Fejau on the Treasury Quietly Running Yield Curve Control
- Fejau’s core call is that Treasury is already suppressing long-end yields through issuance and buybacks—a functional, if not textbook, form of yield-curve control. Since 2022 it has tilted funding toward bills, held coupon supply flat, and financed purchases of illiquid off-the-run bonds with near-cash T-bills, shifting duration risk out of the market much as QE did through the Fed’s balance sheet.
- The new information is escalation: QRA wording shifted from no planned “increase” in coupon issuance to no planned “change,” opening the door to cuts, before long-end buybacks rose from $2 billion to at least $4 billion and Bessent said routine operations would exceed $4 billion. The initial dollars are small, but the signal is “whatever it takes”; after the bond market erased the first move, Fejau framed the response as: “No, no, we’re just getting started, buddy.”
- The emerging “Treasury put” matters because long yields transmit into government interest costs, mortgages, risk assets, and corporate funding. Federal interest payments reached roughly 3.3% of GDP, an all-time high by this measure; Fejau also says equities become sharply vulnerable when the 10-year exceeds 5%, while AI hyperscalers increasingly need corporate bonds after exhausting much of their free-cash-flow funding.
- Fejau expects the dollar to become the intervention’s exhaust valve, making gold and Bitcoin cleaner expressions than US equities. A falling dollar can erase foreign investors’ equity gains in their home currencies; gold is his higher-confidence, more direct hedge, while Bitcoin also benefits from washed-out positioning—though he concedes the latest surge may partly be a liquidation run after open interest accumulated.
- He is overweight, not all-in, gold and Bitcoin, has accumulated them over the last couple of months, and expects to hold for at least six months. Short-term traders might sell some of the pop, but his own stance is: “I don’t think you go through the pain we’ve seen in crypto markets over the last year or so and then you just sell the first pop.”
- Political incentives make further escalation likely through the November midterms, even though Bessent previously accused Yellen of hijacking monetary policy with essentially the same playbook. Fejau calls the hypocrisy real but the outcome structural: having started support in August, the administration has little incentive to stop in September or October, while a later split Congress could absorb blame if inflation and yields reaccelerate.
- Japan shows that explicit yield-curve control becomes enormously expensive and eventually yields to market forces, but fading intervention too early can still be ruinous. Fejau assigns about 5% probability to the maximum escalation—Fed purchases of the bills funding Treasury’s long-bond buybacks—but calls that “textbook debt monetization” and “Pandora’s box”; until proven otherwise, his rule is: “Don’t fight the Treasury.”
1. Treasury has moved duration control from the Fed’s balance sheet to issuance policy
Fejau begins with QE: the Fed swaps central-bank reserves for long bonds and warehouses their interest-rate risk. His useful simplification is that 10-, 20-, and 30-year bonds behave more like risk assets, while a three-month T-bill is cash-like yield with almost no duration risk.
Removing duration leaves the same capital chasing less supply, raising bond prices and lowering yields. It also pushes investors outward on the risk curve: if a long bond offers only 1%, its former buyer may choose equities instead. “That was the whole ballgame” throughout much of the 2010s.
Beginning in 2022, Treasury pursued a similar result before bonds even reached the market. Under Yellen, it issued more bills and constrained coupons: “What does it look like where we just don’t put it into the market in the first place?” Fejau treats deep demand for bills as the enabling condition.
Treasury’s long-term target is about 20% of debt in bills, by Fejau’s figures. The share rose from roughly 17% after COVID to about 22% now—already above that range—while each quarterly refunding announcement, or QRA, communicates how much Treasury intends to issue across maturities.
2. Bill-funded buybacks are QE-like—and the policy signal just intensified
Unlike the Fed, Treasury cannot create unlimited reserves, so it must fund purchases. Its mechanism is to issue highly demanded, nearly durationless T-bills and use the proceeds to buy long-duration bonds.
The buyback program has two components: buying T-bills as a liquidity-provision measure, and buying long-end, off-the-run bonds to remove duration from the market. Once a bond is no longer part of the latest auction, it becomes off-the-run; during the COVID shock, those were the bonds that became especially illiquid.
The first tell came in the latest QRA. Guidance formerly said Treasury did not expect to “increase” coupon issuance over coming quarters; replacing that word with “change” introduced the previously unexpected possibility of decreasing long-end supply. Fejau paired that with yen intervention as evidence of accelerating suppression.
Treasury then announced that long-end buybacks would rise from roughly $2 billion to at least $4 billion. During the interview, Thread Guy read fresh Bessent headlines promising routine buybacks “more than $4 billion”—alongside Bessent’s claim that rates had nothing to do with the decision. Fejau’s interpretation remained: small size, huge signal, with markets invited to test official resolve.
3. The long end is pressuring fiscal arithmetic, borrowers, and risk assets simultaneously
Fejau sees no single cause for rising long yields: war, oil and inflation, the yen, globally higher yields, fiscal deficits, and doubts about Fed credibility all contribute. Inflation has remained above the 2% target for “six, seven years,” weakening appetite for decades of duration exposure.
Government interest payments have reached approximately 3.3% of GDP, an all-time high on the chart Fejau cites. That gives the “Treasury put” a fiscal rationale: higher refinancing costs compound the debt burden even as investors demand more compensation for inflation and duration risk.
His bond-loss example is deliberately simple: buy at $100 with a 3% coupon, then watch rates rise to 5% and the price fall, illustratively, to $90. Holding to maturity can restore par, but forced sellers crystallize the damage—the same unrealized-loss dynamic that mattered during the 2023 regional-bank crisis.
Issuing everything as three-month bills would avoid duration today but create extreme rollover risk tomorrow: each maturity must be refinanced at prevailing rates. It would also weaken the long-term confidence and maturity diversification associated with a reserve-currency bond market.
Fejau dismisses auction-failure and default doom as unlikely because the Fed can backstop the market; the more plausible adjustment is currency debasement.
4. Lower long yields support housing and AI financing—but the dollar takes the strain
Mortgages and housing are anchored to the long end, making elevated yields immediately political. Fejau also points to a corporate channel: hyperscalers initially funded the AI build-out from abundant free cash flow, but as that capacity is “tapped out,” they must issue equity or corporate bonds priced partly from long Treasury rates.
