Gold and the Shadow of Fiscal Dominance

Between Monday, August 17, and Wednesday, August 19, three things happened that belong in the same sentence. The yield on the 30-year US Treasury bond climbed above 5.3 percent, its highest level since 2007. The national debt passed $40 trillion, more than double where it stood a decade ago. And the Treasury Department announced that it would at least double the size of the operations in which it buys back its own long-dated bonds. Over the same month gold rose 13 percent, its third-best month in a quarter of a century. By the textbook, that is backwards. A 19-year high in long-term yields is supposed to be poison for a metal that pays no interest. That gold rallied anyway tells you it was pricing something other than the level of interest rates. It was pricing the arrangement that pays them.

That arrangement has a name economists reach for when a government’s borrowing needs start to shape the price of money itself: fiscal dominance. It describes a world in which debt, rather than the central bank’s inflation mandate, quietly sets the limits of monetary policy. The United States is not living in that world. But in August, for the first time in a long while, markets began to price the possibility that it might be walking toward one — and gold was where that suspicion showed up first.

A bond market asking questions

Start with the numbers, because they have been moving faster than the forecasts. The Congressional Budget Office now expects a federal deficit of $2.1 trillion for fiscal 2026, some $200 billion more than it projected in February. The largest reason is not new spending but lost revenue. In February the Supreme Court ruled that the administration could not impose tariffs under its emergency-powers authority; collections fell, roughly $100 billion of duties have since been refunded, and by July the government was paying out more in tariff refunds than it was collecting. The July deficit was the largest for any single month since March 2021. The debt crossed $40 trillion months earlier than forecasters had expected, some four and a half years after it passed $30 trillion, and the interest on it now runs above a trillion dollars a year.

The bond market has noticed. The 30-year yield first broke above its 2007 high at the end of July, on the day the Federal Reserve held rates steady over three dissents in favor of a hike, and it pushed higher again in mid-August. This is not a uniquely American condition; the World Gold Council observed in its August review that 30-year yields have been rising across economies carrying heavy debts. But the United States issues the world’s reserve asset. When investors demand a 19-year-high yield to lend to it for three decades, the question they are asking is not really about next month’s inflation report. It is about how the bill eventually gets paid.

Two interventions in three weeks

A heavily indebted government has, broadly, three ways out. It can run surpluses. It can grow faster than its debt. Or it can let inflation and suppressed interest rates erode what it owes in real terms. The first two are slow and politically hard. The third is quiet, and it has done a great deal of the work before. That is why markets watch so closely for any sign that a government is beginning to lean on the price of its own borrowing. This summer supplied two.

The first came on July 31, when the US Treasury joined Japan in buying yen to arrest a slide that had taken the currency to its weakest level against the dollar in nearly four decades. It was the first American intervention in support of the yen since June 1998, in the depths of the Asian financial crisis. Officials framed it as help for an ally; Treasury Secretary Scott Bessent said a stable yen mattered to the entire region. The mechanics raised eyebrows. Rather than selling dollars, the New York Fed sold euros from US reserves to fund the purchases, a choice some veteran intervention-watchers called odd, precisely because it invites markets to wonder what else the operation was for. It is not hard to see why investors read more into it. Japan is the largest foreign holder of US Treasuries, and a country defending its currency does so by selling its reserves. Whatever the motive, the effect was to put the US Treasury on the buying side of the yen for the first time in twenty-eight years — and the World Gold Council later named concerns about the intervention among the drivers of August’s gold-fund buying. (We explored how Japan’s policy normalization reaches gold in an earlier piece on the yen carry trade.)

The second came on August 19. Two days after the 30-year yield set a fresh 19-year high, the Treasury announced it would at least double its long-end “liquidity support” buybacks, the operations in which it repurchases older bonds maturing in ten to thirty years, from $2 billion to at least $4 billion per operation, running from September 9 through November 4. The next day Bessent said operations could be larger still, while stressing that the level of yields had not factored into the decision. The market heard it differently. Long-dated yields fell sharply on the announcement, the 30-year dropping nine basis points in a session, before the rally faded within days as investors weighed what it implied. Even the World Gold Council’s monthly review turned to how markets might interpret the buybacks, and the prospect of a cap on yields.

We have been here before

The United States has run this experiment once already, and gold’s part in it is instructive, partly because of what gold was not allowed to do.

In April 1942, to help finance the war, the Federal Reserve committed to holding down the cost of government borrowing: Treasury bills were pegged at three-eighths of one percent, and long-term bond yields were capped at 2.5 percent. The Fed bought whatever it had to in order to hold those lines, and the arrangement outlived the war by nearly six years. When inflation surged after wartime price controls came off, reaching double digits in 1947, bondholders were locked into yields far below it, and the real value of the government’s debt quietly melted. Federal debt held by the public, which had exceeded 100 percent of GDP in 1946, fell steadily as a share of the economy over the following decades. Strong growth did much of that work, but so did years of negative real interest rates, a mechanism the economists Carmen Reinhart and M. Belen Sbrancia would later document as “financial repression.” Only with the Treasury–Fed Accord of March 1951 did the central bank win back control of its own interest-rate policy.

