The 30-year Treasury reached 5.33% on Tuesday, its highest since 2007, as investors demanded more compensation to lend to a government with a growing debt burden. Inflation worries and a wave of AI-related corporate borrowing competing for the same capital added to the pressure.
On Wednesday the Treasury announced it would "at least double" its purchases of outstanding 10- to 30-year bonds. This was a mid-quarter change to a schedule published only two weeks earlier, and a departure from the "regular and predictable" approach to debt management that Scott Bessent himself endorsed in a speech last November. It makes him the most interventionist Treasury chief in decades.
The detail matters. These are liquidity support operations, and they target older securities that are no longer actively traded. Investors prefer the most recently issued bond at each maturity, so the ones behind it become stagnant and thinly dealt. Buying those back is the cheapest way to put a bid under the long end, because it takes the least liquid paper out of the market rather than competing for the paper everyone actually wants.
It worked, briefly. The long end rallied, with the 30-year yield falling around 12 basis points to a low near 5.18%. The Bloomberg Dollar Spot Index dropped as much as 0.8% to its weakest since 12th May, falling against every major counterpart.
By Thursday it had all gone. The 30-year rose seven basis points to 5.26%, back to where it stood before the announcement. The 10-year touched 4.71%, just shy of its highest since early 2025. The entire rally was unwound inside a day.
The Dollar did not recover with it. It fell for a second day to a fresh three-month low.
One thing the week has established is where the Treasury's nerve gives out. Bessent moved with the 30-year above 5.30%, and that number is now public information. It functions as support, and it also functions as a target. Markets are rarely content to leave a stated threshold untested.
Why the Currency Kept the Loss
Capping borrowing costs without addressing the fiscal position underneath does not remove pressure from the system. It moves it, and Deutsche Bank has set out the clearest account of where it goes.
If the price of US Treasuries is not permitted to fall in the way market forces would otherwise dictate, the adjustment has to happen somewhere. For the foreign investors holding a large share of that debt, the remaining variable is the exchange rate. Suppress the move in the bond and you get it in the currency instead.
George Saravelos and colleagues described the buyback, alongside encouragement for foreign central banks to use the Federal Reserve's FIMA repo facility for their reserves, as "soft-form financial repression policies aimed at containing the long-end of the US yield curve", and negative for the Dollar on both counts.
That framing matters because it makes Wednesday part of a pattern rather than an isolated technical fix. The FIMA encouragement followed joint US-Japan intervention to support the Yen last month, which reduced the risk of Japanese authorities selling Treasuries to raise dollars. Separate measures, one objective: keep the long end contained.
Citi arrived at the same conclusion by a shorter route. "The main price to pay for lowering rates in such a way is a weaker currency," wrote strategists including Dirk Willer, who expect further falls.
The market took the payment and declined the goods. "The market is not fully buying the narrative that Bessent can credibly keep long-end yields in check," said Howard Du at TD Securities. Analysts at ING likened the plan to "rearranging deckchairs on the Titanic".
The scepticism is well founded. US debt has just passed $40 trillion, the enlarged operations do not begin until 9th September, and Treasury's purchases are small against a market that size. A buyback changes who holds the paper and how easily it trades. It does not change how much of it exists, or how much more is coming.
Robin Brooks at the Brookings Institution warned that the Treasury is "playing with fire", and that capping yields without fixing the imbalance risks turning a debt problem into a currency problem, with Japan as the precedent. Shoki Omori at Deutsche Bank was blunter: "The Treasury can buy back its bonds; it cannot buy back the dollar."
The Only Real Fix
Everything announced this week is a bandage.
The durable solution runs through the deficit, and there is no way around it. While the government continues to borrow heavily every year, the debt compounds, the supply of long-dated paper keeps growing, and the market keeps demanding a higher price to absorb it. A buyback rearranges that pile. It does not shrink it, and it does nothing to slow the rate at which it grows.
Canobi framed the choice facing policymakers exactly this way. The alternative to managing yields is to tackle the country's structural problems, which is the far harder path, leaving the authorities either to manage the yield or to let the currency absorb the strain. This week they chose the currency.
Which is why the market handed the rally back inside a day. Nothing on Wednesday changed the arithmetic. Until the deficit narrows enough for the debt to stop outgrowing the economy, every intervention treats the symptom, and there is a reasonable case that each one costs a little more Dollar than the last.
There is at least a hint of movement on the point that matters. Bessent suggested this week that fiscal consolidation plans may be coming, possibly centred on a task force to cut fraud in government spending. Most observers are sceptical that this makes any material dent in a budget deficit running near 6% of GDP, and nothing has been announced. It is still the only category of measure that would change the argument, which is reason enough to watch for it.
How Broad the Move Was
Every G10 currency advanced against the Dollar on Wednesday. The Swiss Franc and New Zealand Dollar were among the largest gainers and the Yen rose as much as 1%, none with a domestic catalyst of its own. Options flows showed the strongest demand against the Euro and the Pound.