He cites a consequential threshold: when the 10-year rises above 5%, its relationship with equities becomes significantly negative—further increases tend to accompany falling stocks and other risk assets. Suppressing the long end therefore serves more than Treasury financing; it helps preserve broader financial conditions.
Yet the intervention’s “exhaust valve” is the dollar. If US stocks rise modestly while the dollar falls, a European pension manager can still lose in euro terms. That helps explain why US equities did not perform well after the announcement and why Fejau expects weaker marginal foreign demand for dollar-denominated assets.
Gold and Bitcoin avoid that exact cross-border problem because, in his framing, “they are not US assets.” They express declining confidence in the bond-and-currency complex more directly than buying the Nasdaq and then managing dollar exposure.
5. Gold is the cleaner hedge; Bitcoin is the higher-noise companion trade
Gold historically traded tightly against real rates—higher real rates meant lower gold—and that relationship helped explain its difficult preceding six months. Fejau says another driver now matters: declining faith in US bonds and reserve diversification, including heavy Chinese gold purchases.
Bitcoin is “on its way” toward the same role, but Fejau refuses a single-cause story for its rally. It had already fallen substantially; traders were shorting for another leg down before a hoped-for October “giga bottom”; and accumulated open interest meant liquidations could explain part of the sudden move.
His confidence is therefore asymmetric: gold offers the cleaner one-to-one exposure, while Bitcoin remains “worth a punt,” especially because “there’s no more sellers” after the market’s washout. The macro thesis and favorable market structure reinforce one another without making the attribution certain.
Time horizon governs the trade. Fejau accepts that short-term traders could sell into strength, but he has accumulated both assets slowly over the last couple of months and plans to sit for at least six months. He is “somewhat overweight,” not 50% gold and 50% Bitcoin, and can tolerate surrendering half the immediate gain.
6. Bessent’s reversal is hypocritical, but Fejau thinks the chair dictates the policy
Before becoming Treasury Secretary, Bessent criticized Yellen for using issuance policy as a version of QE that usurped the Fed. Now he is extending that playbook. His earlier “three-three-three” ambitions—3% GDP growth, three million barrels of oil, and a 3% 10-year—make the reversal look especially stark.
Fejau’s concession is blunt: “I don’t like how gaslighty he is about it,” and Bessent damaged his credibility by promising a different course. His deeper conclusion, however, is that pressure from the president, fiscal arithmetic, and market structure would drive almost anyone in the seat toward the same outcome. “It’s the nature of the beast.”
With Trump’s approval ratings described as weak, Republicans likely to lose the House, and the Senate a toss-up, Fejau expects policy to play “Whac-A-Mole” through November. Why deploy support in August, he asks, only to stop during September and October immediately before voting?
His rough post-election scenario is a split Congress followed by renewed inflation and surging long yields around January or February 2027, letting the administration blame legislative deadlock. He remains uncertain whether Warsh is intentionally letting the long end tighten conditions or whether officials are simply “building the plane as we’re flying it.”
7. Japan defines the limit, while the escalation ladder still has several rungs
Japan is Fejau’s model because it is “10 years ahead” in endgame central banking. The BOJ explicitly capped the 10-year and promised unlimited purchases; once inflation returned after COVID, the yield repeatedly hit the ceiling, forcing enormous intervention and successive band widening from roughly 25 to 50 basis points.
Japan ultimately retreated and JGB yields normalized sharply from near zero. The feared market “Armageddon” did not arrive, though the process was unruly. Fejau’s lesson is that direct YCC is not durable—but also that traders can lose heavily waiting for free-market equilibrium to overpower official balance sheets.
Fejau distinguishes technical YCC from the broader label: the US is not formally promising unlimited purchases at a fixed yield, but he considers the ongoing suppression of the long end a functional version. The terminology matters less to him than the signal.
He also notes that bond-market demand is driven at the margin by institutional flows, including hedge-fund basis trades that buy off-the-run bonds in cash and sell futures. The leverage involved may add fragility. As an extreme example, he recalls—while noting he may not have the specifics perfect—that Brazil once reached a point where virtually all issuance was bills.
For monitoring, he prioritizes 10-, 20-, and 30-year yields; the 2s10s curve and whether steepening comes from falling two-year yields or a “bear steepening” long-end selloff; the MOVE bond-volatility index; priced Fed hikes or cuts; and the dollar. Buyback results are public, allowing investors to compare rhetoric with executed purchases.
The maximum escalation would be Treasury issuing bills to finance long-bond buybacks while the Fed itself buys those bills—“textbook debt monetization” rather than ordinary market funding. Fejau assigns that maximum scenario about 5% probability, while Thread Guy jokes that it sends Bitcoin to $10 million. The framework matters more than the number: “We know the maximum level,” several intermediate steps remain, and “don’t fight the Treasury.”
Full transcript
Hey, Zhao, Mr. Felix. What’s up, man?
DG, what’s up, dude? How are you?
It’s good to see you, man.
Yeah. It’s macro season again.
I’ll tell you what, I’ve had a brutal 24 hours LARPing as a macro specialist, so thank God you have joined us. Dude, your episode—
I was listening in, yeah.
—was sick yesterday, by the way.
Thank you. Thanks, man, appreciate it. Yeah, I saw you had it up. You were watching, trying to piece through it, so I have some context. Hopefully I can give you some explanations here.
Let’s fucking go. I mean, I know there’s a lot to break down. It’s a 45-minute episode; there’s a lot in there. I sent the link, but do you want to just start with why the announcement yesterday and what Bessent did was significant, and why the debasement narrative has popped back up with Bitcoin, gold, hard assets, and commodities?
Yeah, yeah. Okay, there’s a lot of context and definitions that we need to get right here to fully explain this.
1. The Fed Removes Duration
For all of the 2010s, a lot of marginal increases in liquidity, or whatever you want to call it, came from the Fed and QE. We had these QE programs. What is QE? What is quantitative easing? It’s the Fed going into the market and directly buying bonds and swapping them for central bank reserves.
