The 30-Year Falls to 5.17% (25 August 2026): The $966.8bn Account Behind the Buyback — and Why the Dollar Rose as Yields Fell

Treasury doubled its long-end buyback on 19 August and the relief lasted two sessions; by 21 August the 30-year was back at 5.27%, a basis point from where it started. Then, on Monday 24 August, CNBC reported that the near-$1trn Treasury General Account is considered available to fund the purchases. The operation did not get bigger. The 30-year closed at 5.23% that day and 5.17% on Tuesday — eleven basis points below the pre-announcement close and seventeen below the 19-year high touched the week before. And the dollar, which had fallen against all six of the other majors while yields rose, rose against all six of them while yields fell.

This piece was written on 19 August, arguing that a yield rise concentrated behind the ten-year point is a term premium story rather than a policy story. It was updated on 20 August when Treasury's intervention tested that mechanism, and again on 24 August when the intervention faded and the front end finally moved. This update covers the sessions of 24 and 25 August, in which the same operation was received completely differently because one fact about it changed.

That fact was not the size. It was the cash.

What actually happened: the funding source, not the size

The 19 August announcement was specific about quantity and silent about payment. Treasury said it would increase, "by at least double", the size of liquidity support buyback operations in longer-dated nominal coupon securities — from $2 billion per operation to at least $4 billion, in the 10-to-20-year and 20-to-30-year sectors, effective 9 September and running through 4 November 2026. It did not say where the money would come from, and the market's default assumption was new bill issuance. Treasury Secretary Scott Bessent had described the operation to CNBC as a "Treasury Twist", which pointed the same way.

That assumption is what capped the effect. Buying long bonds with freshly issued bills swaps duration for supply; it does not bring cash to the transaction. As this post put it three sessions later, the operation worked exactly as designed and the design was small.

On Monday 24 August, CNBC's Steve Liesman reported, citing two senior Treasury officials, that a different funding source is on the table: the Treasury General Account, the government's checking account at the Federal Reserve, already funded out of collected tax receipts. The officials would not say how much, if any, would be used, or when an announcement might come. They were clear that it is considered available.

The Daily Treasury Statement puts the closing balance on 24 August 2026 at $966,849 million — $966.8bn, built up from an opening $933.2bn on deposits of $50.4bn against withdrawals of $16.8bn. Under the previous administration the stated aim was to hold roughly $550bn to $600bn. The current Treasury says it sets the account "consistent with Treasury's long-standing cash balance policy", which is a way of saying the number is discretionary.

Where the move landed this time

Official closes from the Treasury's daily par yield curve. The 30 June column is where this summer's repricing began; 18 August is the last close before the buyback news; 24 August is the day of the funding report.

Three readings, and they no longer disagree with each other the way they did last week.

Read the "vs 18 August" column and the buyback finally has a scoreboard. The 20-year is twelve basis points below its pre-announcement close and the 30-year eleven, against zero at the bill and two at the two-year. That is the monotonic-in-maturity shape a duration operation is supposed to produce, and it took six sessions and a funding disclosure to appear.

Read the "21→25 August" column and the move is much flatter — seven basis points at the two-year against ten at the thirty. A buyback story alone would not touch the two-year. Something else was happening too, and the honest answer is that two other things were.

Read the "since 30 June" column and the summer is still intact. The bill is a basis point lower than it was at the end of June and the 30-year twenty-six higher. Ten basis points of relief does not undo a repricing that ran thirty-six.

What else was pushing yields down

Attribution across two sessions is rarely clean, and pretending otherwise would make this post less useful rather than more.

Two other things arrived on 25 August, and each lowers a yield curve on its own. The Conference Board's consumer confidence index fell to 89.4 in August, down 0.8 points and below the 90.2 consensus, with the forward-looking expectations component the weaker half — the print covered in the consumer confidence piece. Softer expected demand lowers the expected policy path, and that reaches the two-year directly. It is very likely most of the seven basis points there.

Second, crude fell hard. Brent settled 3.9% lower at $88.58 and West Texas Intermediate 3.1% lower at $82.36 as the United States moved toward an economic-pressure strategy on Iran rather than a military one — the mechanism traced in the Iran and oil analysis. Energy is the most volatile input into breakeven inflation, and lower breakevens pull nominal yields down across the whole curve without anyone changing their view on real rates at all.

So the defensible claim is narrow, and it is the one the table supports. The shape of the two-day move is broader than a duration operation explains. The level of the long end relative to 18 August is where the funding news shows up: the 20-year and 30-year are the only points on the curve materially below their pre-announcement closes, and they are the two sectors the buyback actually targets.

