There are a bunch of historical analogies that people like to reach for in order to describe some of the economic trends we're seeing today. There's obviously the period of high inflation in the 1970s and early 1980s, or the disruptions caused by the Spanish Flu pandemic around 1918. But there's also a single year -- 1953 -- which not only contains some eerie similarities to today's economic environment, but also ended up having far-reaching consequences that reverberate all the way to 2022. On this episode, Josh Younger, JPMorgan's global head of asset and liability management research and strategy, tells the origin story of the decisions made in 1953 that helped create the vast repurchase or repo market. At a time when there are plenty of concerns over the stability of the market for US bonds, we go back in time to explore the reasons why repo exist at all. This transcript has been lightly edited for clarity.
Key insights from the pod:
1953 as a seminal year in finance — 03:34
Who sets interest rates, the Fed or markets? — 4:36
A growing deficit and who’s going to buy all the bonds? — 5:40
The inflationary backdrop of the 1950s — 11:00
The Fed-Treasury Accord and freeing the bond market— 13:25
Bringing back the long bond in 1953 — 18:46
Parallels with the Bank of England — 22:06
The role of the dealers and expanding repo— 24:36
Is there anything the Fed can’t do? — 28:15
Repo vs. the call loan market — 31:07
Repo becomes entrenched in the financial system — 33:32
Why does the repo market keep seizing up? — 36:36
Should we worry about bond market volatility? — 39:16
---
Tracy: (00:10)
Hello and welcome to another episode of the Odd Lots podcast. I'm Tracy Alloway.
Joe: (00:14)
And I'm Joe Weisenthal.
Tracy: (00:15)
Joe, I feel like in our current period of high inflation bond market volatility, tension between central banks and fiscal stimulus, all of that, I feel like people tend to have their favorite historical analogies that they reach for to try to explain what we're experiencing now.
Joe: (00:35)
Everyone, whatever it is, always has ‘this is just like this’ and I get it, kind of, you know, naturally this is what we do. But I also think, you know, there's that danger of knowing too much history, which is if you know your history, you think it's repeating.
Tracy: (00:49)
That's right. I mean, the one that everyone seems to reach for is the 1970s, right? Like, ‘oh, this is just like the 1970s. We need a Paul Volcker to come and really tamp down inflation.’ And then the other one that I'm kind of partial to this one the 1918 Spanish Flu.
Joe: (01:05)
Who can forget?
Tracy: (01:05)
The period after that And World War I ended up with high inflation, and then it flipped into deflation.
Joe: (01:12)
You know, the other one that I think about too is the euro area crisis and the tensions, you know, what they had to do sort of plumbing-wise, what Mario Draghi had to do in terms of sort of recreating the European sovereign bond structure on the fly, to improve the transmission of monetary policy, I think some of these issues are also coming up. We really see it in the UK specifically.
Tracy: (01:34)
Yeah, that's a good one. I hadn't considered that, but today on that note, we are going to be speaking about a historic parallel that hardly ever gets mentioned. It is the year -- one specific year – 1953.
Joe: (01:46)
I don't know anything that happened in 1953. I would never, you could have asked me any year in the 1900s and I would not have…
Tracy: (01:54)
Yeah, same.
Joe: (01:55)
So what happened that year?
Tracy: (01:57)
All right, so it turns out 1953 was actually a seminal year for global finance for reasons that I will not get into right now, but specifically because it ended up with a policy decision that still has relevance today. So not only was the US specifically in the 1950s facing high inflation, tension between employment and prices, things like that, but we also had an outcome that has sort of like echoed across the decades ever since.
Joe: (02:27)
I want to learn more. I'm ready.
Tracy: (02:29)
All right. Well, we really do have the perfect guess to explain 1953 and its relevance to today to us. We are going to be speaking with Josh Younger. He is the Global Head of Asset and Liability Management Research and strategy at JP Morgan. He is also a repeat Odd Lots guest.
Joe: (02:45)
He's now getting up there like this is, he's not quite in the lead, but getting close.
Tracy: (02:49)
All right. One of our favorites. So Josh, welcome back to the show.
Joe: (02:52)
It's great to be back in in studio, which is very exciting.
Tracy: (02:55)
So this is just your hobby now? It's researching financial history?
Josh Younger: (03:00)
I need a new hobby, basically. Yeah, it's just for fun and there's so much you can do online, it seems weird to say, but the internet's a wonderful thing. Like, you can pull up basically anything. And the Federal Reserve Bank of St. Louis has been very kind to provide an enormous quantity of documents and data, so you don't have to get dusty in archives, you can just search it.
Joe: (03:17)
This is what I should be doing instead of just scrolling Twitter at night and playing on playing chess online.
Tracy: (03:22)
You should go on the New York Times Wayback Machine and see what bond market volatility was like in the early 1950s. Well, okay, on this note, Josh, why 1953, specifically?
Josh: (03:34)
1953 has a lot of relevance in general. So that's the year where the European countries start opening up their imports to the dollar zone. It's the London debt deal with the Germans after the war around German debt. But it's also one of the years where the Fed is really forced to make a really important decision.
