between two dealers are centrally cleared. When we have a trade between a dealer and a non-dealer, they're not. Then, the SEC took a couple of steps to change that. So, they required the vast majority of repos to be centrally cleared starting in mid-2026. Then, they required cash transactions on what are called inter-dealer broker platforms, which are electronic platforms where dealers and principal trading firms trade with each other. They required those transactions to be centrally cleared starting at the very end of 2025.

Schulhofer-Wohl: Then, in addition, they changed the rules on how dealers handle customers' money to make this all possible. That was where my research came in. Because right now, if you want to centrally clear a Treasury trade involving a customer— and the gist of what the SEC did is to require a trade between a dealer and a customer to be cleared, or between two customers— right now, you would have to post the dealer's money as collateral for that trade.

Schulhofer-Wohl: That's economically unattractive in many circumstances, and so it would be very hard to get it done. The SEC also changed their rules, and this will kick in a little sooner, to allow dealers to post the customer money as collateral for customer trades while still protecting the customer position, ensuring they're not at risk if, for example, the dealer goes bankrupt for some other reason. So, there's a phased approach of all of these pieces coming in, but by mid-2026, all of those pieces are supposed to take effect.

Beckworth: A brave new world in 2026. I look forward to it. So, Sam, we've been talking about the Treasury market, and as you know, as a Fed person, the Fed is an important part of the Treasury market. It bought up a lot of Treasuries, and it bought up a lot of Treasuries because it maintains a floor operating system, or what some might call ample or abundant, depending on where you are. And recently, your president, Lorie Logan, had a speech on this that was really interesting. She approached this from a perspective that I typically don't see. When we talk about the floor system on here, we talk about supply and demand graphs, the flat part of the demand curve for bank reserves, but she did a speech that she gave last year in November at the ECB that was titled, *Ample Reserves and the Friedman Rule.* Help us understand this. What was she trying to articulate in this speech?

*Ample Reserves and the Friedman Rule*

Schulhofer-Wohl: So, let me step back a little bit and just talk about what ample reserves are, and then come back to the Friedman rule. So, historically, before the global financial crisis, for many, many years, the Fed implemented monetary policy with what are called scarce reserves. The FOMC would choose a target for the fed funds rate, and then the Fed was supplying less reserves than banks wanted to hold. So, that put banks at the steep part of their demand curve for reserves, and it put a spread between money market rates and whatever interest would happen to be paid on reserves, which at that time was zero.

Schulhofer-Wohl: So, these were very large spreads, hundreds of basis points usually. Then, the Open Market Trading Desk at the New York Fed would frequently adjust the supply of reserves in response to changing demand from banks and changing what we call autonomous factors, like the balance in the Treasury's account, so that the supply met demand at the right point on that demand curve to hit the target for the fed funds rate. Then, in the GFC, the Fed did several rounds of large-scale asset purchases to support the economy and provide economic stimulus.

Schulhofer-Wohl: And when the Fed buys assets, that also creates reserves. And so, the Fed de facto moved way out of the demand curve for reserves to the flat part of the curve where you can't make small adjustments in reserves and have any effect on the fed funds rate. You just drive rates down to, more or less, whatever interest is being paid on reserves, which was the zero lower bound at that time. Once you're in that environment, the main influence on money market rates, the fed funds rate or other rates, is whatever interest you're paying on reserves and not any fine-tuning of reserve levels.

Schulhofer-Wohl: Then, in 2019, the FOMC decided to remain in a system like that for the long run, a system with ample reserves, and I had the opportunity, actually, to be a visiting officer in monetary affairs at the Board, at that time, and lead the staff work that supported that decision, so it was great to get to see, upfront, how such a decision is made and really support the FOMC's work in thinking about not just, what do we want to be doing over a period of years in response to the GFC, but for the long run, how do we want to make this work? And the basic reasons that the FOMC gave in its minutes and statements at that time, were that this ample reserves regime is really effective.

Schulhofer-Wohl: So, post-crisis changes in regulations and in banks' risk management made reserve demand higher, more volatile, more difficult to predict. Additionally, the Treasury's balance got a lot larger and more volatile because of changes in how they manage their risk. And so, that made it much harder to envision a regime where you would make fine-tuning adjustments in the supply of reserves in order to hit a point on the demand curve. In addition, this ample reserves regime, where you just meet banks' demand at close to money market rates, has been effective in controlling rates and transmitting to financial conditions across a wide range of market environments.

