When Bonds Stop Behaving Like Bonds

A “core” borrower, repriced

Last week, France’s 10-year government bond yield rose above 5% for the first time since 2002. The extra yield investors demand to hold French debt instead of German debt reached its widest level since the 2011 eurozone crisis. For a brief moment, France paid more to borrow for ten years than the United States did.

Perhaps most striking: markets now price France as a riskier borrower than Italy. In 2011, Italy sat near the center of Europe’s debt crisis and France was considered part of the region’s stable core.

None of this was driven by a central bank. It is the bond market asking for more compensation to lend to a large, wealthy government running persistent deficits amid political gridlock. That shift matters well beyond France, including for how diversified portfolios here at home are likely to behave.

It isn’t just France

France is the most visible case of a broader trend. Long-term borrowing costs have been rising across developed markets: 30-year U.K. government yields have approached 6%, and the U.S. 10-year Treasury has traded near 5.2% this month.

Much of this increase reflects the term premium: the extra yield lenders require to tie up their money for many years with a government whose finances they are less certain about. Research from several major asset managers attributes the rising term premium to persistent government deficits and lingering inflation uncertainty.

In other words, bond investors are no longer simply asking, “Where will the Federal Reserve set rates?” They are also asking, “How confident am I in this borrower over the next decade?”

Why the hedge depends on the shock

Whether bonds protect a portfolio depends on what kind of shock hits the market.

The relationship between stocks and bonds is not a fixed law. It reflects what the market is most concerned about at the time.

What the data shows

We have seen this before, and recently. In 2022, the S&P 500 fell about 18% and the 10-year Treasury lost a similar amount, one of only a handful of years since 1928 when both declined together.

The pattern has returned in 2026. One analysis finds that daily stock and bond returns moved in the same direction this year (a correlation of +0.42 through early August), compared with an opposite relationship (−0.36) from 2016 to 2021. State Street Global Advisors places the three-year global stock-bond correlation in the 89th percentile of the past 35 years. During the bond selloff in early October, stocks and bonds again fell on the same days.

From the 1970s through the 1990s, stock and bond returns usually moved in the same direction. For roughly two decades after that, bonds offset stocks. This year, the relationship is back above zero.

An important nuance

The picture is not all negative for U.S. bonds. When stress in France spiked this month, investors moved money into U.S. Treasuries and German Bunds, pushing Treasury yields lower for a time. Treasuries still acted as a safe haven during someone else’s fiscal problem.

The harder question is whether they would play the same role if the fiscal concern originated here. We don’t think anyone can answer that with confidence. But it is the right question to ask of any portfolio that relies on bonds to cushion a stock market decline.

Having been there before

For roughly two decades before 2022, bonds reliably rose when stocks fell. Many of today’s portfolio models, risk tools, and rules of thumb were built during that period, and they quietly assume it will continue.

I began my career on a bond desk during the dot-com era, around the time that negative relationship took hold. By 2022, I was Chief Investment Officer at State Street Bank, watching it break. One of the clearest lessons from that year was how many experienced decision-makers had never worked through a period when stocks and bonds fell together, and how far the effects of rising rates reached across a balance sheet.

That experience doesn’t let us predict what comes next. It does make us skeptical of any assumption that has only been tested in one kind of market.

What this means for how we build portfolios

Bonds are not broken. At today’s yields, high-quality bonds pay meaningful income for the first time in many years, and that income has real value in a portfolio. What has changed is the set of jobs we can count on bonds to do.

In practice, that leads us to ask a few questions of every portfolio we manage:

The relationship between stocks and bonds has shifted before and will likely shift again. If growth becomes the dominant concern, bonds could quickly return to their traditional role. Our aim is not to bet on one outcome, but to build portfolios that don’t depend on any single assumption holding true.