The “Sell USA” Trade Is Real.

The World is Backing Away

Every month the Treasury Department releases a dataset that almost nobody reads. It doesn’t generate cable news segments. It doesn’t have a memorable acronym that fits on a chyron. It’s called the Treasury International Capital report—TIC data—and it is, in my view, one of the most honest documents the federal government produces. No spin. Just money moving across borders, recorded in black and white.

That’s not to say it’s without controversy. We’ve reported on how there is a massive discrepancy between the Federal Reserve and the Treasury as to the providence of certain dollars. But we’ll save that for another day because it doesn’t alter the top line data. It’s important, just not for this analysis.

I report on TIC data regularly, usually as a chart of the week or folded into a broader economic piece. This time I’m breaking it out on its own, because the March numbers tell a story that deserves its own space. It’s important to note these figures cover March 2026 and were just released in mid-May so there’s a lag. We’re always six to seven weeks behind. That said, March is notable because it captures the first leg of what is now becoming a protracted oil shock—the early weeks of the Strait of Hormuz closure and the geopolitical repricing that followed. The market’s nervous system was already firing. This data is the readout.

What TIC Data Actually Tells Us

The TIC report tracks cross-border flows of money into and out of U.S. financial assets—Treasuries, agency bonds, corporate bonds, equities, and short-term instruments. The most important distinction in the report is between two categories of foreign buyer: private investors and official institutions.

Private foreign investors are hedge funds, asset managers, pension funds, sovereign wealth funds acting in a commercial capacity, and retail investors abroad. They move on yield, risk appetite, and opportunity. They’re responsive. They’re also flighty—when conditions shift, they can exit fast.

Official institutions are central banks and government entities acting in their sovereign capacity. These are the patient holders. When a central bank buys Treasuries, it’s not trading—it’s managing reserves, signaling confidence in the dollar system, and absorbing U.S. debt as a long-term geopolitical and economic commitment. When they sell, the signal cuts in the opposite direction.

The March data shows a split that should command everyone’s attention. Total net TIC inflows came in at $150.7 billion, which sounds reassuring until you open the hood. Private foreign investors were responsible for $162.1 billion in net inflows. Foreign official institutions—the central banks and sovereign entities—posted net outflows of $11.4 billion. The “patient” money is leaving. The nimble money is replacing it.

The March Numbers

Dig into the line items and it gets more specific. Foreign official institutions sold $37.9 billion in long-term Treasury bonds and notes in March. That followed $25.3 billion in net sales in February. They also dumped $12.2 billion in short-term Treasury bills. Normally, when sovereigns trim long-duration holdings, they rotate into bills temporarily—reducing interest rate risk while staying in dollars. When they sell both ends of the curve simultaneously, it suggests something different: not a duration adjustment, but a dollar reduction. They’re not repositioning within the U.S. system. They’re stepping back from it.

The rolling twelve-month picture for official Treasury bond holdings is -$39.3 billion, so this is the continuation of a structural trend that has been building for years—accelerated now by a geopolitical environment in which holding large dollar reserves is increasingly a liability rather than a purely strategic asset.

The country-level data, which lags one additional month, fills in the picture. China and Hong Kong combined have shed more than a third of their Treasury holdings over the past decade—the twelve-month trend is -$96 billion despite a modest February uptick. The Euro Area has hit a record $2 trillion, but that figure is concentrated in financial centers: Belgium, Luxembourg, Ireland, France. These are not sovereign strategic commitments. They are custodial and institutional positions that move with market conditions, and two of them are considered tax havens so there’s little opacity.

The TIC data doesn’t exist in a vacuum. Layer it against the recent Treasury auction results and a coherent picture emerges.

This past week the Treasury sold $25 billion in 30-Year bonds at a 5% yield—the first time since 2007 that a 30-Year carried that rate. That followed weak auctions across the 2, 5, and 7-Year notes in March, and tepid demand for the 3 and 10-Year offerings last week. The contrast with February is striking: in mid-February, just before the Iran conflict escalated, a 30-Year Treasury auction saw the highest demand ever recorded in its history. Six weeks later, the market is demanding 5% to absorb the same instrument. That is a dramatic repricing in a very short window.