And gold? Americans could not own it. Private holdings had been called in by executive order in 1933, and the dollar price was fixed by law at $35 an ounce. The one asset that might have signaled what financial repression was doing to savers had been removed from the scoreboard. When the scoreboard came back — the gold window closed in 1971 and private ownership was legalized again at the end of 1974 — gold spent a decade repricing years of accumulated monetary adjustment at once, climbing from $35 to roughly $850 by January 1980. That story is told in full in our history of the gold standard and Bretton Woods and its sequel, the free-float era.

None of this makes a buyback program a 1942-style yield peg. It is not one, and the scale is not remotely comparable. The point is that markets remember the shape of the road. A government under fiscal pressure first reassures, then manages, and only in extremity controls. The asset that prices the far end of that road is the one standing outside it.

What gold was pricing in August

Look at who bought gold in August and the fiscal reading becomes harder to dismiss. Global gold ETFs took in $18 billion, the second-largest monthly inflow on record, lifting holdings to a record 4,189 tonnes. European funds had their strongest month ever, and within Europe the standouts were the United Kingdom, with its second-largest monthly inflow on record, and France, with its largest — two countries whose own government-bond markets have spent the past year under visible strain. North American funds posted their third-largest month on record; we look at what brought those investors back in a companion piece. When the World Gold Council listed the drivers of the inflows, it named concerns about the yen intervention, rising Treasury yields and fiscal sustainability, European sovereign-debt worries, and price momentum. Only the last of those is about gold itself.

Nor was anyone buying on the expectation of a rescue from the Federal Reserve. The Fed spent August edging toward a rate hike, not a cut. Three officials had dissented in favor of raising rates at the July meeting, and at Jackson Hole Chair Kevin Warsh said he would be hard pressed to describe financial conditions as restrictive and that, until the Fed was confident inflation was falling, “we have work to do.” Under the old framework — gold as a bet on falling real yields, which we examined in why that relationship broke down — that is the worst possible backdrop. Gold rose 13 percent into it.

There is a tension here worth naming plainly. One arm of the state, the Fed, is leaning toward higher short-term rates to contain inflation. Another, the Treasury, is using buybacks in a way markets read as leaning against higher long-term rates. Officially the two are unrelated: one is monetary policy, the other debt management and market liquidity. But they push in different directions, and a market watching two arms of the same government pull against each other goes looking for an asset that belongs to neither. It is the same instinct that has made gold a beneficiary of turbulence in the Treasury market before.

---

What would make this wrong

A long read owes its readers the other side, and here it is substantial.

The buybacks are small. Four billion dollars per operation is a rounding error against a $40 trillion debt and a Treasury market that trades hundreds of billions of dollars a day, and buybacks change the mix of what the government owes more than the total. The United States borrows in its own currency, runs the deepest bond market in the world, and has a central bank whose new chair has just gone out of his way to call its 2 percent inflation objective “a firm, fixed target.” A 30-year yield of 5.3 percent is high by the standards of the past two decades and unremarkable by the standards of the decades before them.

And the fiscal premium in gold is plainly conditional. From late August, a hawkish speech at Jackson Hole, a strong jobs report, a hot inflation reading and a fresh oil shock sent gold down three weeks running and back below $4,300, as the odds of a September rate hike climbed. When the real-yield channel and the fiscal channel collide over a matter of days, the real yield still tends to win. So the honest version of this argument is not that fiscal dominance has arrived. It is that the market has begun to price the possibility — and that August showed how much money stands ready to move when the possibility looks a little more real.

The scoreboard is back on

For anyone who holds gold, or is weighing it, the practical question is which risk you are insuring against. If you own the metal as a bet on the next Fed meeting, the fiscal story is background noise, and September has already shown how quickly it can be drowned out. If you own it as a hedge against the slow, polite ways heavily indebted governments have historically reduced what they owe — interest rates held below inflation, a central bank under pressure, a currency allowed to erode — then August was not a trade. It was a reminder of the reason.

The last time the United States carried debts this large relative to the size of its economy was in the years after 1945, and the most important difference between then and now is not the size of the numbers. It is that in 1946 gold’s verdict was invisible, because its price was fixed by law. In 2026 the verdict is printed every second of the trading day. What it shows next depends on choices in Washington that no forecaster can make for you, which is exactly why the long record is worth studying before you decide what role gold plays for you. We have set that record out across 768 years of prices, and in the story of gold’s first real all-time high since 1980.