The yield move was equally broad. UK and European rates rose again on Thursday, sitting around the multi-year peaks reached in recent days. Andrew Canobi at Franklin Templeton described a "synchronicity of forces arguing for higher yields, steeper yield curves", with every large developed market facing fiscal pressure and stubborn inflation at once.
This is a long-end problem that happens to be worst in the currency that has just tried to intervene against it.
The Fed's Awkward Position
To fund the buyback, the Treasury has to issue more short-dated bills. That is borrowing rather than money creation. The debt is the same size, it is simply owed shorter. But bills reprice with the funds rate almost immediately, so Washington has just made its own interest bill considerably more sensitive to whatever the Federal Reserve does next.
That leaves Warsh with two uncomfortable roads.
If he tightens, he lifts the front end into a Treasury now funding itself there, works against the stated objective of containing yields, and invites the political pressure that follows.
If he does not tighten despite financial conditions having eased, the reading is that monetary policy is being set around the government's financing needs. That is the definition of fiscal dominance, and it is precisely what term premium exists to price.
One road raises the cost of the debt. The other raises the compensation demanded to hold it. Neither helps the currency.
Markets are not pricing this directly, because there is no instrument for it. The signature is visible all the same. The Dollar took a heavy hit on Wednesday with front-end rates barely moving, and September pricing has hardly shifted through a hawkish set of minutes. The adjustment is landing in the currency rather than in rates, which is what you would expect if the view were that the Fed cannot resolve this cleanly.
None of which makes a rise impossible. Nothing prevents a central bank tightening into fiscal pressure, and a Fed that restores inflation credibility could compress term premium rather than lift it. Hiking is not unambiguously against the Treasury's interests. It only looks that way at the front end.
What July Left Open
Some of the pressure in the long end is self-inflicted, and it dates to the press conference of 29th July.
Warsh reaffirmed the 2% target that day and was emphatic about it, insisting there is no soft target. What unsettled the market was that he questioned the yardstick at the same time. He has long preferred a trimmed mean measure to the Fed's own PCE gauge, which he once dismissed as a "rough swag", and he returned to the theme. Bloomberg reported him hinting at a broader inflation roadmap and stoking investor angst. Fortune wrote that the press conference had shone a spotlight on confusion over how the Fed actually measures inflation.
The target and the measure are not the same commitment. Changing the yardstick can change the answer without anyone touching the target, and an investor pricing thirty years of risk notices the difference.
The second gap was the reaction function. He did offer one in words: "Any central banker, when he or she sees underlying inflation moving higher, he or she is more inclined to tighten policy. That's my reaction function." What he did not offer was an instrument or a trigger. He shortened the policy statement, cut guidance on the path, and told markets to "play the ball, not the referee". MUFG described him channelling an inner Greenspan.
Markets can price a rule. They cannot price a disposition. As Gianluca Benigno put it, a rule that never triggers is indistinguishable from having no rule at all.
The reaction that day is worth recalling, because it is the same trade running again this week. Long yields rose, short yields fell, the Dollar weakened, gold gained and equities declined. MUFG called the move at the 30-year one of the largest selloffs at an FOMC meeting in over a decade, and attributed the bear steepening to one fear in particular: that the Fed would forgo rate rises in favour of balance sheet policy instead.
Which is the awkward part. Warsh left open whether he would use the balance sheet. The Treasury has now effectively used it for him, removing duration from the market and funding it at the short end, which is the operation the Federal Reserve would otherwise have conducted itself.
New Chairs are usually allowed a first stumble and this is not out of the ordinary. It is also repairable. The two questions left hanging in July are precisely the two he can answer on Friday.
Why Jackson Hole Is the Catalyst for the Dollar
Kevin Warsh gives his first Jackson Hole address as Federal Reserve Chair on the morning of Friday 28th August, mid-afternoon UK time. The symposium runs from the 27th to the 29th under the theme Financial Innovation: Implications for Payments and Policy, which is not what currency markets will be listening for.
The speech matters more this week than it did last. If fiscal policy has just demonstrated that it cannot hold the long end, monetary credibility is the only anchor left.
Note what Thursday's move was not. Long yields rose sharply while front-end rates were little changed, and futures still price only a 34.6% chance of a September rate rise, barely moved by minutes showing several officials wanted a rise in July and many more expecting one if inflation does not fall. The market is not repricing the Federal Reserve. It is repricing the fiscal position.
There are three outcomes.
He acknowledges the buyback as an easing of financial conditions. There is a mechanical case that he should. Deutsche Bank compares the operation to the Federal Reserve's operation twist: to fund the removal of duration from the market, the Treasury has to issue more short-dated bills. The net effect is looser financial conditions, which all else equal argues for offsetting tightening from the Fed. Evercore ISI reached the same place, arguing the measure "resembles FX intervention" and "adds risk to the Fed rate path". Acknowledgement would read as hawkish, support the Dollar, and open a Treasury-versus-Fed story that has been dormant all year.