The proximate cause is inflation—back-to-back hot CPI and PPI prints. But the structural cause is what the TIC data illuminates: the traditional sovereign buyer is stepping back, leaving primary dealers and private foreign capital to absorb a debt load that only grows. When the marginal buyer is more price-sensitive and the supply keeps expanding, you get exactly what we’re seeing—yields rising to attract the capital that used to show up regardless.

What It All Means to Us

The bond market meltdown is not an abstraction and should never be treated as such. Yields are rising on both ends of the curve simultaneously—the short end reflecting inflation expectations and Fed policy uncertainty, the long end reflecting deficit financing risk and the erosion of sovereign demand. When both ends move up together, it is called a “bear steepener” and it is one of the most hostile configurations for a government carrying nearly $39 trillion in debt.

Here’s how the math plays out. Every 100 basis points of additional yield, or 1%, adds roughly $390 billion in annual interest costs once existing debt rolls over at higher rates. The 10-Year has moved from roughly 4% at the start of the year to above 4.5% last week, and we recently reported that RBC has flagged 5% as the threshold at which equity multiples historically compress. In layman’s terms, that’s the point where bond yields theoretically break the stock market.

Because Treasury yields are the benchmark for everything—mortgages, auto loans, corporate borrowing—this is not a problem contained to markets. The average APR on a new credit card offer is 23.75%. U.S. household debt hit an all-time record of $18.8 trillion in Q1 2026. Credit card balances stand at $1.25 trillion as we’ll cover in the chart of the week from Stephen Semler. So the consumer was already stretched before this yield surge.

This is where the term fiscal dominance enters the frame. The short version: when the government’s debt is so large that the Fed’s rate decisions are effectively overridden by bond market dynamics, monetary policy loses its traditional power. The Fed doesn’t set the long end of the curve. The bond market does. And the bond market is pricing a government that cannot stop borrowing, in an inflationary environment, with its sovereign creditors quietly stepping back.

Here is where I want to be fair to the data. The U.S. remains a preferred destination for private foreign capital in equities and corporate bonds. Private foreign investors put $67.3 billion into U.S. corporate bonds in March and the twelve-month flow into U.S. equities from foreign private investors remains strong. The world still regards American corporations as a relatively safe haven for productive capital. That is a meaningful distinction—it is not a full retreat, and it matters. Until it doesn’t, and investors either look elsewhere because we’ve overextended the AI trade that’s supporting the stock market right now, or everyone just moves to cash because there’s nowhere left to hide.

But that’s not all.

Note what’s happening on the other side of the ledger. U.S. residents increased their holdings of foreign securities by $15.2 billion in March, primarily foreign bonds. So domestic capital is already searching abroad for returns.

For the investor class, this is a genuine dilemma: bonds aren’t safe, cash erodes to inflation, equities are priced for a future the bond market is actively disputing. The AI-driven melt-up may continue a while longer—BCA Research is projecting a potential further 30% rally before a 2000-style correction. But when institutional money retrenches, the effects cascade well beyond portfolios. Credit tightens. Business investment slows. Hiring freezes. The consumer, already carrying record debt at 21% APR with real wages below inflation, has very little cushion left.

That’s why TIC data is a gut check and should be reported on more often. It doesn’t tell us exactly when or how this resolves. What it tells us is that the world is watching, the patient money is moving, and the numbers don’t lie.

Is There a Limit?

It’s the ultimate triple whammy. Wages are decreasing against inflation and forcing more Americans to charge life on the credit card. And the rates being charged have never been more exorbitant. Less money, more expensive items, higher cost of debt. This is the insatiable appetite of the capitalist doom loop when the economic systems that are supposed to keep some semblance of balance go haywire.

Source: Polygraph

I like following Stephen Semler’s work because he’s another primary data digger. Semler doesn’t just rip charts from Fed, he digs into the notes and puts things in context.

For example: “CPI has several shortcomings, one of which is the exclusion of debt. Debt is experienced as a rapidly increasing cost at the household level but isn’t incorporated into the price index the way other household costs are. This further strains the link between real wages and real economic security—the ability to reliably make ends meet.”

As if stripping out food and energy from “core inflation”, the metric around which policy decisions are built in this country isn’t bad enough. Factor in that the cost of debt service for the average household is “off book” so to speak and there really isn’t an inflation and purchasing power gauge that truly measures the depth of the American household financial crisis.