Bản chép lời đầy đủ của The Greatest Ever Panel on the World's Most Important Market

used to have a lot of central banks, a lot of sovereign wealth funds. They're still there, but compared to stick buyers, retail like that has grown a lot more. Nellie, does that change the way you think about debt versus you know some years. 00:03:08 Speaker 6: Ago absolutely so said prices will adjust, there will be a buyer. But it used to be decades ago we just had a much more stable investor base central banks, foreign funds. Now it's like the non bank what we would call the non bank financial institutions. It's hedge funds for various reasons, private funds who use treasuries for liquidity risk management. So the minute things get volatile, they'll want to sell treasuries to help manage their own positions. And so the investor base has changed. There will be buyers, but it could change the price and change the way prices fluctuate. You know, there just can be much more volatility given the changing investor base. And that's something that Treasury, who has to issue the debt regularly. We when I was at Treasury, probably two hundred and fifty auctions a year. They think about that, and it does affect how you think about bills versus longer term coupons and all that. 00:04:15 Speaker 7: I guess I would it's definitely saying the same thing I should start with. I thought i'd get away from disclaimers when I left the FED, but I have to say, hey, disclaimer which is this is not investment advice and place that has lots of positions, and nothing I say should implicate what positions we may have or not have that said, I you know, I think it's a similar way to ask the question is why are they buying the debt because the market's going to clear the price. We may or may not like that price, but prices used to fluctuate like all over time for various reasons. I mean, during the Civil War we had a captive demand base because if you wanted to be a bank, you had to buy treasuries, and yet the price moved right, And so for me, it's are you buying a security to hedge a liability that is of similar duration to the thing you're buying. Are you in it for the long haul? And classic examples like a life insurance company which has very long term longevity indexed is the term of art, right, It's like, as long as you people are alive, there's going to be life insurance. For companies, they have to buy debt of similar length and they're going to be very stable. They might be price sensitive, but probably less so. And at the end of the day, they have this liability that has to get funded. Banks to the same extent have these very long term liabilities deposits or long term liabilities, as we talked about that on one of the episodes. So they need long term assets to hedge the long term liabilities. Because you have bank accounts, you can get your money back whenever you want, but you tend not to, right, So that's a long term liability. A hedge fund is not in it for ten years, because that is not the nature of the business. They are responding to price signals and relative value. Treasury trading is really just a response to price signals where the market is attempting to find the lowest cost buyer. There's this great book from the nineteenth century which is inspiration for Freeman on that a Freeman Night. But like it's an interesting story which called Feeding Paris, which is by Bustiacht and a French economist, and he was saying, if one person was responsible for feeding Paris, everyone would die, because it's impossible feed a million people if you're making all of these decisions on your own. So price signals get the food to where it has to go when it has to go there. And so like the miracle of the price mechanism is the fact that Paris wakes up every morning and has food eat. And it's still true, right, I mean, cities are complicated, and so in the treasury market case, the feeding Paris equivalent is basis trades and swap spread trades and every instance of buying a security with levered money, repo and things like that, and hedging the risk with the derivative where the price difference between those things makes that worthwhile. And that's also a signal that we don't have enough of those liability hedgers who are in it for the long haul. We have to find somebody else. 00:06:48 Speaker 4: What are the data points we should be looking at, because if I look at the ten year yield, you know, it's something to do with the long term trajectory of monetary policy, and that's going to fluctuate for various reasons inflation growth, et cetera. If we want to capture some of these other dynamics such as the change and who are the buyers, or just the desire to even own US dollar denominated debt assets, what should what else should we be looking at? 00:07:13 Speaker 5: Well, So the way that I look at US treasuries, assuming that there's not real credit risk, right, yeah, I would, I would still I would still argue that there's still not credit risk more than a couple of basis points that's embedded in the current yield of say the ten year treasury, then ten year treasuries. 00:07:30 Speaker 3: Again, the way that I look at it, it have to be somewhere around nominal GDP growth. 00:07:34 Speaker 5: Right, So basically at the trajectory of what is the growth rate of the country in the longer run, and that's what the market is going to spit out, plus or minus, like you said, some kind of liquidity or either premium or discount. Now I would argue that with treasuries. To Josh's point, right there is that markets that have deep liquid funding markets, deep liquid derivatives markets, in order for someone to hedge that risk, you tend to get better outcomes and lower yields because of that. So you know, we did a study I actually when I was back at Credit Sueez, I did something actually for the for a World Bank study about what is liquidity and just about every single OECD government bond market in the world. And what you determined is bid offers were tightest when you had deep in liquid funding markets like repo and