That sounds wonky, but what’s important is that the Fed has unlimited amounts of central bank reserves. That’s something that’s pretty much money-like between different banks. It’s a dollar that’s just—
Okay.
—between different banks. So they swap that, and then they buy a bond. Typically, for QE, it’s long-duration bonds.
When we talk about duration, it gets mixed up a lot with maturity. We don’t need to get into the wonkish math around duration, but an easy way to think about it is as similar to maturity: 10-year, 20-year, 30-year. Those act more as a risk asset than something like a T-bill. A 3-month T-bill acts more like cash.
Got it.
When somebody’s trying to buy a long-end bond, like a 10-year or 30-year bond, they’re weighing that against different things, like more risk assets, because you’re taking on more risk around your expectations of inflation.
There’s only a certain amount of capital in the world that wants to own risky assets like that, such as a long-duration bond, versus the short end—cash that earns a yield without much duration risk, or interest-rate risk, which is another way to say it.
Everybody wants that. Who doesn’t want a 3-month T-bill that pays them, I don’t know, 4% or whatever it is these days?
And when I’m looking at the 30-year chart, I’m looking at the riskier bond?
Yeah.
Got it.
More interest-rate risk that you’re taking on.
Got it.
For years, to get financial conditions looser and really stimulate, they would try to take duration out of the market. Supply-demand dynamics are really useful here to think about. When you’re taking duration out, that means there’s the same amount of capital chasing after less and less duration.
Yeah.
That means higher demand for those bonds, which means lower interest rates. It also means that if there’s this idea of going out on the risk curve, somebody who’s buying that long-end bond might say, “Okay, well, now I only get 1% on it. I’m going to go buy equities instead, because I might get a better return there.”
That was the whole ballgame for a long time.
And just so I’m clear, taking duration out means the Fed is just buying these bonds and locking them, basically taking them out of the market—buying and holding them?
Yeah, locking them on their balance sheet and holding them. They’re warehousing that duration risk on their balance sheet, and they’re the Fed. Who cares, right?
Yeah, yeah.
Versus in the market, it matters a bit more. So that was the name of the game for a long time.
2. The Treasury Changes the Game
Then, in 2022, we started to see it shift to, “Okay, so that’s just taking it out. What does it look like if we just don’t put it into the market in the first place?” That’s what the Treasury was doing. That’s what Yellen started to do.
Instead of issuing a lot of long-duration bonds into the market and making it absorb that duration, what if we just issue a lot more short-term T-bills, like 3-month T-bills, instead? It’s basically like giving them cash in some ways.
There’s unlimited demand for T-bills, really. Everybody wants T-bills. Instead of issuing a lot of long-term debt, we’re just going to constrain the amount of duration we’re putting onto the market.
That’s what Yellen was doing over the past few years. Ever since then, we’ve had this interesting regime where they provide guidance on whether they expect to see any sort of increase in the amount of long-end duration bonds they’re going to issue, and they’ve held it flat for quite a few years now.
This is where Bessent gets a lot of shit, because he wrote op-eds before he was Treasury Secretary criticizing Yellen for doing this and saying that you’re basically playing monetary policy from the Treasury side of things.
This was in 2024?
Yeah, around there.
Is there any rule or standard for the quantity of issuance for each vehicle?
Yeah, kind of. There’s a long-term target where they want to have about 20% of the total debt issued be T-bills. After COVID, about 17% of the total debt was T-bills, so there was valid room to go from 17% up to that long-term average of around 20%.
We’re at about 22% now. Within that context, we’re above the long-term range.
Every quarter, we have this thing called the QRA, the quarterly refunding announcement, where they provide guidance on how much they plan to issue and what durations or maturities they’re going to issue.
They’ve starved that long end by just providing as little as possible. Then the next thing they added on top was this idea of Treasury buybacks.
If you think about QE, during QE the Fed can just swap the bond for a central bank reserve.
And these are essentially dollars?
Yeah. People make arguments about whether that’s net-new money creation, or money printing, or not. I don’t know; we don’t need to get into it.
But when the Treasury does it, they don’t have unlimited amounts of money like the Fed does, so they need to fund the buybacks with something.
Got it. Okay, okay, okay, okay.
The thing they’re funding it with is T-bills.
Got it.
Like we said, there’s an unlimited amount of demand for T-bills. They issue a bunch of T-bills with no duration, or almost no duration, and then they buy back—
Which have infinite demand, just like cash.
Yeah, pretty much. There’s an insane amount of demand.
And who’s buying these T-bills? Because it’s a massive size, right? Who is it?
I mean, it’s everybody. It’s banks; it’s people. U.S. citizens can go to TreasuryDirect—
Are you buying them as a retail trader?
Yeah. You can go to TreasuryDirect, participate in the auction, and buy those T-bills.
Got it.
Or when you go into a money-market fund—say you have some cash you want to put in a money-market fund—you give it to the money-market fund, and then they go out and buy the bills. They pass that yield on to you.
Okay. Okay, okay.
Now we have these Treasury buybacks. They started them over the last couple of years in small amounts, super-small amounts.
One portion of those buybacks was to buy back T-bills, and that was a liquidity-provision thing. We don’t really need to get into it. But the other component is basically the long end.
In the long end, there are 2 types of bonds: the off-the-run bonds and the on-the-run bonds.
The super simple way to think about it is on-the-run bonds are the bonds that were just issued.
When bonds get issued, typically it's at, like, $100 par value. But say interest rates go up, then that par value goes below 100, so it's trading at a discount to reflect those higher interest rates.
So once those bonds become owned, or old and not part of that latest auction, they become off-the-run bonds, which are much more liquid. Those are the things where, when we've seen Treasury—like during the COVID crash—we saw people talk about the Treasury market freezing up. This is what froze up: the off-the-run bonds. Nobody wanted to buy those things. They're super, super illiquid.
So now what they're doing with these Treasury buybacks is they issue a bunch of bills, and then they go into the market and buy those illiquid, long-duration, off-the-run bonds, and then take them out of the market. So they're removing the duration out of the market. It's quite similar to QE in many ways.