Schulhofer-Wohl: So, it's working well right now, where we're away from the zero lower bound. It also works well in the transition to the ELB. So, in March 2020, the economy was deteriorating, the FOMC decided to do very large asset purchases, both to support smooth market functioning and then, eventually, to provide economic stimulus. That increased reserves, and there was just a very smooth transition into that environment. You didn't have to worry about, okay, we're changing from controlling rates by fine-tuning reserves to some other mode, and how do we make that change?

Schulhofer-Wohl: Because it's just all one approach. The FOMC stated in 2019 that this ample reserves approach is really effective. And then, President Logan's speech really took a different angle on that, but one that leads to the same conclusion. She was speaking in the context of a really international conversation about, how should central banks, in general, choose to implement monetary policy? So, lots of central banks around the world have been facing this decision as they come out of, both the response to the GFC, and then the response to the pandemic, and many central banks expanded their balance sheets and then have to think about, what do we want our balance sheet to look like in the long run?

Schulhofer-Wohl: So, she was speaking at an ECB conference on the issue and thinking about the broad theory and public policy rationales of this. And she turned to a broader public policy argument for why ample reserves is a desirable regime. It goes back to the Friedman rule. So, the classic Friedman rule that you learn in grad school, it says, well, look, money, like currency, doesn't pay interest. So, in an environment where the nominal market interest rate is positive, that's effectively a tax on holding money, because if you hold money, you don't get interest. If you held any other instrument, you'd get interest.

Schulhofer-Wohl: That's inefficient, because people need money to make payments and do transactions. And so, you're taxing something that can be supplied at very low cost. Why would you want to put that kind of drag on the economy? So, Friedman said that it's optimal to have a zero nominal interest rate so that money market rates are the same as the interest rate on money, which he's assuming is stuck at zero, and so that there's no implicit tax on having money. Well, if the real interest rate is positive, then a zero nominal interest rate means that you have to be running deflation.

Schulhofer-Wohl: And so, the shorthand for Friedman rule is often that it's optimal to have deflation. But it's all coming from this idea that you don't want to tax the supply of money when it's very easy to supply more money if people need it to efficiently do business. And that's an idea about the role of central banks that's existed for a long time. If you look at the original purposes of the Federal Reserve System, one is to supply an elastic currency. So, it considers the idea that currency is really important for the economy to work, and you don't want to restrict it unduly. So, in her speech, President Logan pointed out that, once you pay interest on reserves, the way you should think about what it means to not have an implicit tax on liquidity is different.

Schulhofer-Wohl: So, in our economy, there's currency, which certainly doesn't pay interest, but for the workings of the financial system, what's much more important are bank reserves, which the Fed does now pay interest on, unlike when Friedman was thinking about that. And so, in that environment, to not have an implicit tax on liquidity, you just want the interest rate on reserves to be close to money market interest rates. If you do that, then you're not either paying people to hold more reserves than other instruments, nor implicitly taxing them on their choice to hold reserves.

Schulhofer-Wohl: And so, an environment with ample reserves, where banks' demand for reserves is amply met with money market rates close to interest on reserves, delivers that efficient outcome. That was the argument in President Logan's speech. It, of course, gets you to the same place. If you're going to amply meet banks' demand for reserves at close to money market rates, you're going to be on the flat part of their demand curve or a relatively flat part of their demand curve. So, you're going to be in this environment where the primary influence on money market rates is the administered interest rate on reserves and other administered rates, which is how the FOMC defined ample reserves.

Schulhofer-Wohl: So, it's another way of getting to the same place, but it's in terms of what's efficient for the financial system. And I'd say that it's also, really, just the classic argument for, why should central banks pay interest on reserves in the first place? If you're going to have banks holding reserves, it doesn't cost a central bank any large amount to create reserves. There are arguments— For example, Annette Vissing-Jørgensen has pointed out that if you create reserves, you have to buy assets to back them. There's a cost of removing those assets from the financial system, but the cost of creating reserves is certainly not enormous. And so, why would you want to massively penalize the holding of reserves by having hundreds of basis points spreads between reserves and market rates? And so, that's the classic argument for paying interest on reserves, but in President Logan's speech, it really leads you to an argument that ample reserves is an efficient approach.

