Why the Dollar Fell Despite Doubling Treasury Buybacks

If someone claims they are paying off debt by taking out new loans, would a creditor feel reassured? In August 2026, when the U.S. Treasury doubled its Treasury buyback program from $2 billion to $4 billion per session, the market’s answer was “no.” The decline in 30-year yields lasted less than a day, and the dollar fell to a three-month low.

Looking at the numbers, the reaction is even more dramatic. Before the August 19 announcement, the 30-year yield was 5.34%, the highest since 2007, and it fell by up to 9bp to 5.18% immediately after the announcement. However, by the next day, it reversed about half of that decline, trading at the 5.24% level. During the same period, the dollar index, which measures the dollar’s value against six major currencies, fell to 98.8, its lowest level since May, while gold and bitcoin rose in response.

It is rare for a currency to weaken following an announcement intended to protect interest rates. Gerald Gan, CIO at Singapore’s Lead Capital, interpreted this as “a signal that a weaker dollar is being tolerated.” I believe this discrepancy can be resolved by looking at which side the Treasury is standing on in the bond market.

Why did the Treasury buyback only suppress rates for a day?

Because the Treasury is a seller of bonds, not a buyer; even if they buy them back, they cannot reduce the total amount of bonds circulating in the market.

Comparing this to Federal Reserve quantitative easing highlights the difference.

The Fed is an external buyer that prints new money to make purchases, so the circulating supply actually disappears, and interest rates structurally decline.

The Treasury is an issuer carrying a chronic deficit. Since it does not have cash set aside to buy back long-term bonds, major investment banks expected that the Treasury would issue more short-term Treasury bills (T-bills) to fund the buybacks. Secretary Bessent himself called this measure a ‘Treasury Twist.’ It merely swaps long-term debt for short-term debt, meaning the total volume of debt remains unchanged.

The scale is also small.

With this expansion, the scheduled buybacks from September 9 to November 4 increased by $14 billion, from $69 billion to $83 billion. Out of a $32 trillion Treasury market, long-term bonds account for $5.5 trillion, and the additional $14 billion represents only 0.25% of that. The market returning to its original price in a single day appears to be the result of calculating this ratio.

There is also a boundary set by the Treasury itself.

Officially, the purpose of the buyback is not to artificially lower interest rates but to support liquidity in the Treasury market. In a sense, it is by design that an intervention that did not intend to be read as a promise to suppress rates failed to do so.

It wasn’t interest rates that wavered after a day of Treasury intervention, but confidence in the currency.

Why did rates rise again in September when the buyback was tripled?

The amount of money actually moving was small compared to the size of the announcement, and the market priced in that difference.

Treasury Secretary Scott Bessent said in a CNBC interview, “Current long-term Treasury yields do not properly reflect the fundamental economic conditions of the U.S.” and suggested the buyback could be increased beyond $4 billion per session. Even as he repeated that “we will increase the size of the buybacks,” bond selling did not stop.

There was also speculation that cash in the Treasury General Account (TGA) could be used for buybacks.

When its balance of $935 billion was mentioned in early September, Secretary Bessent drew a line, stating, “We haven’t bought a single bond.” This meant there was no large-scale purchase using Treasury cash.

Then, on September 9, the Treasury set the maximum purchase limit for the next day’s buyback at $6 billion—three times the limit from a month earlier.

However, after the announcement, the 30-year yield rose to around 5.30% during the day, and the yield on the 10-year Treasury auction that same day hit 4.834%, the highest since August 2007. Instead of rates falling as buybacks increased, the market’s disappointment that the scale failed to meet expectations was priced in first.

U.S. National Debt of $40 Trillion: A Number Heavier Than Treasury Buybacks

The reason the market trusts numbers more than the Treasury Secretary’s words is that those numbers are growing too quickly.

According to a Seoul Shinmun report citing U.S. Treasury data, the total U.S. national debt surpassed $40 trillion for the first time on August 19. It doubled from $19.95 trillion in January 2017 in less than 10 years. Why debt has ballooned this much is covered separately in Five Reasons Why U.S. National Debt Is Growing.

There is little sign of improvement ahead.

The Congressional Budget Office (CBO) 2026–2036 Budget Outlook projects that the federal budget deficit will grow to $3.1 trillion by 2036, or 6.7% of GDP. Even looking only at Treasury bonds held by the private sector, the ratio rises from 101% of GDP (in 2026) to 120% (in 2036), surpassing the record set immediately after World War II. Net interest costs will exceed the defense budget at $1 trillion this year and reach $2.1 trillion by 2036. Unless spending is cut, issuance volume will continue to increase.