We've had a steady amount of buybacks, and then obviously the context now for the market is, in recent weeks, we've seen long-end yields going higher.
Yeah.
Everybody's been freaking out.
Is this primarily war, oil, inflation-driven, or—
It's a lot of different things. It is that, for sure. It's—
The yen situation?
Yeah. It's the yen situation. I mean, globally, yields are going higher, generally speaking. There are questions about fiscal spending, excessive spending, deficit spending. There are questions around Fed credibility because the inflation target's been above 2% for, like, 6 or 7 years now. So there's a lot of different things moving it.
Yeah. Okay.
But the fact of the matter is that when the 30-year yield is getting above the levels it was reaching over the past couple of weeks, that starts to have a lot of pretty significant impacts on markets.
Okay.
The big signal that happened this week is that every quarter, the Treasury meets for the QRA, where they decide how much of each tenor they're going to issue. They came out with that QRA a couple of weeks ago, and it was pretty vanilla, other than the fact that the guidance we talked about around how much they're going to issue was always about saying, “For the next few quarters, we don't plan to increase coupon issuance. We don't plan to increase the amount of long-end bonds that we're going to issue to the market.”
They made a change to the word “increase.” They changed it to “change.” So by changing it from “increase” to “change,” suddenly they're saying, “What if we actually decrease the amount of long-end bonds that we issue?” That was not on anybody's cards. Nobody was expecting that.
So that was really interesting. It was the first signal. That's what really started to get me excited a few weeks ago. That, paired with the yen intervention that was going on, was like, okay, they are moving to elevate or accelerate the amount of manipulation or suppression of the long end.
And then, lo and behold, we get this announcement yesterday that, literally two weeks after they just met for their quarterly announcement, suddenly, out of nowhere, they're like, “Look, we're going to double the amount of 30-year—or of long-end—buybacks that we're doing. At the very least, we're going from $2 billion to $4 billion.” And there's this, like—
At least $4 billion, right?
I know you're saying this in the stream. Yeah. Very, very aggressive language. And it's like, okay, that just came out of nowhere. That's interesting.
I'm sure that came from the top. Treasury is political. Bessent is under the Trump administration. So, out of nowhere, after their QRA meeting, they come out with this news. That was just kind of the full green light of, like, okay, look, they are going to throw everything at this.
Going from $2 billion to $4 billion is not a lot, but the signal is huge, and the language is huge. In the same way that QE—you know, back in 2020, yes, they were doing a lot of QE, but a lot of the move was done by the market through the signaling mechanism. The markets do a lot of the work.
So the signal here is, “All right, we're going to do whatever we can to keep the long end lower.” And because there are a lot of important things happening that are associated with the AI build-out, the AI build-out was largely funded by these hyperscalers because they had so much free cash flow.
But that's been tapped out now, so now they need to either issue equity or issue corporate bonds into the market. Corporate bonds are priced off the long end, or at least attached to the long end.
Oh.
That's why people are starting to tie the two together: okay, is the amount of—
So it's fucking expensive for them to tap more markets for cash.
Yeah, and so then the credit spreads widen. But if you try to keep the long end lower, that does help with keeping that whole issuance that's probably going to be accelerating going.
There are a lot of different things at play. I know I'm kind of jumping everywhere, but the fact of the matter is—
No, it's great.
The fact of the matter is that it's a major signal of the commitment that's ongoing.
And then the last part of that commitment is like, okay, well, if we're doing all of this in August, and we have the midterms in November, just think through the game theory and the incentive structures. Why would they throw everything in August, and then in September and October just be like, “Nah, we're done. We're good. That's it,” right before the midterms?
To me, it's clear that we're in the early innings of this escalation, and the market, especially the bond market, is not going to jump on the wagon right away. You can see a lot of that. Obviously, when this buyback got announced yesterday, yields came lower pretty quickly.
Yeah.
We've already given back a lot of it.
They're pretty much at the same price before, right?
Yeah. We're going to have to see a lot of this. It's going to take a while before the market is convinced of their resolve.
But in the meantime, it's like, okay, how do I ride the coattails of them trying to tamp this down? For me, it's those classic debasement assets: Bitcoin and gold.
What's really interesting is that, basically since that announcement happened, U.S. equities haven't really done all that well.
At all.
And my theory is two things there. One, what they did yesterday—the dollar just nuked—and typically, a lower dollar means higher risk assets, generally speaking.
Yeah.
But if you look at where the flows have been in recent years into U.S. equities, it's been a lot of foreign buyers. Foreign buyers make money in a different currency, and then they need to buy U.S.-denominated assets.
If a bunch of European pension managers are long a bunch of U.S. stocks, and then the dollar goes lower, that means the euro goes higher. So, on a currency basis, even though U.S. equities might go up a bit, because of that dollar decrease, they actually probably lost money.
This outright intervention is them trying to manage that long end, but at the same time, it obviously leads to a lower dollar. So, it leads to marginally less interest in exposure to U.S. assets, which is why I think it's really important to consider that gold and Bitcoin are not U.S. assets. They are more pure expressions of this questioning of—
Whoa, we just got an announcement.
Bessent: “We will routinely do buybacks and increase the size of buybacks.” Bessent: “It will be more than $4 billion.”
Whoa.
Whoa.
There you go. So the bond market was testing him today. They gave back, and he's like, “No, no, no, we're just getting started, buddy.”
Whoa. So—
So this is the signal.
Wow, dude. So they're—
Yeah.
He basically said, “Call my bluff.” The bond market goes up, and he says, “Fuck you guys, I have infinite money to throw at this.”
Pretty much. There's this famous thing that happened in central banking in the 2010s in the ECB. Mario Draghi had this “whatever it takes” thing. This was during the European debt crisis, and people were asking, “What are you going to do? What are you going to do?” Shit was hitting the fan, and he was like, “We'll do whatever it takes, and believe me, it'll be enough.” It's this famous quote.
I kind of feel like it's the same situation here. Bessent has some work to do to convince the market, as you can see.
Yeah.
But he’s right back there in the headlines right now. He’s going to work to convince them.
Okay, I’ve got a bunch of questions for you. By the way, you pretty publicly came out and said you were buying Bitcoin about a month ago, if I’m not mistaken.
