There is no shortage of frightening explanations for a 5% ten-year: runaway inflation, unsustainable deficits, a flood of bond issuance from technology companies. Telis Demos went looking for which one the data actually supports, and the answer is the dull one.
Investors sell longer-term bonds and demand a higher yield partly because they believe shorter-term bonds will be paying them more for a while. To agree to lock money up for longer, they need more payback. Based on available market signals, that appears to be the main factor behind the move of recent weeks — and that, in turn, is determined by what the Fed is expected to do.
What the inflation gauges actually say
This is the part that cuts against the panic. The gap between the yield on a nominal 10-year Treasury and the yield paid over inflation on 10-year Treasury Inflation-Protected Securities — the so-called breakeven rate — is around 2.4%. That is higher than in 2019 or early 2020, when it sat under 2%. But it is well within the range it has traded in since 2021.
It is not a perfect gauge, since it also reflects liquidity differences between the two markets. But it matches other measures. The five-year, five-year forward rate — what investors expect inflation to be over the five years starting five years from now — is also around 2.4%, per Fed data. Goldman Sachs (GS) economists wrote in August that five years of higher inflation following a long period of low inflation has “left long-term household expectations back where they were in the mid-2000s before the last two cycles.”
Inflation expectations, in other words, have not come unmoored. That matters more than almost any headline.
The term premium, which is the deficit story
The extra yield investors demand to hold long bonds rather than short ones, beyond the expected path of rates, is called the term premium. It is the closest thing markets have to a gauge of worry about government issuance and long-term risk.
It has risen considerably in recent years — meaningfully higher than in early 2022, before the Fed began raising rates post-pandemic, on the San Francisco Fed’s estimate.
But it does not explain the latest move. Since the end of July, that measure of the term premium is down slightly, while 10-year yields have risen.
Read that twice, because it inverts the popular story. The most recent leg of the bond selloff is not the market demanding a bigger risk premium for lending to a heavily indebted government. It is the market repricing where the Fed will sit.
Jonathan Hill, head of U.S. inflation market strategy at Barclays (BCS): “The biggest repricing recently has been the market’s understanding of where average policy rates might sit.” A month ago futures priced a 16% chance of three or more quarter-point increases by next September. Now it is roughly 85%.
Which is why it could unwind
Demos lists what would bring long yields down, and none of it requires a miracle.
The single biggest factor pushing the Fed to raise right now is oil prices. Any letup in Middle East tensions would therefore be a release valve for the Treasury market. A couple of increases now, paired with reassuring language, could lower long-term yields by allaying the fear that the central bank is behind the curve. A pullback in AI capital spending would reduce bond supply. An outright recession would bring yields down fast, at a price nobody wants to pay.
His conclusion is the line to remember: despite what deficit hawks have insisted, “it won’t take an outbreak of fiscal responsibility in Washington to bring yields down from their recent highs. Many other things could do it.”
That is not a reason to stampede into Treasurys at 5%, and he says so. It is a reason to hold a bond position you could live with under either outcome — which is the only kind worth owning when the mechanism is this well understood and the timing is not.