The growth print that started the week, and the one that ended it

The front end had moved on 21 August because a growth number arrived, and growth is a front-end input in a way that bond supply is not. At 09:45 EDT that day, S&P Global's flash US PMI put the composite output index at 56.0, up from 54.5 in July and the fastest in 52 months, on data collected 12–20 August. Services business activity hit 56.8, a 20-month high, against manufacturing output at 51.9, a 13-month low. Chris Williamson, chief business economist at S&P Global Market Intelligence, wrote that the survey pointed to annualised third-quarter growth approaching 3.0%, up from 1.5% in the second quarter, with the largest signalled payroll increase since January 2025. Input cost inflation, meanwhile, was the slowest since February.

Four sessions later the Conference Board's consumer confidence index went the other way, and the two-year gave back seven of the five basis points it had gained. That sequence is worth holding onto, because it is what a data-dependent front end looks like from close range: a survey month is a signal about direction, not a settled fact, and a curve that reprices on each one is telling you it has no anchor from the central bank to fall back on.

The dollar broke the carry read twice, in opposite directions

The previous version of this piece ended with a specific, falsifiable line: "A move that finally appeared in the two-year and the bill would be a different animal with a different sign for the dollar."

The move appeared on 21 August. The sign did not change. Then the yields reversed, and the sign did not change that time either — it just pointed the other way. Using the European Central Bank's daily reference rates, here are the two episodes side by side.

Six for six, then six for six again the other way. A pure carry model gets both weeks backwards, which is a stronger result than getting one week backwards, because it removes the easiest excuse. One wrong week is noise. Two consecutive wrong weeks in opposite directions means the variable being tested is not the one setting the price.

Two things separate what happened from what the carry channel predicted.

The first is size, and it is the most boring and probably the largest. In both episodes the currency moves were several times the rate moves that were supposed to cause them — five to nine basis points across the curve on the way up, seven to ten on the way down, against 0.18% to 1.32% in major pairs. And the two-year never left its range: it peaked at 4.37% on 23 July and has spent August between roughly 4.15% and 4.25%. Movement inside a range does not reprice an expected path, and an expected path is what carry is a claim on.

The second explains the sign, and it is why the fiscal framing has held up better than the carry framing across both weeks. A long yield can fall for two completely different reasons. It can fall because growth or inflation expectations have weakened, which is bad for the currency. Or it can fall because the market has raised its estimate of the issuer's capacity and willingness to manage its own debt stock — a claim about the balance sheet behind the currency, and good for the bond and the currency at once. The 21–25 August move carries fingerprints of both: consumer confidence supplied the first, the General Account report the second. The dollar's direction says which one the market weighted.

There is also a governance question sitting underneath all of this, and it has not been answered. CNBC has reported that Fed Chair Kevin Warsh has expressed a preference for the open market determining rates, and the July minutes — covered in the FOMC minutes post — showed several policymakers ready to raise. Paul Stanley, managing director and founding advisor at Arca, put the market's reading of that bluntly in CNBC's coverage on 21 August: "It seems as though Warsh wants the market to do the tightening for the Fed, and that's really what is happening with the recent surge in bond yields."

If the bond market is doing the tightening, then a fiscal authority buying back the long end is, in a limited and indirect way, doing the loosening. Those are two arms of the state pulling on the same yield from opposite ends. Nothing about the last two sessions settles what that means; it is simply the thing to keep an eye on, and Friday's keynote is the first scheduled opportunity to hear it addressed.

What a buyback does, and what changes when the cash is already sitting there

A buyback is the borrower repurchasing its own outstanding debt. No money is created, and the total stock of debt does not shrink. Peter Boockvar of One Point BFG Wealth Partners called it "just a rearrangement of the maturity schedule of Treasuries" in CNBC's coverage, and on the original design that was exactly right.

The quantity the market was choking on was never debt in general but duration — long-dated paper that has to be held by a slow-growing pool of pension funds and insurers matching multi-decade liabilities. Swapping a 30-year bond for a three-month bill leaves total debt unchanged while cutting the duration the market must absorb. The term premium is the price of that duration, so it falls. What the funding question changes is the second leg of that swap.

Neither route is quantitative easing, and the comparison gets made loosely enough to be worth one sentence of precision: QE expands a central bank's balance sheet with newly created reserves, while drawing down the General Account moves existing reserves from the Treasury's account at the Fed into commercial bank reserves. Same balance sheet, different composition. That is a liquidity operation, not the central bank financing the government.