So it's where a lot of their desire to get out of the bond market -- I'm sure we'll talk about what they did during the Second World War. It's a very controlled market. It's very heavily managed by the Fed on an active basis. They really pegged yields in the long end. And getting out of that is complicated. And ‘53 is when the market really tests them. So we didn't have the ‘Don't fight the Fed’ mantra quite yet at that point.
Joe: (04:11)
But this is a really interesting philosophical question right here, because there's always this question today or in any period, to what degree is the rate that we see on a 10-year, on a 30-year, on a two-year a policy rate versus a market rate. Before we get to ‘53 specifically, can you just talk a little bit more about what you're saying about how rates across the curve were essentially policy rates?
Josh: (04:36)
When we think about a 10-year rate, like you could do one of two things. You can roll Treasury bills every three months for 10 years, or you can buy a 10-year bond. So in a perfect world, those two things are connected through expectations. And then there's this question of should I involve a little premium on top of that, right? The term premium, which is I'm locking my money up for 10-years. I might be wrong in my expectations, and so I should probably be paid for that. At least a little excess over the expectations. And so they're connected. Most of the 10-year rate is an expectation. Those expectations are often wrong, but, you know, that's a separate issue. But there is a policy lever and on the other hand, there's the market lever, which is a term premium, the demand to actually buy the bond in the first place.
Tracy: (05:18)
So talk to us then about the state of the bond market. You know, coming out of World War II into the 1950s, you mentioned that the Fed had imposed yield curve control, basically to finance the war time deficit. They were trying to move off of that. Again, these are all sort of familiar themes to anyone in 2022, but why was that difficult for them?
Josh: (05:40)
Well, so let's get a sense of scale, right? So in 1941, there's about $40 billion worth of Treasuries outstanding. In 1945, there's about $240 billion worth of Treasuries outstanding. So that's about double the prewar GDP. If you scale that to 2019, say, okay, Covid expenditures on the scale of World War II, that's about $40 trillion in net Treasury issuance. So we're talking about a lot of debt. And the question is who's going to buy it?
Joe: (06:08)
So what we're dealing with is kind of nothing. We got this.
Josh: (06:10)
In that context. And so that's where, I guess the analogy breaks down, not to start there for the episode, but like, it's a very difficult problem, which is who's going to buy the equivalent of $40 trillion today in debt? And one of those participants is the Fed. So what the Fed says is, further into the war effort, we're going to peg yields specifically the front end, the one year point and the long end. So every 10-year bond, for example, yields two and a half percent or less because I'm willing to buy it at two and a half percent. Every Treasury bill yields three eighths of a percent. That's roughly where the curve was at the beginning of the war. There was no like particular thought given to that in terms of like fair value pricing. It was just saying wherever it is in 1941 is where it's going to stay until conditions allow.
Joe: (06:53)
Real quickly, how much actual buying did they have to do? Versus the declaration of the price essentially taking care of most of it?
Josh: (07:00)
Yeah, not that much. I mean, they bought about $20 billion, the Fed did over the course of that period. Most of those were in bills, actually, because bills at three eigths are not very appealing. Ten-year bonds at two and a half are a lot more appealing. So what the market did was they extended, and actually the Fed was a net seller of long end bonds towards the end of the war because yields were dipping below their target. So it wasn't all buying, but they owned something like 70, 80% of the bill market by the end of the war. And they didn't really have a chance to get out so easy because when you get out of that situation, like you can't just turn it off. And so there was this outstanding question, which is, how are we going to get out of the market and not completely blow up bond markets in the first instance? How do we do this in the right way?
Tracy: (07:42)
Well, so how, how did they try to undertake that transition? I mean, nowadays you think about the Fed trying to wind down its balance sheet, embark on quantitative tightening. There's a lot of communication that comes ahead of that. Was it a similar thing trying to broadcast to the market, ‘This is how we're going to do it. Please don't freak out’ ?
Josh: (08:01)
Yeah, it was really fraught. So Ken Galbraith has done a ton of work on this, and they really did it in fits and starts and very slowly. So no one ever really imagined they'd do it cold turkey. That was never really considered. So what they said is, ‘Well, let's start with bills.’ And then the Treasury said, ‘Well, we don't want unpeg bills. We like bills at three eighths. And there was a back and forth over a couple of years, and by, I think it was ‘47, essentially the committee went to the Treasury and said, ‘We think this makes sense. We'll figure it out for you, but we're going to do this in two days.’ And so it was a kind of unilateral decision by the committee. There were still sweeteners for the Treasury, but the Fed basically told the Treasury we're doing this whether you like it or not and then the debate moves on to the certificates of indebtedness, which roughly the one year point.
So they said, ‘Well, can that be released?’ And by around the early fifties, the front end of the curve out to the one year point was free to float. So bill yields came up into the 1-2% rate, the one and a half percent range. So it was kind of a market rate, but the long end was still at two and a half. And they hadn't figured out how to facilitate that. And that was most of the market, even though the Fed owned most of the bill market. So that made it a little more straightforward. But at the long end, you know, that was most of the outstanding debt at the time.
Joe: (09:22)
So I'm just trying to understand a little bit more how much of the challenge of, for the Fed, of extricating itself to some extent from the bond market, was about the rates specifically and what that would do to the economy were rates to jump up versus whose balance sheet these assets would be on?