Beckworth: Okay, so, [there is] a lot more I could go there with that. I have a lot of questions, but for the sake of time, I want to move on and talk about a paper that you wrote, as we alluded to earlier, on the banking turmoil of March 2023. The title of this paper was, *Deposit Convexity, Monetary Policy, and Financial Stability.* Tell us what you found there, because I think it's really interesting, and I think it also has implications for some of the developments we see going forward. 

*Deposit Convexity, Monetary Policy, and Financial Stability*

Schulhofer-Wohl: So, that was a really fun paper to write, and it was an example of the kind of integrated thinking that we're trying to foster at the Dallas Fed. So, I wrote that paper together with Emily Greenwald, who is the head of our bank supervision department, and Josh Younger, who's a markets expert at the New York Fed. We, in that paper, make one simple point, but with one really big piece of jargon and then a few big implications. So, the big piece of jargon is convexity. Convexity is when the duration of a financial asset or the value of a portfolio changes non-linearly with interest rates.

Schulhofer-Wohl: For example, mortgages, from the perspective of a lender, have negative convexity, because if interest rates go up, the borrowers are less likely to refinance. That means that the duration of those mortgages increases. Then, when it increases, the lender is more exposed to duration risk, and any further increase in interest rates would make the value of those mortgages drop even more. So, that's an example of negative convexity. Our simple point is that bank deposits are hugely convex from the perspective of the bank, and that that hasn't been appreciated in the industry.

Schulhofer-Wohl: So, why does this matter? The duration of deposits is central to how banking operates. If you have a checking account or a savings account, a demand deposit, in principle, you're allowed to take the money out anytime, but we all know that people don't, by and large, do that. They leave those deposits for a long time, and the fact that deposits tend to stick around is what enables banks to make longer-term loans. It's just central to banking, knowing that, on average, deposits will stick around or that you can use them to finance long-term loans. But it means, as a banker, that you've got to get right, exactly, how long will those deposits stick around, and what is the risk that you're exposed to in terms of, not necessarily the money leaving, but that you'll have to pay a higher interest rate to hang on to it, because the market hasn't moved.

Schulhofer-Wohl: Standard models that are used in the banking industry assume what we call a constant beta. So, they assume that the interest rate a bank has got to pay on its deposits moves proportionately with market rates. If market rates go up 100 basis points and the beta is 0.5, then the assumption is that the bank has to pay 50 basis points more, for example. And they assume that it's just proportionate no matter how far or how fast the market rates rise. And so, that proportionality might make you think that there's no non-linearity or no convexity, but that's really wrong for two reasons.

Schulhofer-Wohl: One is just bond math. Any fixed coupon bond, anything that pays a fixed interest rate, shortens in duration when rates go up, because you're discounting the future of more. That's one source of convexity. Then, the other is that, empirically, the interest rates that banks pay on deposits are convex. So, as market rates rise, deposit costs rise non-linearly for banks. They don't actually have a constant beta. And so, the combination of those two means that their deposits, in fact, expose them to this huge convexity risk, which is, as rates go up, deposit duration is getting shorter and shorter, which makes them less and less able to fund long-term investments.

Schulhofer-Wohl: So, why does that matter? One, is it amplifies monetary policy transmission, because as rates rise more and more, banks' lending capacity is reduced and so that tightens the bank lending channel. So, it's an amplifier that we argue hasn't been fully appreciated. Then, the other is that it increases financial fragility. So, if banks are exposed to that risk and don't fully appreciate that they're exposed to that risk, then they won't manage the risk that they face right, and if rates rise rapidly, which, of course, we had a recent episode of over the past few years, [then] that's going to pose a challenge to them that they weren't prepared for. And so, our second point in the paper is that it’s really important for banks to be prepared for and manage this risk.

Beckworth: So, does the banking turmoil of 2023 fit this model?

Schulhofer-Wohl: I think that it's certainly an example of the convexity in deposit costs. You can see that, as rates rose, the rates banks had to pay on deposits went up faster and faster. There were banks that had a number of challenges. This was one of them. I'm not going to claim that this was the only challenge that banks faced at the time. A lot of vulnerabilities were exposed, but seeing what happened in that episode certainly is what motivated us to think about this issue of, how do deposit rates change as rates rise, and they go up faster and faster, and what does that mean for banks?