Treasury issuance is increasing, but the hands willing to buy are shrinking.

Why do bond vigilantes demand higher rates on U.S. Treasuries?

Because the buyers who would absorb the growing issuance volume have changed.

In the past, when the U.S. issued bonds, central banks of allies and trading partners quietly absorbed them into their foreign exchange reserves.

With geopolitical conflicts and economic bloc formation overlapping, their purchasing power and willingness have both diminished, and the trend of various central banks reducing the dollar’s share of their foreign reserves continues. That void has been filled by private hedge funds that are skeptical of U.S. fiscal policy.

Investors who sell bonds to drive up interest rates when they judge that the government is spending recklessly are called ‘bond vigilantes.’ The premium they demand for taking on long-term debt is the term premium.

The term premium is a value that only decreases when fiscal projections change, so it does not move with a few purchases. This is why the Korea Center for International Finance concluded that the effect of this measure would be short-lived unless the fiscal deficit outlook improves.

The Triffin Dilemma: The Path of Treasury Buybacks to Dollar Weakness

The contradiction pointed out by economist Robert Triffin—the proposition that a reserve currency country erodes its own credit as it supplies money to the world—explains this flow.

For the U.S. to maintain the dollar as a reserve currency, it must continue to release dollars through current account deficits and Treasury issuance. Once the released dollars and debt exceed a manageable threshold, the issuer’s ability to repay its debts begins to be questioned.

At this very stage, if the government directly suppresses interest rates, the market does not view it as an action to protect credit. It is interpreted as a signal that the government will sacrifice currency value to ease the debt burden, leading to the sale of that currency. This is the path through which the August buyback drove down the dollar instead of interest rates. How this dilemma shakes the post-dollar order is discussed in The Dollar After the Triffin Dilemma.

Reserve currency status and debt credibility are on the same scale.

Confidence in a reserve currency comes not from the issuer’s assurances, but from the calculations of market participants.

It is only maintained when the calculation that borrowed money can be repaid holds true, and the Treasury Secretary’s remarks cannot change that equation. The fact that Fitch in 2023 and Moody’s in 2025 successively stripped the U.S. of its top credit rating is the result of the same calculation.

The Impact of Dollar Weakness on Retail Investors and the National Pension Service

For Korean investors calculating assets in won, the issue of dollar confidence translates directly into foreign exchange gains or losses.

For retail investors (Seohak Ants) invested in U.S. stocks, even if big tech stock prices rise, if the dollar weakens rapidly, the won-denominated return is reduced by the exchange rate loss.

One must distinguish whether stock price increases are the result of growing corporate value or an illusion caused by falling currency value.

Since a significant portion of the National Pension Service’s and the Bank of Korea’s foreign reserves are in dollar-denominated Treasuries, the combination of dollar weakness and Treasury valuation losses creates a burden on national-level assets.

It is in the same vein that the price of gold, which was well over $4,000 per ounce immediately after the buyback announcement, jumped to a three-month high, along with bitcoin. When confidence in fiat currency wavers, money moves to scarce assets without an issuing entity, and the market calls this a currency devaluation trade. However, it is easy for judgment to become one-sided when jumping on such a narrative. The trap of that is discussed separately in Why Your Brain Is the Worst Financial Advisor.

Cards to Reverse U.S. Treasury Yields: AI Productivity and Fed YCC

There are two escape routes mentioned by the market, but neither is a card that can be played immediately.

First is the path of lowering the debt ratio by increasing nominal GDP through AI-driven productivity gains. It will take quite some time for hyperscaler capital expenditures to lead to industrial-wide productivity and tax revenue.

Second is Fed intervention. If the Treasury’s stopgap measures run out and long-term rates spike to the 6–7% range, making mortgage defaults and corporate bond market stress a reality, the Fed might take on the burden of inflation and engage in emergency quantitative easing or yield curve control (YCC). However, by that point, it is highly likely that the financial market will have already experienced a major shock.

The results of August and September have shown that the issuer buying back its own bonds cannot change the market’s calculations.

The expanded buyback program continues until November 4, and the Treasury will release its next quarterly debt issuance plan in early November.

Looking at the actual purchase details and the direction of the term premium revealed then, it will become clear whether this intervention was a policy or merely a signal.

Note: This article is general information based on public data available at the time of writing and is not an endorsement of specific investment or financial products. Figures and policies may change later. (Last check: 2026-09-28)