Josh: (09:40)
Yeah, it was both. So the Treasury just doesn't like the idea of paying more, which is reasonable. I mean, they have a deficit to fund, and by the late forties, early fifties, the deficit’s expanding again. So they pay down their debt pretty fast. And then by the early fifties, you've got the Korean War, you've got a tax bill that actually Truman vetos and is passed over his objection to cut taxes. And so there's a fiscal expansion going on. So that's the Treasury's incentive. Now, a lot of their documents suggest they were also sensitive to the Fed's concern around inflation. And inflation was under control for about a year after the war because the Office of Price Administration basically fixed the price of things. They tried to renew that organization to keep price fixing in place. Truman decides it's not a strong enough bill, he vetos it, but they can't come up with something that can pass the veto. So it just expires. So all prices are released at the same time. So shock therapy.
Joe: (10:32)
Why don’t they just do that now? I don't get it.
Tracy: (10:34)
Well, okay, on that note, I mean the inflation of the late 1940s, early 1950s, how much of that was sort of supply chain issues, you know, the transitioning of the US economy from a wartime footing to one of peace, and then I guess they went back to war relatively quickly given the Korea situation, versus actual debt monetization because you know, the Fed was in the market buying back Treasuries?
Josh: (11:00)
Yeah, it was it's hard to make an attribution. I haven't seen a good attribution, but you know, the other thing is the Fed bought $20 billion worth of Treasuries during the war. Commercial banks bought $70 billion. So commercial banks are the real price fixers in the Treasury market during the Second World War. And when a commercial bank buys a Treasury bond, they create deposits on the other side of that. So they didn't let loans roll off and replace them with Treasuries. They had new assets. So the leverage in the banking system doubles in three years to 1941 and 1944. So this is new money because deposits are money, right? And some will talk about, you know, money's a spectrum. So deposits are very close to the real paper money side of that spectrum. And so there's a view, at least at the Treasury and the Fed, that because banks are so heavily supporting the market, they own 60% of marketable debt, that that expansion of their balance sheets, expansion of the money supply has generated a lot of inflation. So that's on the demand side, there's too much money. We can debate monetarism, I guess on a different episode, , but there's a big expansion of the money supply. And at the same time, you go from guns to butter and back, and that's disruptive. So that's where the 2020 analogies come in. You're really moving the economy from a very different production model in a year. And then you're trying to bring it back. And, and these things don't start up that quickly.
Joe: (12:15)
This is all so familiar. So, okay, so the Fed doesn't want to be fixing the price of all these assets. There's also this concern about too much commercial bank leverage -- valid or not. So who else is there, what's the next step here in terms of who enters the market?
Josh: (12:34)
Yeah, so the non-bank investors are really domestic. So something like less than 1% of the market for Treasuries at the end of the war. And into the late fifties is owned by international investors. So you have to find insurance companies and, you know, other, there's not a great deal of granularity in the old data. So it's basically like, it's a bank, it's an insurance company, it's the Fed, it's international, or it's somebody else. And so, you know, we don't have PIMCOs and Black Rocks in the world at the time. And corporations are very important. Corporations are very cash rich, they have a lot of money to invest and so to find a place to put it, but ultimately two and a half percent in a world where inflation’s at 10% is not particularly appealing.
Tracy: (13:11)
So what happens in the bond market? I can imagine that, you know, the combination of two and a half percent being not that appealing plus the Fed basically making it fairly clear that they would like to step back from the market, I imagine that's a recipe for some drama.
Josh: (13:25)
Yeah. What became the Acord, or the Fed-Treasury Accord, the Treasury-Fed Accord, I'm not sure what the order was supposed to put in, but that's in 1951. And that's the result of a lot of congressional pressure in the early fifties. So in 1950 the, the banking committee at the Senate convenes a hearing. They make recommendations, you know, ‘thanks for being helpful kind of thing.’ And the president gets involved and there's just a lot of concern that monetary policy is worsening a very disruptive inflationary environment. And by early ‘51, there's just enough pressure within the Treasury to get along that they figure it out. And one of the most important characters in this debate is Bill Martin. So Bill Martin used to run the New York Stock Exchange in the thirties. So he's comes from a dealer background, he comes from the street, he goes in and runs the Ex-Im Bank for Truman. And then he's brought into Treasury by Snyder, who's the secretary of the Treasury at the time, initially as an assistant secretary for International Affairs. But he quickly becomes kind of a fixer. So he’s kind of doing everything and he's tasked by Snyder with liaising with the Federal Reserve to figure out a way to get out the market that’s amenable to both parties. Like what's the best way to do this, because we kind of have to do.
Joe: (14:36)
So what did what did he do?
Josh: (14:38)
So he brokered an arrangement that came out as a very vague commitment. So they basically just say, ‘Look, we've reached a full accord.’ They don't say what that accord is other than to say we're going to stop monetizing the debt. So the question is, how are we actually going to do this? But they do put that up publicly. They make a public commitment to free the bond market. And a couple days later Martin is actually nominated to chair the Fed because McCabe steads down. So he switches sides, and he's confirmed pretty quickly and by early April he's the chair of the board of the Fed. And he's very committed to this inflation fighting thing. And in his oath of office statement, he says inflation is the greatest threat, including the enemies beyond our borders, or something to that effect, which, you know, in April, 1951, three days earlier the Rosenbergs have been convicted.