when you had derivative markets. So you look at Italy that basically didn't have a derivative market that was particularly deep in liquid versus France, which did, and a Spain that did. Actually, so Spanish spreads were actually tighter than Italian spreads, not that the yield levels might have been the same, righting, But the difference is those deep liquid like ancillary markets around things matter, and that's where the US is unlike any other country in the world, because we have all of those things in abundance that very few other markets have, you know, And I think that's one reason why it's going to be difficult for people not to be involved with treasuries, either as a liability management tool or as a trading instrument. 00:09:05 Speaker 6: Uh well, oh, Mellie, please, I was just going to add, I think just to emphasize, you know, it is long term how to think about yield's long term nominal GDP growth, but there's a lot of uncertainty about that growth and that comes and you know, that fluctuates, and so if you're uncertain about inflation, even if you have an expected path of inflation, if it's high, it might be more volatile. Or if you're uncertain about policies, any kind of policy either you know, whether you're going to support the dollar or you're going to support the US as a safe haven, or you're going to support debt or try to reduce debt. That adds uncertainty. So then treasury is you know, like in long long run it is nominal GDP, but in the meantime you're kind of going to fluctuate what these we call premiums or discounts, you know, depending on how much uncertainty there is about that. I tend to think there's a fair amount of uncertainty about that. Right now. 00:10:00 Speaker 2: Can you convince Joe that there is such a thing as the term premium? 00:10:06 Speaker 6: Well, yes, because so because if you define term premium as the expectations hypothesis less, whatever the current yield is, there's a residual, and that is a term premium. Then you just try to define. You try to use things you know about to explain the residual, but there's always something left, and that, to me is a term premium empirically. Empirically, I don't know if I'm going to convince you. 00:10:34 Speaker 5: I think I called it on Bloomberg Radio. Actually I called it the dark matter of the treasury market. Right, that term premium must exist. The question is do we measure it properly? 00:10:45 Speaker 3: Right? 00:10:46 Speaker 5: And that's the art of it as opposed to the science of term premium. 00:10:50 Speaker 7: So I like the easiest possible way to do this, which is just to ask people what they think short rates are going to be of a long run and what long term rates are going to be tomorrow, and the Philly Fad does this recorder and there is as the Philly Fed. Okay, so they just ask economists to make predictions as to what they think this that or the other thing I'm going to do. And there's like inflation and GDP growth and all these other things.

But once a year, I think the first quarter, so we probably get that either now or soon. They ask tenure average teable yields, and then they also ask about the tenure yield, and so you're just literally asking people. There's a lot of bells and whistles you can put on these models, and some of the models with bells and whistles incorporate the survey data. Some people just look only at the survey data. Some people do just the modeling. But in all these cases there's a residual doesn't mean it's positive. 00:11:35 Speaker 3: Is the really key thing? Term premium can be negative. 00:11:39 Speaker 4: You can see why I'm unsatisfied. 00:11:41 Speaker 3: Yeah, like this is the thing is dark matter. 00:11:44 Speaker 4: They ask these surveys, yeah, which doesn't really like they ask they ask a random survey. Sometimes it gains negative. You can see why, Like I'm skeptical, Like, yeah, I'm not totally satisfied by any of this. 00:11:56 Speaker 5: No, there's a difference between the two year yield and the ten year yields, so therefore that difference also true. 00:12:02 Speaker 6: That could be the expectations of But you can write down what you think or a survey of what you think is between the two and ten and there's usually a residual and can be positive or negative. 00:12:16 Speaker 3: And having got we can. 00:12:18 Speaker 6: Often be explained correlated with things like inflation, expectations. 00:12:22 Speaker 7: Or other kinds of uncertain I tell you from experience with both dark matter. 00:12:26 Speaker 2: And turn pre josh was an actual. 00:12:31 Speaker 7: Both deeply unsatisfying it with dark matter in the from the physics perspective, well, we. 00:12:37 Speaker 3: Don't know what it is. 00:12:39 Speaker 7: There were attempts to explain it away in various like trying to hang on to the old way we think about the world is full of stuff that we can touch and see, but those never worked and there's just too much of it. And then don't even give me a start on dark energy, which is the opposite, right, And so I worked for for someone Hopkins years ago who for his PhD thesy it was told to confirm other experiments to measure the size and shape of the universe and part of that was weighing it. And so he did that experiment using supernova, which is a different way to do there's lots of with different ways to do things. Got a negative number, super unsatisfying negative mass density of the universe, which he immediately say like, okay, well this was a waste. Why did I spend two years doing this? Instead he ran with it and it turned out it was super real and it got a Nobel Prize from that outcome, which I'm not saying we'll come from term premium. Well, but sometimes the deeply unsatisfying thing is the more you dig into it, the more it's real. And I think that any way you slice that information, either literally asking people or trying to model what the market's telling you in some super sophisticated way, you always come up with a residual. Now,