The scale objection has not gone away either, and the people who made it were making it about the operation rather than the account. A $4 billion operation cap sits against a single $25 billion 30-year auction. Mohamed El-Erian, per CNBC, described the purchases as "small in both absolute terms and relative to net issuance", and Krishna Guha of Evercore ISI wrote that the operation "changes almost nothing in terms of the fundamentals in particular the unchanged need to finance the tidal wave of hyperscaler debt in addition to very large government deficits." Both remain true of the operation. What the last two sessions repriced is the optionality behind it — and Treasury officials named its cost themselves: running the account down leaves less cash for a debt-ceiling impasse, and rebuilding it later means selling more bonds to do it.

Nor did this summer's pressure come from Washington alone. Japan's Ministry of Finance published a 30-year JGB yield of 4.05% on 17 August with the 10-year at a 30-year high, and the Bank of Canada's long-term benchmark printed its 2026 high at 4.14% the same day — having bottomed on 27 February, the identical day the US 30-year bottomed at 4.64%. Two long bonds on different continents turning on the same date is not something one country's debt management office explains, and one country's checking account will not fix it.

The equity leg

Long-duration equity traded the discount rate in both directions, which is the cleanest sign check in this whole piece. In the week to 21 August, with the term premium at 19-year highs, the Nasdaq fell 2.1% against the Dow's 0.8%. On 25 August, with the long end rallying, the order reversed: the Nasdaq Composite gained 0.66% to 26,151.30 against the Dow's 0.3% rise to 53,577.40, with the S&P 500 up 0.32% to 7,677.28. Chip stocks rallying ahead of a large semiconductor earnings report — the setup covered in the Nvidia Q2 preview — is a competing explanation for Tuesday specifically, and an honest reading leaves it in.

Barclays' checklist, quoted here two weeks ago, now stands differently. Anshul Pradhan's four conditions for a constructive long end were a downside fiscal surprise, slower AI-related issuance, a shift in Treasury's issuance strategy, and a sustained run of soft activity data. The third has now ticked more convincingly than the 19 August announcement alone managed. AI issuance is still accelerating, the deficit is what it was, and the fourth remains contested — a 52-month high in the composite PMI on 21 August against a soft confidence print on 25 August is not a run of anything yet. Currency-by-currency detail sits on the dollar, euro and yen pages, and the method behind the scoring is on the about page.

What is next, and what would actually settle it

Three dates now carry the question, in ascending order of consequence.

July PCE lands on Wednesday 26 August at 08:30 ET, and its composition matters more than its headline — the setup is laid out in the core PCE preview. The August flash PMI said input costs rose at their slowest pace since February; a PCE print that agrees narrows the case for the front end to keep rising, and one that disagrees widens it again. Then Fed Chair Kevin Warsh's keynote on Friday 28 August, the event set out in the Jackson Hole preview — the only thing on the calendar capable of addressing whether the front end is pricing a central bank that intends to act or one content to let the bond market act for it.

And then 9 September, which is the real test and the one nobody can front-run with an opinion. That is the first enlarged operation. Everything priced so far is priced off a report about an intention, and Treasury officials themselves said it was too early to judge the market impact, since nothing has been bought. Two things become observable that day and not before: whether the operations are actually sized above the $4bn minimum, and whether Treasury says anything about where the cash came from. A large operation with no offsetting bill announcement is the confirmation. A minimum-sized operation alongside routine bill issuance means these two sessions priced a headline.

For the currency leg, the diagnostic from the previous version has to be replaced rather than sharpened. Watching the two-year for a carry signal has now failed twice. The better question is whether a US yield move is a balance-sheet move or a demand move, because the dollar has traded the first and ignored the second all summer. A long end falling on credible debt management, with the front end steady, points one way. A whole curve falling because the American consumer is slowing points the other. This week contained both, which is exactly why the dollar's gain was six-for-six in direction and under half a per cent in size.

The takeaway

The buyback bought two sessions when the market thought it was funded by bills. It bought ten basis points at the long end when the market learned there is a $966.8bn account standing behind it — and the operation itself never changed.

That generalises past this trade. An intervention is priced on its escalation capacity, not its announced size. Nobody was disputing that Treasury could buy $4bn of bonds; they were disputing what would happen on the day the long end sold off again. A discretionary cash pile answers that question in a way a fixed operation cap cannot, and it answered it without a single bond being purchased. The 9 September operation is where the answer gets tested against cash rather than intent.

The dollar leg is the part worth carrying forward. Two weeks, two curve moves in opposite directions, and a carry model that got the sign wrong both times. A yield is not a currency signal on its own — the reason behind the yield is. When the long end falls because the borrower looks more competent at managing its own debt, the bond and the currency rally together. When it falls because demand is weakening, they do not.

Before deciding what a yield move means for a currency, find out where on the curve it happened, and then find out why.

Educational macro context only — not investment advice.