You don't get Joe Welch saying, ‘Have you no decency, sir?’ to McCarthy until ‘54. So this is the Red Scare, and he's saying inflation is worse than communism, or implicitly. And so, he's very much committed to the cause. But the commitment is out there. It's just the question of like, how are we going to execute this in practice? And that's where, you know, the changing of the guard, Treasury, the Eisenhower election, people have to get into the seat for the next administration to actually affect the change in policy because they ultimately need to agree. Snyder didn't like the idea of issuing above two and a half percent. So the Treasury kept flooding the market with short term debt. And that was partially for regular debt management purposes. They do some things around non-marketable debt to get bank debt, bank holdings down. But ultimately there's just a lot of institutional tension. At one point Truman takes Martin aside at some event and says, ‘you're a traitor, right? He gets very much in his face. So it’s a very fraught situation.
Tracy: (16:21)
I feel like I'm listening to a really good, like, campfire story.
Joe: (16:24)
This is good. I don't even want to ask any questions. I just want to hear the whole story.
Tracy: (16:28)
Just like ‘what happens next?’ I mean, on that note, you can imagine at some point the Treasury goes back to selling longer term debt, right? And how does the market take that?
Josh: (16:39)
So that's an Eisenhower administration thing. Part of Eisenhower's election is about inflation. Actually, one of the first television commercials that ever got sent to a general audience is about inflation. It’s Eisenhower is saying, ‘my grandmother doesn't like the energy of inflation. That's why I say it's time for a change.’ That’s a central plank of his presidential campaign.
And he nominates not only Humphrey who's committed to fighting inflation, but a bunch of ex-Fed people. So he brings in people in from the New York Fed and other places to be senior advisors, undersecretaries. And so, you know, the news media at the time speculates the Fed has like won this argument as to the relative benefits of low cost of debt service versus inflation. So inflation is the priority. So the changing of the administration changes the priorities. And they come in and they say, ‘We need to find non-bank investors. We need to find a way to sell long-term debt that doesn't inflate the size of the banking system and by extension the money supply and by extension the price pressures.’ And their solution is to reintroduce the bond. They hadn't issued long-term debt since ‘45. And they want to bring back the bond. And the question is how do you figure out the right way to do that? Because the market's not used to buying, I think it was supposed to mature in ‘83 or something like that. So it's like a 30 year bond and they haven't done a 30 year bond in a really long time. So who's going to buy it? How are you going to price it in a way that brings in interest from non-banks specifically.
Joe: (18:11)
I mean, this point is interesting because when politicians talk about fighting inflation today, they're talking about various fiscal policy levers. I mean, we need to cut spending here or we need to, you know, expand oil drilling or something like that. But it sounds like in the Eisenhower administration was like, they're trying to solve it via the financial system or via plumbing in some way.
Josh: (18:33)
Yeah, everything was basically, well, there was a fiscal lever as well, I should say. But that's always a little hard though -- to say, ‘Oh, higher taxes because inflation's higher’ and that doesn't feel great for anybody.
Tracy: (18:45)
Again, relevant to today.
Josh: (18:46)
Relevant, yeah. But the debt management strategy of the Treasury, when the Fed is buying debt at a fixed price, debt management is money management because they're in the market to buy at any price. They buy Treasuries. When the Fed buys Treasuries, they print new money with which to do so. And so those two things are connected and this becomes a technocratic problem. There is actually quite a bit of public debate about debt management, more so than you would expect given the current climate or at least focus on, you know, how big the 20 year sector is, and things like that. But you know, it's not like a political story all the time, but back then it would've been. But you know, I think the key here is there's an alignment of interests.
And so Humphrey comes to the market and says ‘we're going to bring back the bond. And by the way, we're pricing it cheap, so get involved.’ So we're going to price it about a quarter of a point cheap to where you would've otherwise expect it, given where like the victor, two and a halfs were trading, the victories were issued in ‘45. So everybody's interested, right? So they're saying we want to blow out reception. They get five times over subscription. So they get five and a half billion of orders for roughly a billion of paper. And they're feeling really good about it. The problem is, it turns out that a lot of that five and a half billion was speculative. So people were trying to buy cheap bonds and flip 'em for a half a point. And so they called them free riders back then, which I haven't heard. But it's kind of a fun analogy.
Tracy: (20:16)
Could I ask, who did they think they were going to flip them to?
Josh: (20:19)
Unclear. Somebody else? It’s just a question of, you know, if you think the bond is trading fundamentally cheap, then there's probably someone else at what you think the par value should be. And so their bet was, the expectations of the market were for a lower yield, they could buy it a slightly higher yield. They could flip it to someone with different expectations and as long as you got filled, you'd be happy, as long as you got your order in sufficient size to make that interesting. The Treasury also makes a mistake in filling orders on a proportional basis. So if you put in an outsized order, for example, as part of the fixed rate placement, then you got a big allocation. And the problem is if you're a spec account, that's not necessarily what you want it to happen. Um, so a lot of bonds get sold almost immediately.
Tracy: (21:03)
Yeah. This sounds familiar. This is like big bond funds padding their order books with corporate issues nowadays.