the question is what is that term premium telling you? And can you find consistent ways to measure it and track it? And this positive and negative thing is clearly the case, and you know there's different microeconomic ways to explain why that should or should not be true. It really comes down to uncertainty. So and is the uncertainty correlated with yields. So if I don't know what's going to happen in the future to the economy, is that uncertainly greater or lesser when the rates go up or down and that naturally generates these dislocations. 00:14:25 Speaker 2: Can you talk about the existence of something else, which is bond vigilantes? So we just heard teleb talk about the deficit, and yet I feel like the notion that there are investors that you know, wake up one morning and say, oh, wait, I'm really worried about the deficit. Today's the day I'm gonna, you know, sell all my bond exposure. That probably doesn't happen that often. And then secondly, Nelly, I would be very interested in your take on this. But you know, when you were at Treasury, did you sit in the office going like, oh, the bond vigilantes are going to get me. I better be disciplined with my issuance schedule. 00:15:00 Speaker 3: Was that a question for everyone? 00:15:03 Speaker 6: Okay, Well, let me just no, I didn't sit there with with that, And I was at the Fed for thirty years before I went to Treasury. And you do care a lot about bond fields. I mean, it's sort of fundamental to the way monetary policy works. It's fundamental to the way you issue treasure, but you don't think about it on a daily basis, but it really influences how you view events like these scarce events and if these like you know, shocks that you weren't which by definition you're not expecting. But if you've got a system where there's a lot of leverage and you have an unexpected shock, people are going to make trades and change positions and that's when you worry. But it's not an ongoing thing. So those kinds of to sort of prevent that, you spend a lot of time as a policy maker, where do we understand where the leverages and how can we keep it manageable and make sure they can keep their funding. This is goods to the point of funding being you know, fundamental to being able to trade treasuries. So it's kind of a bigger picture, but it's not a daily thing. I don't know, but but it's important. I actually think it's a really important market disciplining mechanism. 00:16:23 Speaker 5: Yeah, the level of debt matters, right, So the bod fingilantees like, there's no group of people who get together at a bar and say, hey, We're. 00:16:29 Speaker 3: Going to go sell treasures. 00:16:30 Speaker 5: Today's the day, Yeah, exactly, like hey, tomorrow, you know the debt is going to be too big, Let's just sell treasuries. The issue, I think is manifests itself in multiple ways. And one is this this steepening of the yield. 00:16:43 Speaker 3: Curve that we've seen. 00:16:44 Speaker 5: Right in a normal environment,

you'd expect that anyway if the Fed Reserve was expected to cut rags, which it certainly has. But at the same time, you know, you do have a growing fear that when you have two trillion two point five trillion dollar deficits every year and we wind up in a a in a debt trap where interest rates and the interest on the debt ends up being so large that the fiscal agents in Washington will have to do something about it. But the market hasn't yet forced them into it. And I think that that's that forcing the government to actually act and do something is really what might have to be the impetus for you to actually get some kind of fiscal response. The challenge is political, right, and that is because fifty plus percent of our debt is interesting of excuming of our spending by the federal government is Medicare, social Security, and interest on the debt. Well, those are hard things to contend with, right, It's really really difficult. 00:17:45 Speaker 7: So I believe in bond vigilantes is not in a US context. And what I mean by that is when we talk about vigilantes, we're really referring to the nineties EM crisis, where the concern was I'm not going to get my dollar. There were dollar bonds. I'm not going to get these dollars back because the party to this debt doesn't have them and can't get them at a reasonable price, and so the bond will default. And therefore I want to get ahead of this default because you know the classic bank run. I want to get out before everyone else is before I'm stuck. In the US context, you don't have that problem. So the question is who's going to wake up and selling why? 00:18:16 Speaker 3: I'm saying why again. 