Gold and Bitcoin, yeah. To be honest, initially it was because I felt like we were at peak hawkishness. People were pricing in rate hikes in July and September, and I didn’t really think those were going to play out, so it felt like a good moment.
On top of everything that’s happened in the last month with the intervention stuff and the QRA, it’s clear that you want to own debasement assets. Both of those charts have been wrecked over the last few months because of those hawkish fears and higher real rates. So I got in on that, and now my confidence in them is building a lot.
3. The Treasury Put Takes Shape
Can you explain the significance specifically of the 30-year and the price that it’s at? Why now? Why is Bessent saying, “We will empty the kitchen sink to defend this line” on the long end?
What are the biggest implications of it being really high or really low, and why are they all out on defending this level?
It’s nebulous where the line gets drawn in the sand. Everybody has their theory of where that line is, or where the Treasury put is, as we can probably start calling it.
One way to look at it is interest expense as a percentage of GDP. Obviously, the other side of the equation of all this debt that we’re issuing, higher inflation, and higher yields is more interest expense from the government. Just last week, we saw interest payments as a percentage of GDP hit an all-time high.
What percentage? What’s the number?
I tweeted it the other day.
3.5% of GDP is interest payments?
Yeah.
That’s nuts.
Let me find the chart. It’s close—3.3% or so. That’s an all-time high. Obviously, on a nominal basis, interest payments have already been higher, but it’s important to look at them as a percentage of GDP. That’s what broke out.
That, paired with higher inflation, starts to lead to a loss of confidence in the long end. In this secularly higher-inflation regime, nobody really wants the long end because they don’t want to take on that duration risk.
If I buy a 10-year bond at, say, 3%, and then rates go to 5%, I’ve lost money. Even though I’m earning—bond market stuff gets really wonky really quickly—you can have a 30-year bond where you’re getting a coupon of, say, 3%, but you can lose money on the price.
Say that bond you bought at $100 par value, at a 3% rate, goes to 5%. Say it goes down to $90, so you’ve lost $10 on that.
Got it.
It gets even wonkier because that only matters if it’s an unrealized loss. You heard about this back in 2023 during the regional banking crisis. A lot of the issue was that because inflation went higher and rates went higher, all these bonds were worth less at their par value.
I see.
Eventually, if you get to maturity and keep getting those coupons, it will mature at $100. You don’t actually realize the loss unless you have to sell. The whole name of the game is avoiding that situation.
But you’re not holding it for 30 years.
Most people aren’t. Pensions own them, and some people are.
So if you’re willing to go the distance, you don’t lose.
Yeah.
But most people don’t want to go the distance.
Yeah.
Got it.
The other side of the equation is that, in a modern economy, a lot of what makes a reserve currency comes down to whether there’s long-term confidence in your bond market and demand for that debt.
It can be really risky from a risk-management perspective for the Treasury to just issue all debt as T-bills—as 3-month T-bills. There’s no duration, and so on.
I was about to say, why don’t we just take the 30-year out entirely and only issue short-term debt?
There was, I think, a moment in the early 2000s when they didn’t issue any 30-years. The 20-year is relatively new, from the last decade.
The reason you don’t do that is because it’s basically what a banana republic looks like. It exposes you to much higher rollover risk. Say that 3-month T-bill expires in 3 months. If interest rates are higher, suddenly that’s going to cost more.
Got it.
You’re exposing yourself to rollover risk. They’re trying to manage risk across the whole maturity spectrum. If we issue a bunch of 30-year bonds at 2%, even if rates go to 5%, we only have to pay 2%. That’s what Bessent is thinking about.
But if we issue a bunch of bills at 3% and rates go to 5%, when those bills roll over in 3 months, suddenly we have a high rollover risk, and we have to reissue those bills at a higher yield to maturity.
There are a lot of different aspects to it. People get really doomer-y and think there might be a Treasury auction failure or that the government might default on the debt. Those aren’t likely.
What’s a Treasury auction failure? How does that work? No bidders?
Yeah, but it’s not even worth contemplating.
Okay.
There are backups. Literally, the Fed will go into the market and cover it if they need to. It’s not going to happen.
What will happen is this debasement that we’re seeing, and the exhaust valve of that is the dollar. Therefore, riding the coattails of debasement assets like gold and Bitcoin is where you want to be.
That was what I was going to ask next. There’s also another TradFi tweet: “U.S. Secretary Bessent: Rates have nothing to do with buyback decision.”
Why are debasement assets the biggest beneficiaries of this? How big of long-term beneficiaries are they of the decisions being made right now?
Gold is a lot cleaner from that perspective. It has a lot more history behind it. Its correlation has changed over time. It used to be really tightly linked with real rates: when real rates go higher, the price of gold goes lower, and vice versa.
That’s why it’s struggled over the past 6 months. Real rates have been going significantly higher, and that’s been hurting gold.
But there’s another component to gold now, which is a lack of faith in U.S. bonds, or diversification by other countries. China has been buying gold like crazy. It’s a very clean exposure because it’s not as exposed to cross-border capital flows, like the thing we talked about with our European pension managers.
Yeah.
It’s just gold. It’s not like buying the Qs, where the Qs are priced in dollars and then you have to wonder about currency hedging. It’s just, “I’m buying gold.”
Buying gold.
Bitcoin is on its way to that as well. This is the whole thing about markets. Bitcoin ripped, and it’s awesome. I bought a lot of Bitcoin for this reason, and I’m happy right now.
But how much of it is because of this? How much of it is because it’s already down a bunch, and a lot of people started shorting the bottom because they expected one leg lower? Then you buy the giga-bottom in October.
Over the last week, a lot of open interest has built up in Bitcoin. You could probably attribute a lot of this liquidation run to that as well.
It’s always hard to piece apart, but I feel like it’s a cleaner read on exposing yourself to this. Obviously, I have higher confidence in gold being one-to-one tracked in that, but I still think it’s worth a punt on Bitcoin as well, especially because the market is just so washed out. The market-structure side of things is just like, yeah, there are no more sellers.