Josh: (21:09)
Yeah, it was actually so concerning that they delayed allocations which had never really been done, or at least for a long time, so the Fed could go through the orders and make sure that they all made sense. So, you know, so they were a little nervous from the start. But the bond eventually prints in, I believe it was late April, initially it's trading. But it starts to weaken pretty quickly. And in particular the market starts to become pretty dysfunctional. And you know, one of the fun things about the New York Times in the fifties is they published bid ask spreads for every Treasury issue on a daily basis.
Tracy: (21:37)
Oh man, imagine the poor reporter that has to make those phone calls.
Josh: (21:40)
It was after the sports section. So they had their priorities straight. Actually the whole business section was after the sports section.
Joe: (21:47)
Are there any papers that even have stocks anymore?
Josh: (21:49)
Probably not. But they had FX forward pricing in Europe, so three month forwards on sterling, they had on a roughly weekly basis, they had daily bond market pricing. So I mean, this was the only source. You had nowhere else to go.
Joe: (22:02)
Okay. So the bond starts trading weak, then what?
Josh: (22:06)
So they're faced with a choice. So inflation's still pretty high. This is going to sound familiar. So you have market functioning issues, right? So the solution to market functioning issues is to buy bonds. And there's two problems. The first is the Fed is committed to buy only bonds in the short end. That was part of the agreement when they came to the Accord, they said, ‘we're only going to intervene in the market in the short end.’ And that's specifically to provide reserves. So it's not about target prices for long term bonds. So we're going to let the market find the price. So they have to decide if they're going to buy long bonds, because that's where the pressure is. Bid as spreads are widening, volatility's picking up. And then they have to decide if they want to increase the size of the money supply at a time when inflation is running hot.
So you have you have a discrepancy in the policy goals, which is fighting inflation. You should be tightening. But to fix market functioning, you have to ease. And so they're faced with this really existential choice. And ultimately, the world kind of bails them out in the sense that the business cycle turns mid-year, right around when this is happening. So they buy a bunch of bills, they do a bunch of repo, which we'll talk about, I'm sure shortly. But they do a bunch of financing. They offer financing to dealers and they lower the reserve requirement on banks. So they grow the size of the banking system and they buy a bunch of Treasury bonds. So the second half of the year banks increase their holdings by $2 billion.
Tracy: (23:24)
Do they buy the long end though? Or just T-bills?
Josh: (23:26)
They bought the long end as well. So, banks were sort of interested in not the long end, long end but like intermediate type paper, you know, three to five years. And they also buy bills. And ultimately the question is can we take the excess paper out of the market, which doesn't have a home and dealers can't warehouse, and resolve the functioning issues and kind of come at the problem again another day?
Joe: (23:49)
Extremely BOE in terms of what's going on right now, of this dual tension of, like, we need to maintain financial market stability, but also we're in an anti-inflation stance. And so any perception of balance sheet expansion is seen as working at cross purposes.
Tracy: (24:05)
Totally. And also, I mean, just the idea of we need to expand the buyer base for US bonds as well. That's kind of familiar. There's been a lot of chat recently about who is going to buy bonds given interest rate volatility and you know, a backdrop of higher inflation and all of that. Okay, back to the story. So the market tests, the Accord, the Fed kind of backs down and then what does it do? Does it try to like go back to the period of the Accord? Or does it try out some new solutions to this problem?
Josh: (24:36)
Yeah, so they have two problems. One is just who’s going to buy the bonds. And the other is the dealers are clearly not capable of intermediating a market that large, right? So they’re clearly are not able to warehouse securities in enough size to really dampen volatility, which is what dealers are supposed to do, right? They're supposed to find a buyer and a seller and if they can't find each other the same day, they hold that thing in their inventory. A big problem the dealers had was financing. So for most of the prior 70 years, dealers have been funded most by what are called call loans. Call loans are kind of this intuitive concept, which is if I'm a dealer, I need a certain amount of money borrowed every day, I'll post collateral against it, but I'll probably borrow it from a bank and I'll just change the balance on a daily basis and I'll pledge whatever bonds I have as collateral to back that loan.
So that was the call loan market. The problem with the call loan market, it’s kind of expensive. And when bill yields and Treasury yields were below the call market rates, it meant that inventory was negative carry. Negative carry means it costs you money to hold inventory. You don't make interest income at least, to hold inventory. So it becomes very expensive to run a dealer when you have interest rate expenses on top of the salaries and infrastructure and things like that. And so this has been a problem in the forties as well. And that's when they brought back the repurchase facility.
The repurchase facility dates back to 1917 when they used it to support the first World War effort. And they were concerned that there wasn't a market for Treasuries back then. And so they used repurchase agreements, which are the buying and selling of a bond at different prices that kind of mimics a loan, but in two transactions. And so that allowed the Fed to do two things. One is it allowed them to lend money to non-banks, which is critical. And the second thing is it allowed them to do that below the discount rate in principle. So they could do it at a rate that was consistent with where bills were trading if they were below the policy rate.
Tracy: (26:28)
So this is kind of a legal workaround, right? Because I imagine the Fed isn't really supposed to be lending money directly to non-banks. I mean, the whole reason that banks have regulations and things like that is so that they can interact directly with the Fed and they have that sort of safety backstop. Is that right?