00:18:18 Speaker 7: And they will sell because they are forced to sell. And we've had the Repo vigilantes so to speak, strike in twenty twenty and in twenty twenty five, and they were forced to sell for a variety of reasons. One was just the increase in the volatility market in general, and then there were margin calls, especially in twenty twenty where they were de levered, and the question then it becomes like are we heading for that kind of scenario? And the reason why the debt growth matters is because these repo vigilantis are not worried about the credit. 00:18:48 Speaker 3: Of the bonds they hold. 00:18:49 Speaker 7: They're worried no one will buy them from them because the banking system or the dealer of the bank affiliated dealers that are to be supposed to be on the other side of these trades won't have capacity, and every trade's going to keep ticking cheaper and cheaper and cheaper, and they're going to be in a difficult like sort of mark to market situation. But that's a very different set of considerations, and it's sort of related to overall growth in the debt, but it's also related to the structure of the market at how it places. 00:19:15 Speaker 4: Since we're here and we're just clarifying things for me that I've always wanted to, you know, learn about for years, over ten years of been sitting at

my Bloomberg terminal. Every once in a while you get a red headline and it's talked about like bid to cover in the tail, and I can never tell if any of these auction statistics really make it difference, like oh, terrible auction and there's always a good auction. How should I consume that information? How useful is that or for whom is that useful? 00:19:43 Speaker 5: So so we actually started just earlier this year in Bloomberg Intelligence having a grading methodology where we actually grade these from DTA A plus, and you know, we look at a variety of the bidding metrics in order to do that and how they compare it to history. So one of the big things that you've seen, and this goes to Josh's issues about structure, you go back about ten twelve years and you saw that primary dealers were the biggest buyers of coon coupon debt. Today they're the smallest. So you actually in the recent auctions, for example that we just had this week with a seven year auction earlier today, we had five year yesterday, the dealers only bought about ten percent of the bonds, whereas if you went back to twenty twelve twenty thirteen would have been they would have bought forty to sixty percent of those auctions. So the bidding metrics matter, and it matters because you can see where the primary demand is coming from. And we know now that you know, dealers, because of the changes in market structure that have occurred, particularly since the institution of Basil three, are much smaller buyers, and you know, basically end users are much larger buyers, and some of those are high frequency traders or maybe people who have repo books and kind of need to fill them by getting some collateral. Well, so all of those bidding metrics matter, but the tails will show you that the market was mispriced at the time that the auction closed versus what the aggregate demand was at that auction. And that's that tail is the single most important thing to look at, followed by then some of the details in there about who was actually purchasing, and then you know how much they did for. 00:21:25 Speaker 2: So, since we brought up market structure, it is true that the treasury market has experienced a number of volatility events at this point, which is weird because in theory it's supposed to be a pretty boring, kind of staid, old fashioned market and it's been anything. 00:21:39 Speaker 3: But you're telling me that I've been boring. 00:21:41 Speaker 2: I'm so sorry. 00:21:42 Speaker 6: I'm so sorry. 00:21:43 Speaker 3: Well, not anymore. That's supposed to be. 00:21:46 Speaker 2: Supposed to be, and we have all these things that have been put in place after every single volatility event, like you know, the RRP of the standing repo facility. We just had a change to the supplementary leverage ratio to help dealer banks hold more treasuries. Why do we still seem to have these all events happening. 00:22:08 Speaker 3: I guess we should have them sometimes. 00:22:10 Speaker 7: So the idea that treasury markets never had all events, I mean, you go back to the nineties and their massive all events in like two thousand and three is a massive mortgage extension, there was a surprise seventy five basis. 00:22:19 Speaker 3: Point hike in the night. There's always been these events. 00:22:21 Speaker 7: I think the difference now is it's harder to pinpoint a fundamental source. Like usually back then you could say, oh, this was the GSEs, this was the FED hiking rates in a way that people didn't expect. Now there's like this whole process of trying to figure out why this is happening, and it tends to happen very quickly, and it tends to disrupt a lot of relationships. But like I think, in one sense, this is stuff that's been happening in the past, it's just the market is much larger, the banking system's ability to provide that offset is. 00:22:53 Speaker 3: Is lesser, and the. 00:22:58 Speaker 7: Frequency with which trades happened it's just really gone up. I mean, like the markets are very active now. But I think that's all kind of a symptom of the issue, which is it's kind of like a just in time supply version of treasure markets, which is you have dealers can't hold a lot of inventory, so they have the match trades really efficiently. It used to be if you didn't know the buyer and the seller, you just hold it overnight. Now the high frequency traders do that for them in a very efficient, fast paced way, and then the dealers are trying to get hedge funds and set through the price mechanism to hold inventory on their behalf because basis trades are basically what dealers used to do, and that's all very fragile, and so that combination of things generates these shocks because the whole that arrangement. 