Beckworth: Yes, they rise disproportionately faster than one would expect. So, this, to me, underscores the importance of how a bank's depositor base is its franchise, and why it's so important to manage it, treat it well. I've had a guest on the show recently, Steven Kelly, and we were talking about the increased push for using the discount window as a way to park collateral there to meet your liquidity requirements. He's all for that, but he made this point. That's great from a systematic perspective, but it wasn't going to solve, say, for example, SVB's problems.

Beckworth: Its franchise was being undermined, depositors were leaving, and so it's important that you really have a handle on your bank franchise, the value, which is your depositor base. And if you are completely funded by the Fed, it raises questions about your future. So, understanding this risk, which you're saying hasn't been appreciated as much as it should have been so far, may prevent future crises like this. 

Beckworth: Sam, in the time we have left, I want to move towards some other work you've done, which, to me, is also very relevant to monetary policy, just like the previous paper we talked about. You've done a lot of work on labor mobility in the United States. You have several papers. For example, you have a 2017 International Economic Review paper, *Understanding the Long-Run Decline in Interstate Migration.* You have several others there, but tell us why this is important to the implementation of monetary policy.

The Importance of Labor Migration for Monetary Policy

Schulhofer-Wohl: So, a little bit of history— After the global financial crisis, as I'm sure you recall, unemployment was very high, but, also, job openings were pretty high. And so, unemployment was unexpectedly high given how many job openings [there] were. So, the Beveridge curve, which relates vacancies and unemployment, had shifted out, and there was a lot of debate about why that was, why these unemployed workers weren't matching to these vacancies. One hypothesis you heard a lot in the media and elsewhere was that many people were underwater on their houses because of the housing crisis.

Schulhofer-Wohl: And so, they— it was hypothesized— couldn't sell, and move to wherever they were going to get the new job, whether it was a new job that was just one that they weren't well suited for, or in some part of the country that was doing well. For example, there was the fracking boom going on at that time, and so there were jobs in the oil and gas industry in certain parts of the country. So, the hypothesis was that there are all of these workers, they'd love to have jobs, they can't sell their house, they can't move. The Census Bureau had also published data showing that there was a very sudden drop in the internal interstate migration rate in 2006 and 2007, which lent some credence to this idea that people weren't moving or couldn't move.

Schulhofer-Wohl: And it was important for monetary policy to understand whether that hypothesis was right, because if it was, then monetary stimulus wasn't going to be able to bring down the unemployment rate much, because if you stimulate the economy and more jobs are created, that's nice, but people still can't move and can't take them. We needed to understand whether that hypothesis was right or not. And so, I did a bunch of research on this, largely with Greg Kaplan, who overlapped with me briefly at the Minneapolis Fed and then went on to a very distinguished career in academia.

Schulhofer-Wohl: We found two key things. One is, we found that that big reported drop in internal migration in 2006 and 2007 was almost entirely a data glitch. It wasn't an actual drop in migration, and we showed where that data glitch came from and how you could correct the data to measure it correctly. When you made that correction, you saw no sudden drop there. So, just at face value, nothing was going on. What you did see, though, was a very smooth long run trend decrease in migration from the early 1990s through about 2011. So, by one measure, the migration rate in the US fell about in half, interstate migration, over those two decades.

Schulhofer-Wohl: Since then, it's been largely flat. And so, in some other work, we raised the question of, why did it drop? And the question was, basically, did it drop because the cost of moving went up or because the benefit of moving went down? If the cost of moving went up, this could be very bad for the economy, because it could mean that there are people who are not moving to places where they'd be better off, either for their work or just because they like living there, and if we could get that cost of moving down, that would be a good thing. On the other hand, if they're not moving because there's no benefit to doing it, then there's nothing to worry about here.

Schulhofer-Wohl: So, we wanted to understand, was it cost or benefit? And what we found in our research was that that trend decline in migration was mostly because the benefits of migration had fallen, not the costs. Labor markets had gotten more similar around the country. People also had better information about the amenities and what it was like to live in different parts of the country. And so, they were less likely to make a mistake— move somewhere, not like it, and go back. That's actually something you see remarkably a lot of in the data— that people move to a place, they stay there [for] a year or two, and then they go back where they came from.