Josh: (26:44)
Yeah. Now Carter Glass of Glass-Steagall, was instrumental in the original Federal Reserve Act. And he said ‘this is not intended for non-banks, but in the Depression they find out that like sometimes you need to. So Section 13(3), which becomes very famous in 2008, that's the authorization that allows them to do all kinds of interventions. It specifically allows for the lending or extensions of credit to non-banks. So in 2008 that includes, you know, primary dealers, that includes a variety of real economy participants. And in 2020 they do the same thing. So, the main street lending facilities is in principle a 13(3) facility, the problem with 13(3) is it's only allowable under exigent circumstances.
So this is not a business as usual facility, right? This is, if it really comes to it, you can do pretty much whatever you want, but you need to be at least A) of the opinion that the world's about to collapse. And B) you have to be able to demonstrate that credit was not otherwise available.
So that's clearly not the case for dealers in the fifties. The workaround is, well this is a repo, it's a purchase and a sale. This is an open market operation. This isn't section 13(3) at all. This is section 14, which allows me to transact with primary dealers. And so yes, this has the economic features of a loan, but it is fundamentally a purchase and sale. They had to do a little bit of jimmying with it in the twenties to make it consistent with legal opinions. I think it's puttable as opposed to like specific maturities and they try to work around some legal interpretations. But fundamentally it's just a way to lend money to non-dealers in a format that gives the committee a lot more flexibility.
Joe: (28:15)
Just a brief sort of theoretical question, is there ever really a limit to what the Fed can do beyond the creativity of lawyers?
Josh: (28:23)
Well, that's anything basically.
Joe: (28:25)
But that's what I mean. Like in the end, you know, we look at these documents and there are debates about what they did.
Tracy: (28:30)
Right, it's like, don't lend to non-banks, but actually you totally can.
Joe: (28:35)
The constraint is creativity.
Josh: (28:37)
Yeah. Creativity and Congress. So Congress can give you very specific constraints and the courts can in principle stop you. I think it's tough with the Fed because the question is what's the cause of action? Like, who's going to sue them and say you shouldn't have lent this deal or money. So I'm not sure how much this has been tested in court. This is where my lack of a law degree is probably worth noting. But ultimately the interpretation of these rules is ultimately is an interpretation. Fed has a general counsel, they write opinions and there's an internal process, but they can do whatever they think to be ultimately legal.
Tracy: (29:08)
So the Fed's repo facility is created in 1917, but then in the 1950s as you just described, it presumably gets a big boost because they've settled on it as a way to solve this problem of ‘how do we sort of settle the debt market without sparking debt monetization and another bout of inflation’ ?
Josh: (29:28)
Yeah. So what's interesting here is the Fed doesn't do a ton of repo, but what they do is they demonstrate their willingness to use it. They have a big internal debate, they write a bunch of memos and they say, this is something we're committed to doing at the bill rate or higher. So basically they're saying, we are here for the market to provide repo to primary dealers at this specific administered rate. We're just going to say what it is. It's not an auction, we're just going to tell you where we'll lend to you. And so that's important as a backstop. So it's not so much that the Fed is financing dealers, it's providing a liquidity backstop at a specified rate that gives other market participants willingness to participate in the repo market. So most funding in repo by the end of the fifties comes from corporations.
And those corporations had bank deposits. Bank deposits were limited by Reg-Q, a Depression era regulation saying we don't want banks competing with each other for funding, that leads to bad outcomes. And so they don't like their bank deposit yields, they go the repo market, they get wholesale funding rates, they get much more attractive yields. And so the Fed gives the market confidence that they can participate in this sort of money-like, or deposit substitute. Actually the New York clearinghouse is really worried about this in the late fifties. So a consortium of banks say you're cannibalizing our funding, we don't want people doing repo, we want people keeping deposits with us.
Joe: (30:50)
So, sorry, walk through or make it a little more clear. So this is sort of what replaced the call loan market. What does the gap in terms of like financing cost between these two things and what did that open up in terms of, you know, how much balance sheet or how much capacity the dealer community had?
Josh: (31:07)
Yeah, it wasn't always a lot, but it was predictable. And the Fed could control it, and the call loan market was prone to, they called them ‘call loan panics.’ So like you'd call your New York bank and they wouldn't have money on hand, and so you'd call someone in the Midwest or some like not non-New York, other city bank or, you know, I don't think the world banks were really participating in this, but it was, it was just hokey in a lot of ways and somewhat unpredictable. And they ultimately ended up going to fed funds markets often to plug intraday liquidity gaps. And repo was better than that for that purpose. But the price could just be maintained by the Fed specifically, and that was key. So if the Fed can control the price, then they can make sure it's profitable to run a dealer and they can expand their balance sheets.
So there's some data on dealer balance sheets. I think turnover is a better measure. So turnover as a fraction of the overall market, you want to scale it to the whole market, between 1950, I guess ‘55 might be the earliest day we have, to the late sixties, it goes up by multiples. So the dealers are able to move, move the debt around much more easily. That's important because they're a distribution mechanism. If you want non-banks buying Treasuries, you have to find them. And dealers are the mechanism by which to get them the paper. And so that's when non-bank ownership starts going up a lot. So banks go from half the market to 40% to 30% by 2005/2006 to like three or 4% of the market. So it's a very successful policy. Dealers are able to do a lot more volume, a lot more turnover. Bid-ask spreads stay relatively tight until relatively recently. And by the 2008 financial crisis, US banks are not a huge fraction or even a dominant fraction of the of the Treasury market. They have a lot of non-bank participation.