What was the biggest criticism of Yellen when she was doing similar actions in the market, like in this paper?
They basically believed that it was a version of QE that was usurping monetary policy. The criticism was that Yellen was hijacking monetary policy from Powell and the Fed. Suddenly, it was a political institution deciding monetary policy in some ways.
Got it. Bessent himself—literally, Bessent was the one who was most vocally critical about this.
Yeah. Yeah, he wrote about it.
So he’s doing it right now. Is this a desperation sort of play?
I think it’s just like, man, this is the whole game theory of why you never—like, you’ve heard Tyler on Four Guys talk about—
Yeah.
—the Ponzi continues. This is the Ponzi. It’s the game theory that it’s easy to criticize from the outside, but when you’re in that chair and suddenly you have pressure from your president and all these different vectors, it’s easy to criticize from the outside. But once you’re in the seat, it becomes very different.
Wow.
I don’t like how gaslighty he is about it, but I also understand that if you put anybody else in that seat, they would probably come out with the same outcome. I think the game theory is kind of set there. It’s the nature of the beast; there are bigger forces at play.
He probably didn’t do a good job for himself by criticizing this and talking about how he was going to change everything. He had these ideas of “3-3-3,” right? 3% GDP growth, 3 million barrels of oil, and a 3% 10-year yield. He kind of shot himself in the foot by talking about all these great things he was going to do and then just—
That’s what makes it very hypocritical. But the fact of the matter is that these events are kind of— the game theory is set. The playing field is set. I think this is the outcome regardless of how you play it or who the player is.
What about doing this? I just read the chapter of More Money Than God where they talk about Soros breaking the Bank of England, and Bessent and Druckenmiller are actively intervening with the yen. They’re trying to prop it up instead of devaluing it.
I guess they’re trying to prop it up instead of devaluing it, which is like—the devaluing is the cardinal sin, but adding to it is a little bit unspoken. I watched this Patrick Boyle video. I laugh as a macro guy, but what do you think of their ability to prop up the yen market while also fighting this yield battle on the long end? Is it too many things at once, or are they sort of independent of each other and have no real relationship?
4. Japan Shows the Limits
No, they’re all interconnected. It’s really hard to fight market forces like this with direct intervention. You have to throw a lot of money at it. There’s a reason the Fed balance sheet became trillions of dollars over the last decade. It takes a lot of money to move the market.
I know in the roundup you’re looking at, Quinn was talking about going to look at Japan to understand where this goes. The reason he said that—
Price gold in JPY.
Yeah. And the reason he said that is because Japan is kind of 10 years ahead of everybody in terms of endgame central banking in many ways. They did the yield-curve-control experiment, where they directly said, “The 10-year is not going above this level, and we’ll buy unlimited amounts to make sure of that.”
It took a lot of money for them to do that, and eventually they had to give up on the whole thing and reverse out of it. It’s hard. I don’t know how tenable it is. Eventually, you would think that you can’t keep kicking the can down the road and that eventually we have to get our house in order here.
There’s a long history of people who have been smoked in that idea, in that trade of trying to fade it. So yes, right now the market is testing Bessent and seeing his resolve, but I don’t see that ending anytime soon. I think he’s going to continue to try to keep all these things afloat.
How tenable it is, I don’t know. Maybe this is the max. I don’t think so, because I’d be surprised, like I said, if they threw everything they could at it in August and then just sat on their hands for 2 months into midterms.
But it’s not easy to do these direct interventions. Eventually, free markets do find an equilibrium. A lot of money gets burned in the interim, and that’s where we can make some trades.
How did Japan get out of it? What did the unwinding look like?
It happened over the last year. After COVID, they finally started to get inflation again, because obviously, for decades, they had just been in this stagnation. They started to get inflation again, and what started to happen was that with direct yield-curve control, almost every week they started to hit the upper band. Suddenly, the BOJ had to spend a stupid amount of money trying to defend this level.
Whoa.
Initially, it was like, “Okay, we’re going to widen the band before we intervene.” Now we’re changing the band and moving it up from, say, 25 bips to 50 bips. They just kept ratcheting higher.
That’s why the JGB long-end yields have gone up so much: there’s a lot of normalization to do to go from basically zero to reflecting where inflation was. A lot of people thought that would just cause Armageddon in markets, this normalization of the JGB market, but it didn’t. It was still somewhat unruly.
I think the big lesson there was that direct yield-curve control is not really a useful tool. That gets into one thing I wanted to talk about with everybody: as soon as these things happen, everybody tries to jump on what the right acronym or definition is. Is it QE? Is it yield-curve control?
Your favorite. You love QE.
Yeah. The issue with those discussions these days is that people use QE and yield-curve control to mean different things, rather than the direct definitions they were originally conceived for. That’s fine. I’ve come to relent on it because this is how language evolves.
Now QE just means any sort of liquidity easing, and yield-curve control refers to any sort of suppression of the long-end market, however we do it. We don’t have direct, formal yield-curve control where the BOJ says, “The 10-year is not going above 4%, and we will spend unlimited amounts of money to make sure of that.” That’s not what we’re doing, but we’re certainly doing some version of yield-curve control.
I would make the argument that we’ve been doing it since 2021. Have we been in a regime of yield-curve control? I would say yes, but obviously, in the technical definition of it, we haven’t. I don’t think that matters as much. The signal is that there is some sort of control of the yield curve.
Yeah, yeah, yeah. Okay.
Yeah, yeah.
What do you think is significant about this happening 2 and a half months before midterms? What do we need to know about that midterm date, and what needs to be solved or unsolved before we get there?
We all know that Trump’s approval ratings are pretty bad right now.
Yeah.
There’s the war that he got himself stuck into.
The 72-hour war.
Yeah. The odds aren’t looking good. It looks like the Republicans will probably lose the House. It’s a toss-up for the Senate.
The long end is a reflection of the housing market, and mortgage rates are anchored to that long end. That’s a one-to-one connection to the average consumer.
Got it.
High borrowing costs are not great for the average person. They’re not stoked.
I would map it toward this broad idea of them trying to play Whac-A-Mole as much as they can in the lead-up to it, because they want to make sure they win the Senate, maybe win the House, and they’ll throw whatever they can at it.