00:23:35 Speaker 3: Can collapse very quickly. But you know, at the end of the day, like the. 00:23:41 Speaker 7: Size of the market is growing faster than the dealers have capacity to use. 00:23:46 Speaker 6: No, yeah, I just to provide like a policy maker's perspective, like you just step back. There's just been so many changes in technology and then the changes in the buyer base. We talked about the structural change on who buys now versus inn So like in twenty fourteen, there was something called a flash rally and the treasury I remember that, remember, and like no one understood why the treasury yield went up and down like thirty basis points in two minutes and reversed, and it was It kind of scared the public sector, you know, the government officials like how is this possible? What is the trade? And had to do a lot with these new high frequency traders. It took a lot of time to like dissect what happened. So that was even before there was a lot of treasury debt. Now we have more treasury debt and there's just you know, the volume. But I guess I would also separate the I would make a distinction between volatility events and then market ill liquidity events, just because if the mark if news is volatile, there's new changes in the

economy, you would expect treasury yields and prices to be volatile. They should, They're supposed to reflect that, and I think lot of what it's been happening recently. But the concerns are when you can't transact easily and quickly because you've pulled in more dealers that they have pulled in more than they might normally would just because of the higher volatility. So you should always get a little well, you should always get a little less liquidity when things get all little, you know, just because risk is higher. But it's when they sort of stop making markets or stop posting or something then and you can't actually transact. Those are the things that the policymakers really care about. 00:25:37 Speaker 7: There's this balancing thing where we want treasure marks, we deep in liquid. Deep in liquid means it's inexpensive to transact, which means the dealers don't make much money per trade. So the old joke like we're making losses, but we'll make effort and volume kind of thing, and like hopefully not that. 00:25:50 Speaker 3: But the. 00:25:52 Speaker 7: Response that if you want low transaction costs, the way you get that, and so a functioning business is leverage. And this has been the case for you know, seventy five years since the Treasury fed a cord that this was always the core issue. And so when you leverage constrain banks, and even if the bank isn't leverage constrained, when the desk is leverage constrained. When leverage is a zero sum game within the institution, which is kind of what these leverage ratios do. Everyone's fighting over the same resource and that process introduces friction. And at the end of the day, I think these all events are mostly just time slippage. Like if you have to think about things for too long, the market can run away from you. So you know, in twenty twenty, if you had to spend two days figuring out who gets incremental balance sheet, a lot can happen in two days in March of twenty twenty. And these very human experiences are kind of a drive today. 00:26:40 Speaker 5: And we talked about this on the show that we did back in late April, about the April event, and that time slippage is exactly a big thing part of what happened when right before you fell asleep on April ninth, Right, it's because like, look, you can't call the New York dealer desk to get more dealer balent sheet at eleven thirty at night in New York time, and when you're trading in Hong Kong, right' It's it's hard to do that. So so you get these vol events that are create a liquid markets, but only at certain points in time, right, and then that always gets armed away. 00:27:16 Speaker 3: You know, people are are. 00:27:18 Speaker 5: You know at the end of the day where we're definitely not price takers, right, there's a lot of people who are you know, basically want the price of their of the asset to reflect the risk that they're taking. And so you're going to get these instantaneous shifts and expectations when you get a news event, when you get a headline from you know, Donald Trump, and you think that maybe the dollar is not going to be the reserve currency anymore. That's going to affect dollar assets regardless of where they are in the world. 00:27:59 Speaker 2: This has been an another episode of the Authots podcast. I'm Tracy Alloway. You can follow me at Tracy Alloway. 00:28:04 Speaker 4: And I'm Jill Wisenthal. You can follow me at the Stalwart. Follow our producers Carman Rodriguez at Carman armand dash Oll Bennett at Dashbot and. 00:28:12 Speaker 3: Kill Brooks at Kilbrooks. 00:28:13 Speaker 4: For more Oddlots content, go to Bloomberg dot com slash odd Lots where we have a daily newsletter and all of our episodes, and you can chat about these topics twenty four to seven in our discord discord dot gg slash outlines. 00:28:26 Speaker 2: And if you enjoy Odd Lots, if you like it when we do these live recordings, then please leave us a positive review on your favorite podcast platform. Thanks for listening. 00:29:01 Speaker 1: Eight