Tracy: (32:52)
So this is the amazing thing because I think nowadays we're accustomed to thinking about the repo market as a source of potential instability and a big component of the shadow banking system. So you think back to 2008, the repo market was kind of ground zero for a lot of the problems that that occurred in mortgage-backed securities. And then in the decade since 2008, all we've heard from the Fed and the Financial Stability Board is ‘Oh, we need to get a handle on shadow banks. We need to reform the repo market. We need to make everything better.’ But as you are describing it, this was a direct policy decision made in the 1950s to solve a specific problem.
Josh: (33:32)
Yeah. It becomes something that's really entrenched. So the repo market is sort of the solution to their problem in lots of ways, but by the time the solution becomes very effective, now they're committed to the repo market. So, you know, a great example of that is in ‘82 when a bankruptcy court finds that the collateral associated with repo is subject to an automatic stay. What does that mean? It means if a dealer goes bankrupt, this court will seize their collateral and hold it until the bankruptcy is resolved. And that doesn't take a day or two, that takes a long time. And people had just sort of assumed that that wasn't the case. Because they're like, ‘Oh, this is a purchase and a sale. Like why would this be subject to a stay? I just have like a bond that I sold you and you're going to sell it back to me. We'll just do that.’
And the Fed panics basically in ‘82, they lobby Congress aggressively to get specific protections for repo and other similar instruments basically legislated. So there's a bill in ‘84 that's meant to reform the judicial nomination process that actually has within it a protection for repo that exempts it from the automatic stay, so give preferential treatment for repo specifically in bankruptcy. And that is necessary, Volcker argues directly to Bob Doll at the time, to stabilize the whole financial market, you need to do this. So it's just an example of where you sort of find a good solution. Now it's entrenched and the more it presents issues, the more you have to step in to provide other protections or specifically that market.
So it is a shadow banking system in the sense that it is treated like a money alternative, like a deposit alternative. And it's sort of wrapped with successive layers of protection. First there's the Fed liquidity backstop, then there's bankruptcy protection. You know, banks for example, have a different treatment under bankruptcy. Now repo has different treatment under bankruptcy. And by the time you get to 2008 and especially in 2020, the Fed is doing both sides of the repo market. So the Fed's doing the standing repo facility, right? And the reverse repo facility. And so now it's very much a policy lever in lots of ways, whether or not that's desirable is a separate question, but ultimately repo becomes the plumbing because it's so effective in, in facilitating the initial policy goals. The Martin fa.
Joe: (35:39)
So this is really interesting. And I guess just to sort of piggyback on Tracy's last question, I mean, we do seem to have these recurring bouts of instability today in this part of the market. The market that exists solved a certain problem at the time. What is the trade off? Is it that what we're still living with today? Because it seems like the sort of like old regime, while maybe it had issues with inflation etc., the sort of the less market-based regime, so to speak, at least it was more stable.
Josh: (36:12)
Yeah. I think that there's different phases of cognitive dissonance here. So, you know, if you're going to pull dealers in and say you're the critical intermediation mechanism for the Treasury market, and not only are you critical to intermediating it, but getting things out of banks and into non-banks, is like critical to financial stability and more important than the enemies beyond our shores. And that's what they're saying. Do to fight the Soviets we need repo. basically.
Tracy: (36:35)
Again, relevant to today
Josh: (36:36)
So the problem with that is you need a regulatory system that recognizes the relative importance you're placing on dealers. And that took a long time. So dealers are regulated by the SEC. It's an investor protection mandate. So basically if and when a dealer goes bankrupt, we want to have funds on hand to resolve all outstanding trades. Banks are a safety and soundness mandate, which is like, you can't go bankrupt, so let's make sure you don't because you're critical to the functioning of the economy. And so those two things don't entirely mesh.
And by 2006/2007, there's a series of changes to how dealers are regulated and all of a sudden they have a ton of leverage. You know, 40 times leverage is one that everyone always quotes, but like, turnover in the Treasury market skyrockets. Because not only are dealers allowed to take more leverage, but also Treasury, specifically Treasury repo, don't really count towards risk-based capital requirements. And so there's just a lot of capacity to intermediate Treasury bond trading and repo trading in 2008.
Everything gets sucked back into the banking system. So all these independent dealers, with the exception of the smaller ones, but like either you're either out of business or you're part of a bank holding company. So Lehman goes out of business, Bear [Stearn]’s absorbed by JP Morgan, you know, Merrill goes to Bank of America and all of a sudden they're all subject to that bank safety and soundness requirement, or at least within the context of the holding company as a whole. And so the market gets consolidated behind the bank regulatory permiter becomes like formally associated with banks.
That's fine as long as Treasuries don't consume a lot of, you know, what we often call in the industry like resources, meaning capital and liquidity. Treasuries are risk-free, or at least nearly risk free, depending on how it's hair cutyted and so forth. And so, if it doesn't add to your risk-weighted assets, which was the binding constraint on banks for a long time, you’ve got a lot of elasticity. You can grow and shrink a Treasury balance sheet very straightforwardly. Leverage constraints come later and they are inconsistent with that, or at least under some circumstances.