Say they gave up on this whole idea of the buyback and intervention regime that we’ve been in for the last month.
Long-end yields are going to start to surge again. Interestingly, I looked at a chart yesterday showing that when the 10-year gets above 5%, the correlation between that and equities flips negative pretty significantly.
Oh.
Above 5%, if we keep going higher, equities and risk assets generally go lower. That flip happens above the 5% level, most concretely.
Say they gave up on this whole thing, and then we see long-end yields really go nuts. One thing to say on that is I think another aspect of why we’ve seen long-end yields go higher is because the Fed is refusing to hike rates. Warsh talks a lot of tough talk but actually refuses to make it happen.
Is part of the reason the long end is ripping because Warsh is sort of like, “Make me hike”?
A little bit. I think he wants to let the long end do some of the tightening in some ways, too. I don’t know. One day I’ll wake up and feel like there’s some concerted plan, like 4D chess going on, and then another day I’ll wake up and be like, “Nobody knows. We’re just—”
Nobody has any clue.
—we’re building the plane as we’re flying it. I don’t know.
Okay.
Maybe some days I’ll think they’re trying to let the long end go a little higher to try to tighten financial conditions, as opposed to focusing on the short end. But then other days I’m just like, I don’t know. It doesn’t feel like there’s a cohesive plan, so it’s hard to say.
It feels to me like they’re incentivized to really try to navigate this tightrope until the midterms, for sure.
Got it.
So that gives us kind of this green light where they’re just going to keep ramping until then. Then, okay, what happens in the midterms? Say we get a divided, split Congress. Then you’re in deadlock.
Say over the next few months, because of this refusal to hike, this liquidity easing, and everything else that’s going on, inflation starts to reaccelerate higher. Suddenly you get to January or February 2027, and you have a split Congress. Inflation is getting so out of hand that long ends start to surge again.
Then it’s like, “Well, I don’t know. This is a split Congress. What can we do? We can’t get ahold of spending. Blame it on the Dems.” It’s almost like—
Yeah.
—they’re going to rip it until then, and then there’s going to be a split Congress, and they can just be like, “I don’t know. We can’t do anything. Congress is split.” Then blame it in that regard.
Damn.
I don’t know. That’s a super rough framing, but I feel like it’s decently likely.
Okay. What do you think from here, for people trading, especially people who are long Bitcoin and gold—
Uh-huh.
—what do we need to watch as far as how this is evolving, announcement-wise and yield-wise?
Mm.
What are the most important charts for us to pay attention to?
Obviously, the yield curve—looking at what the long end is doing, whether 10-year, 20-year, and 30-year yields are going higher or not. I think it’s useful to look at the 10s-2s curve, so the 2s-10s, or the 10-year and 2-year.
Okay.
And whether that’s steepening or not.
Mm-hmm.
And why it’s steepening. The 2s-10s can steepen for two reasons: either the short end, the 2-year, is going lower because we expect cuts, or the long end is going higher, which is called a bear steepening, because of everything that’s been going on.
We’ve been seeing this steepening over recent weeks. Then, obviously, yesterday we got the announcement, and the 2s-10s just collapsed lower and started to flatten. So keep an eye on the 2s-10s and how that’s tracking.
Keep an eye on the MOVE Index, which is the volatility of bonds and what’s going on there. I think that’s always important. Look at expectations for the Fed funds rate, including rate hikes or rate cuts.
Mm-hmm.
Obviously, we had a couple of hikes priced in by the end of this year, and that reached its climax over the last few weeks. Then this started to reverse lower, which will lead to more dovishness on the margins and higher-risk assets.
I think that’s partly why we’ve seen such a big rally over the last couple of weeks. It’s just that re-rating of hike expectations into something more neutral.
Yeah.
That’s super important. Look at the dollar. Like I said, a lot of what they’re doing with the bond market—the exhaust valve is the dollar.
Yeah.
That’s why we saw the dollar go lower yesterday. It’s up a bit today because the market’s, like we talked about, pressing Bessent to see his resolve. Keep an eye on that.
Those are the big ones overall, I’d say.
So are you fully bulled up on BTC and gold? How are you feeling at this stage? Are you going risk-off here and taking some off?
No. Time horizon is always so important when talking about these things.
If you’re a short-term trader, it probably makes a lot of sense to sell a bit of it on these pops. I’m not really interested in doing that. When I look at those two markets and how washed out they’ve been, what’s happening on the macro side of things, and then the idiosyncratic crypto stuff—the four-year cycle, Saylor selling and having a pretty good reserve of US dollars—there’s less risk there.
The DOTs have all sold, been washed out, or closed down. A lot of those are ticking boxes for me. I’ve been slowly buying the two over the last couple of months, and I don’t really care what happens over the next little while. I’m willing to hold those for probably at least 6 months.
That’s just sitting there, and I’m trying not to overthink it. It’s not like I’m 50% gold and 50% Bitcoin. I still have some other stuff, like random equity plays—
Yeah.
—and that sort of thing. I’m not really all that exposed to the AI stuff right now.
Yeah.
I rode the memory trade, and I think the bottom is in there, but I don’t know if we’re going to rip back higher again.
See a bounce, yeah.
I think it’s basically doing what gold did over the last 6 months. Some healing needs to happen, so I’m not really interested in being there.
I’m somewhat overweight gold and Bitcoin, but still looking at other stuff—a few single names and that sort of thing. I’m in it for the long haul.
I don’t think you go through the pain we’ve seen in crypto markets over the last year or so and then just sell the first pop. I’m ready to sit for a while. Maybe I’ll give back half of the gains that just happened, but I think in 6 months it’ll be a lot higher, so whatever.
I was really just trying to deduce that you seem to think this is a really big deal—at least potentially, a really big deal—that could spiral in a crazy way for these debasement trades.
I think so. I mean—
Do they publish buybacks? Will we know publicly how much they have purchased?
Yeah. Whenever they do buybacks, they provide the results. That’s all available for sure. Yeah, no, that’s a big deal. I mean, you know, my boy Andy Constan—he’s a—
I was reading his tweets yesterday. I was GPT-ing his tweets. He was freaking out yesterday.