Schulhofer-Wohl: But over time, a lot of the drop in migration was people making fewer of those return moves, and if you think what's going on there is that they discovered something that they could have known before moving that they didn't like, then they're better off if they don't have to move there, have that bad experience, and move back. So, our research found that a lot of this trend decrease in migration over those couple of decades was because the benefits were smaller, rather than the costs going up.

Beckworth: So, how does this relate to people who argue that the reason folks don't move to areas where maybe they would have better opportunities is the cost of housing? There is this understanding— I think it's widely shared— that we don't build enough homes and, therefore, housing prices are high. How does your research relate to that?

Schulhofer-Wohl: It's important here to distinguish net and gross migration. So my research was essentially about gross migration, which is people [in] moving opposite directions at the same time. That's important to the economy because it lets people be in the places that they're well-matched to, either for labor market reasons or because they just like living there. Then, net migration is when, in the aggregate, people are moving from one part of the country to another. That's also important for the economy, because it lets workers get to the places that are growing fastest, that are highest [in] productivity.

Schulhofer-Wohl: So, both of these matter. Historically, gross migration is an order of magnitude larger than net migration. There's just tons of people moving in opposite directions at the same time, but they are both really important for the economy. Conceptually, housing supply shouldn't matter much for gross flows, because you can just imagine two people who need to switch places could trade houses and that would be that. Of course, they don't actually trade houses, but you can imagine it. So, housing supply is more about this issue of net migration, and are the areas— the parts of the country that are most productive and growing the fastest, able to bring in workers?

Schulhofer-Wohl: And are workers able to come into those places and have that opportunity? Or are they unable to do it because there's not enough housing? There's a bunch of research, I think, that is particularly associated with folks like Enrico Moretti, Chang-Tai Hsieh, Peter Ganong, Danny Shoag that's found that housing supply constraints, particularly in coastal parts of the country, have limited net migration into those places. In turn, that can limit aggregate economic growth. So, that's where housing supply comes in, is in the flexibility of the economy to bring in workers to a place where productivity is higher or growing faster.

Schulhofer-Wohl: And coming to the Dallas Fed, for me, it's just been fascinating, as an economist, to see the really rapidly growing Texas 11th district economy and the rapid inflows of people from other parts of the country. It's just a key thing that you see every day on the ground in the district. And of course, one factor in that, among many that are driving it, could be the relative ease of expanding housing supply in Texas. Though, I would acknowledge that, even in Texas, there can be constraints in particular places in the short run, and there can be concerns about affordability. But when you compare it to other parts of the country, it's been faster housing supply growth. The Texas economy, in particular, is certainly not my area of research, but the Dallas Fed does a ton of fascinating research on this, and I really encourage people to visit our website and take a look at that if they're interested in learning more.

Beckworth: Okay, now, just circling back, again, to the importance of this for monetary policy. So, the way that I often approach this is thinking through the lens of optimal currency areas. So, you have a one-size-fits-all monetary policy for the United States. The Fed tightens, all rates go up, and maybe that's appropriate for, say, the state of Texas, which is booming, but it's really hammering Michigan, which might be in a structural recession.

Beckworth: And why labor migration might help is that people might move out of the area hit hard to the place where they would find work. So, it's kind of an offset. If you have a one-size-fits-all monetary policy, you need either perfectly synchronized business cycles across all states, or you have these buffers, these shock absorbers— labor mobility, fiscal transfers, or price flexibility. So, is that the way that you approach it, too? You think of this as an important shock absorber or a way to handle differences across the States? 

Schulhofer-Wohl: I think that's exactly right. Net migration is an important shock absorber in an integrated economy, like the one we've got. We wouldn't be able to have an integrated US economy, with all of the enormous benefits that flow from that, if people were not able to move around as the economy evolves in different ways in different parts of the country.

Beckworth: Okay, well, with that, our time is up. Our guest today has been Sam Schulhofer-Wohl. Sam, thank you so much for coming on the program.

Schulhofer-Wohl: David, thanks so much for having me. I really appreciate the opportunity.