Tracy: (38:37)
Well, this is exactly what I wanted to ask about. So today there is this big question mark over the Treasury market about who is going to buy and also intermediate the bonds. So we have a lot of interest rate volatility. Dealer inventories are lower than they have been historically. What exactly is going on there? Like the repo market exists, it experiences spasms from time to time, but the Fed comes in and for the most part seems to fix them or at least set up new programs aimed at fixing them. Why is this still happening? Why do we have that concern?
Josh: (39:08)
So it's a little different every time which is unfortunate, but ultimately…
Joe: (39:10)
It's why we keep having episodes. Content.
Josh: (39:16)
But you're seeing banks buy more Treasuries now, right? So like there is a certain amount of increased monetization of the debt through the commercial banking system, not necessarily through the Federal Reserve. And so that's helping, or it was helping for a while. It's not so much helping anymore. But I think the issue is repo is a form of money substitute that has the implicit backing of the Fed, but is not really wrapped in the same kind of cloak as deposits. And so deposit funding is stable and sticky funding at low cost. Repo funding is fine until there's an issue. And then it's sort of more prone to instability than more traditional forms of bank funding.
And as you go further out the spectrum, you know, we're talking about Treasury repo, then you can talk about mortgage repo in 2007, we could have talked about non-agency mortgage repo. And then there's a question of the collateral credit quality as well as access to liquidity. And so repo is just not money in the sense that deposits are. And so if banks are being called upon to intermediate Treasuries as a core banking activity in effect, they're still funding it with a form of funding that's reserved for dealers and that's the investor protection world. So there's a little bit of cognitive dissonance there. But what the Fed is doing is they're backstopping both sides of the market. So repo's perceiving those kind of controls and there are other central banks where repo's the primary policy, right? So this is not, you know, particularly unusual, whether or not it's desirable is a separate question. And whether or not those facilities are effective is another one.
You’re talking about dealer inventories now, I think I was on to talk about this pretty recently. I remain relatively sanguine I guess, and, you know, prices are moving around, but the world is moving around. Like the world is as volatile as the Treasury market is volatile. And so that's not necessarily, I wouldn't say it's a good or bad thing, but it's certainly not unexpected. Dealer inventories being low, off the run trading not being particularly high as a fraction of total. Like, you don't see a lot of the monetization of Treasuries in the form of sales to source cash in the way that you saw in 2020. So could that happen? Sure. If dealer inventories were going to rise rapidly and we had the same kind of dash for cash dynamic, I think we'd run into similar problems. Ultimately those issues have not been fixed, at least on a fundamental basis. But, you know, the market is functioning even if it is illliquid. I might have said precisely that a few months ago, but I still believe it's the case.
Tracy: (41:39)
Oh, we've come full circle then. And I’ve got to say, the idea that the Treasury market is as volatile as the world itself. That's a good quote.
Joe: (41:46)
It’s a good line.
Tracy: (41:47)
Yeah. All right. Well, Josh that was amazing. I learned so much. I can't believe this is what you do, you know, for fun in your evenings, but absolutely fantastic, super educational and I think you're one of the few people who can draw a direct line between something that happened in 1953 and a lot of what we're experiencing today.
Josh: (42:07)
And I should, I should thank Lev Menand as well, who's been really helpful with this. And he does this for a living, so, you know.
Joe: (42:13)
We’ve got to get Lev on some point. I've watched some of his lectures on YouTube and stuff. Really interesting.
Josh: (42:18)
If you want to understand the banking system and where it comes from and like, why we have certain rules…
Joe: (42:22)
Okay, good. We'll make that happen.
Tracy: (42:24)
All right. Thanks so much, Josh!
Josh: (42:25)
Thank you very much.
Tracy: (42:39)
So Joe, that was fantastic. I actually feel like I kind of sat down and listened to a narrative story.
Joe: (42:45)
I love that. Although it confirms this like long view that I've had about all these questions about how many of the solutions to problems are about recreating the exact same thing under a new language. Or when it's like, it's still the same thing, you know? And so it always sort of drives me crazy when it's like, okay, well we want the Fed to backstop it, but we don't want it to increase the money supply measured this way. So put on someone else's balance sheet. But it's still like economically the same thing. All these conversations, they kind of drive you crazy just because it's like, I don’t know, just have the Fed buy it all. It solves the problem.
Tracy: (43:21)
Okaaaaay…
Joe: (43:22)
Well that's a little extreme, but you get it. You know what I’m saying?
Tracy: (43:24)
I agree with you on the branding. Never underestimate the power of branding and giving something a different name, right? So this is now a money-like deposit, except it's not really, but…
Joe: (43:34)
That's the point. So many different things in the end when you like work it back are still just some sort of like either Fed-backed or implicitly or explicitly like Fed-determined instrument. It's just how many layers of pretend do we want to have so that it looks like it's just some thing out in the market.
Tracy: (43:49)
Okay. So we've discovered that Joe doesn't want free markets anymore.
Joe: (43:55)
There’s no such thing. It's all creating the illusion that there's these real markets when in the end it's like, oh, I don’t know, that's always my take.
Tracy: (44:01)
That's true. I mean, the central bank at a very basic level is setting interest rates, right? So, okay. All right. Well on that very high level, philosophical note, shall we leave it there?
Joe:
Let's leave it there.