Dude, he’s the guy who, when these things happen, 9 times out of 10 is like, “Everybody stop freaking out. Not a big deal. Settle down.” A sage man with a gray beard, right?
Yesterday, he was all like—
It’s like the world’s changed.
Yeah. He went long all assets. That’s his kind of max-long signal, and he understands this wonky Treasury issue and stuff better than pretty much anybody I know. I’ve learned a lot from him, honestly.
When he’s talking about it being a big deal, I definitely listen. People I respect are calling it a big deal.
You can get lost in endless fights on FinTwit about the semantics of whether this is yield curve control or not.
Yeah.
It doesn’t matter. We’ve been—
Yeah.
—in a version of yield curve control for quite a few years now. The 10-year should’ve been 30 bps higher than it is pretty much constantly. There is suppression. It doesn’t really matter.
And yeah, people get upset about the political side of it, the hypocrisy, and those are all valid, but it doesn't really change the fact that this is where the cards are, right? We're just playing the hand we're dealt.
All right, I just want to ask you one more question. I'll let you go. Thanks for staying for a while.
Yeah, of course.
This market shift where people aren't interested in the long end at all, and they only want super-short-duration everything—is this just a larger shift where every market is getting more degenerate? Even the least-degenerate markets are getting extraordinarily degenerate. Is that an accurate reflection or representation, or not exactly?
Mm-hmm.
Got it.
Hmm. Maybe a little bit, like retail. I mean, the bond market moves on the margin of big institutional flows that don't really care about that sort of thing.
The biggest driver of bond yields these days, I would say, is honestly hedge fund basis trades and what they do there. I don't know if you've talked about that or looked into it, but the bread and butter of these hedge funds is that they will be a buyer of those off-the-run bonds. They'll buy that in cash and then sell a future on it, and capture that basis.
See.
That has been one of the biggest buyers of the long end. You can make the argument that maybe, with the amount of leverage they use, they're more short-term in nature and it creates more fragility. But I would say the reflection of more T-bill issuance is more just the nature of where this whole debt game has ended up.
Yeah.
I might be wrong on the specifics here, but I think Quinn was telling me this on an episode once a few months ago, so I might not have it perfectly correct. Apparently, in the early 2010s, Brazil got to this eventuality where 100% of their debt issuance was bills. So, zero on the long end.
Zero.
And that is, if you weren't the reserve currency, that's where you start to think about debt defaults and that sort of thing. But I don't really—
Yeah.
—think that's a risk for the US, but that's one direction where the most extreme can go: pretty much all issuance on the short end.
Whoa.
5. Debt Monetization Becomes the Endgame
Another crazy thing, when you think about the sequence of escalation: What is the maximum escalation that we can conceive? Is it the Treasury doing these buybacks where they issue bills and buy the bonds?
But then, okay, think about the issuance. They're selling the bills. Who are they selling the bills to? What if they were selling the bills to the Fed, so not to the market? What if the Fed was buying the bills and then swapping them for central bank reserves?
That is textbook debt monetization. That is full-on money printing. That is the maximum level, where basically they're buying the off-the-run long-end bonds. They're selling the bills, but they're selling the bills to the Fed, so it doesn't really matter—
So, funds buying the long end—
Yeah.
—to keep it out—
The fund buying it—
—to sell more to the Fed. Fuck.
Yeah, that is the maximum level. That is outright debt monetization, to the max degree. Pandora's box is out at that point. So that's the max. I don't know.
Fuck. And that's like Bitcoin goes to $10 million.
Yeah, something stupid. If it's sustained. Obviously, I think in 2020, during the COVID crash, there were a couple of auctions where the Fed was buying directly from the Fed as it was issuing. So there was a little bit of that debt monetization happening.
But if the genie comes out of the bottle, that's the maximum level. I don't know. I'd probably put a 5% probability on that.
Okay. But you would—
But then you have—
If that ever happened, you would just close your eyes and just fucking—
Yeah, yeah.
—hold Bitcoin, oil even.
Yeah. But it's a useful thought exercise to then ratchet back to: What are the levels? I always think in frameworks, so here we are now at this level.
Right.
Bessent is coming out this morning saying we can ratchet to the next level. We know the maximum level, so we can see that there are a few more levels that could potentially happen. I don't know what they look like off the top of my head, but they're possible.
Word.
So yeah, I mean, don't fight the Treasury. There's always, “Don't fight the Fed, don't fight the Treasury.”
Yeah.
I think there's a lot of resolve here, and obviously markets are going to test it. We've seen that today. But I just want to sit and ride the coattails of—
Wow.
—them trying to do this.
Wow. Dude, you're a special talent. I linked a video for your guidance.
Thanks, man.
Who are you going to replace Tyler with?
I don't know yet. Do you have any ideas? I think we'll have some rotating guests for a while that join us.
Mm-hmm.
I find the roundup's always better with three of us, just talking shop and bantering off each other.
I like three, by the way.
Yeah. Otherwise, it's just—
You know who you should get on as a guest? Daniel Tenreiro.
The greatest yen trader of all time.
The laugh was hilarious.
I would actually love that. Yeah.
You should do it.
Yeah. I might just get—I need somebody who's a nice amount of unhinged. That would be really funny.
He is. Have you had Andy Constan on?
Andy who?
Was it Constan?
Oh, Constan. Yeah, he's been on the show a bunch of times.
Okay.
Yeah.
I never listened to the episode.
Yeah. Anyway, we're going to get a rotating crew for a little bit, shop it around, and keep—I don't want to rush into it. But we'll see. Tyler's a special guy. I'm sad that he's behind the veil of compliance now, but he's at a cool new fund, so we've got to respect that.
Fejau, you're the man, dude. Thank you for coming over and giving us some knowledge.
Yeah.
You're on a legendary run.
Anytime. It's nice to see the market be back in a realm I like and I'm decently good at, so it feels good to be back. Thanks for having me.
You're the GOAT, man. Have a good one. Great to chat.
You too.
Peace.
See you, man.
What a legend. Look, I know it's fucking macro schmacro, but you gotta do it.