The Treasury Market Is Strong, But Not In Every State of the World

The U.S. Treasury market is often described as the deepest and safest bond market in the world. That description is true, but it can also make the market sound more automatic than it is. Treasuries are safe in the sense that investors do not expect the U.S. government to miss payments. But a safe asset can still become hard to trade.

That distinction matters because the Treasury market is being asked to do more than ever. The federal government is issuing large amounts of debt. Foreign investors remain major holders, but their demand can change with global conditions. The Fed has moved from buying Treasuries to shrinking its balance sheet. And Treasury yields are reference prices for mortgages, corporate debt, bank balance sheets, pensions, and global capital markets.

The usual public debate focuses on the total amount of federal debt. That number is important, but it is not the whole story. The key market question is: if a shock hits, how far do yields have to move before someone is willing to absorb the Treasuries being sold?

That is the question behind my research with Kristy Jansen and Lukas Schmid on Treasury market resilience. We define resilience in a simple way: a market is more resilient when it can absorb a shock with a small movement in yields. It is less resilient when yields must move a lot to clear the market.

That sounds obvious, but it changes the way we should think about Treasury-market stability. Resilience is not one fixed number. It depends on which part of the yield curve is hit, which investors are selling, which investors are buying, and whether the shock is temporary or lasting. A one-week disruption is different from a permanent withdrawal of demand. A sale by one type of foreign investor is different from an identical sale by another.

The first lesson is that the identity of the marginal holder matters. Foreign investors hold a large share of marketable Treasuries, but their importance for resilience is not proportional to the dollar amount they hold. What matters is the shape of their demand: what maturities they hold, how they respond when yields rise, and how their demand changes with U.S. fiscal and macroeconomic conditions.

This point is especially important when people worry about foreign demand. The question is not only whether a large holder such as China or Japan reduces its Treasury portfolio. It is also what kind of demand is being withdrawn and who absorbs the bonds afterward. A country's share of the market is visible. Its demand curve is less visible, but it is often more informative.

The second lesson is that inflation can affect Treasury yields through more than the standard expectations channel. In the textbook story, inflation raises long-term yields because investors expect higher short-term interest rates in the future. That channel is real. But inflation can also change who wants to hold Treasuries. In particular, foreign official investors may reduce demand when inflation rises, shifting more duration risk onto arbitrageurs and other marginal absorbers.

This matters for how we interpret inflation episodes. A rise in inflation is not just news about the future path of the Fed funds rate. It can also be news about the willingness of major investor groups to hold U.S. debt. If the long end of the Treasury market moves sharply, part of that movement may reflect a change in the investor base rather than only a change in expected monetary policy.

The third lesson is that the design of Fed balance sheet policy matters, not just its size. Quantitative tightening is often discussed as a simple reduction in the Fed's Treasury holdings. But markets may care about two different things. One is the level effect: the Fed holds fewer Treasuries, so private investors must hold more. The other is the insurance effect: if the Fed's retreat signals that it will be less willing to support the market in bad states, then investors and arbitrageurs price a different future distribution of yields.

Those two forms of tightening are not the same. A balance-sheet runoff that preserves confidence in state-contingent support is easier for markets to absorb than one that removes both current holdings and future insurance. This is not an argument that the Fed should always intervene. It is an argument that communication and policy-rule design matter.

These lessons point to a more practical way to discuss Treasury-market resilience. We should not ask only whether the market is large. We should ask which investors are absorbing which shocks. We should not ask only how much foreign demand might fall. We should ask which foreign demand might fall and how domestic investors would respond. We should not ask only how many Treasuries the Fed is selling. We should ask whether the Fed is changing the market's expectation of future support.

The Treasury market has already shown moments of strain: the October 2014 flash rally, the September 2019 repo spike, the March 2020 dash for cash, and dislocations around the 2023 regional bank failures. These episodes do not mean Treasuries are unsafe. They mean the market's ability to intermediate trades is not infinite.

The United States benefits enormously from the depth of the Treasury market. That depth lowers borrowing costs and supports the dollar's global role. But resilience is produced by market structure, investor demand, arbitrageur capacity, and policy design.

The safest market in the world is safe partly because people believe in it. It remains liquid because real institutions and balance sheets stand behind that belief. Confidence is powerful. Capacity is what makes confidence durable.