There is a reason to think it is likelier than the market assumes. A Chair who has already made a point of saying the Fed is not constrained by market prices has every incentive to demonstrate that it is not constrained by the Treasury either. Naming the buyback would be the cleanest way to do it.
He ignores it. Saravelos was explicit on this point: "If Chair Warsh does not recognize the buyback as a factor driving an easing of financial conditions, we would take it as an additional dollar negative driver." Silence reads as acquiescence.
He sounds dovish. The outcome genuinely not in the price. That 34.6% describes a single meeting. It says nothing about the term premium, and the term premium is where all the pressure currently sits. A Chair who signals tolerance of above-target inflation validates the exact fear driving the 30-year.
The hawkish case is a rates story. The dovish case is a credibility story. Only one of those has a floor.
What He Needs to Do
Our view is that the first outcome is not merely the best available, it is the necessary one.
A hawkish tone that names the buyback as an easing of financial conditions would do three things at once. It would reassert that the Federal Reserve sets policy independently of the Treasury's financing needs. It would put a floor under the Dollar. And it would take some of the credibility premium back out of the long end, which is the only part of this the Fed can actually influence.
Clarity would do as much work as tone. Which instrument he intends to use, and what conditions would make him use it, is the question July left open and the one the long end has been charging him for since.
The dovish alternative is worse than a missed opportunity. It would confirm that the one remaining anchor has been given up, and the long end would sell off sharply on the same day the currency did.
Expect two-way risk either way. Nobody knows what he intends to say. He has a handful of press conferences behind him and no set-piece record at all, and a new Chair's first major address is not usually the polished one. Position for volatility rather than for direction.
None of This Is the Fed's Doing Alone
It would be too neat to lay the long-end selloff at the door of central bank communication.
The hyperscalers are issuing debt at scale to fund the AI buildout, competing for the same investor capital that has to absorb record government supply. The Iran conflict shows no sign of winding down, keeping Brent elevated and inflation expectations alive with it. Fiscal deficits across every major developed market are doing the rest, which is why UK and European yields are sitting at multi-year peaks alongside America's.
Warsh inherited most of this. What Friday determines is whether he adds to it.
What the Dollar Still Has Working for It
The repricing is probably overdone. August is thin, and thin markets move further and more violently than the news in them warrants. Add the volume of commentary around Bessent's decision, the financial repression framing, the currency crisis warnings, the comparisons to Liberation Day, and the price carries a noise premium as well as an information one. That is not to say the concerns are wrong. It is to say that one announcement and two days of late-August trading is a poor instrument for measuring how serious this will be. A market that drifts one way into a scheduled event is also the market that snaps back hardest when the event goes the other.
The macro supports are intact. Brent is around $94 with the Iran conflict unresolved, which is Dollar positive and carries haven demand with it. The Bank of England looks more likely to hold than to move, with Bank Rate at 3.75% and a July vote of six to three in which the dissenters wanted a rise. A Federal Reserve increase remains live at roughly one chance in three, and if inflation does not fall that probability rises and the rate differential starts working for the Dollar again.
There is a reading in which the buyback forces his hand. Ask why a Chair would keep holding rates while naming five years of above-target inflation as the main problem facing the Federal Reserve. One coherent answer is that he hoped not to need to move, because the bond market was tightening for him. Rising long yields lift mortgage costs, corporate borrowing costs and the discount rate applied to every asset, and they do it without a single vote being cast. A Chair who told markets to play the ball rather than the referee, and who insisted the Fed is not constrained by market prices, was describing something close to that arrangement.
Capping the long end removes the tightening he was getting for free. Some of his committee already judge that financial conditions "might not currently be sufficiently restrictive" to return inflation to 2%, on the record in the July minutes. If he agrees, he has to supply that tightening himself, and there is only one instrument left.
Near-term AI support. Masahiko Loo at State Street points to AI-driven inflows into US equities as real Dollar demand. Capital committed to the American buildout has to be converted into Dollars before it can be spent, and while the buildout stays concentrated in the United States that flow continues whatever the Treasury does with its bond schedule.
The case against the Dollar is structural, unresolved and slow. The case for it rests on positioning, on an intact set of macro supports, and on a speech nobody has heard. Which of the two dominates is genuinely unsettled, and Friday is where it gets tested.
What to Watch
Friday 28th August. Warsh at Jackson Hole. Listen for whether he acknowledges the buyback at all, and for anything resembling a trigger.
The 30-year, daily. Whether it takes out Tuesday's 5.33% high is the cleanest read on whether the intervention retains any authority.
The Dollar Index around 98.65 to 98.70. ING's support zone, with 99.00 the level to reclaim before any recovery looks convincing.
9th September. The enlarged buyback operations begin, the first test of the measure in practice rather than as an announcement.
16th September. The Federal Reserve decision.
17th September. The Bank of England decision, with Bank Rate at 3.75% and the July vote having split six to three.
4th November. The buyback programme expires and Treasury reassesses.
Any detail on fiscal consolidation. The only category of announcement that would address the cause rather than the symptom.
Speak to Lamera Capital for in depth analysis.