The Bond Sell-Off: Where Do We Go From Here?

The US 10-year Treasury yield has reached 5%. German Bund yields are at their highest levels since 2009. UK gilt yields have returned to levels last seen before the global financial crisis.

After years in which investors debated how low yields could go, the question has changed completely:

How much higher can they go?

The temptation after a move of this magnitude is to look at the absolute level of yields and conclude that bonds must finally be cheap. Perhaps. But a yield is not cheap simply because we have not seen it for twenty years.

For a rates trader, the more important question is what has taken us here, how much of that adjustment has already occurred, and what the curve, relative value and market reaction are telling us about what comes next.

This is a global sell-off

Oil has provided the latest catalyst. With Brent above $108 today, the market has had to reconsider an inflation problem that central banks had hoped was becoming progressively easier to manage. The US 10-year crossed 5%, while 10-year gilt yields reached around 5.44%.

That has produced a dramatic reassessment of monetary policy. In the US, a Reuters poll published today found 85% of economists expecting the Federal Reserve to raise rates by 25bp this week, a sharp reversal from the earlier hold consensus.

But the important point is that this is not simply a Treasury sell-off. Bunds, gilts and Treasuries are all under pressure. That tells us that something broader than the next Fed, ECB or Bank of England decision is being repriced.

Oil matters, but differently in each market

The inflationary impulse from oil is global. Its economic consequences are not.

In the United States, the question is whether higher energy prices feed into an economy where underlying inflation remains elevated and demand relatively resilient.

In Europe, the euro area is a major energy importer. Higher energy prices therefore produce both an inflation shock and a negative terms-of-trade shock. The ECB potentially has to tighten into weaker growth to prevent a temporary energy shock migrating into wages, services and inflation expectations.

The UK has another combination again: sensitivity to imported inflation, persistent domestic wage and services inflation, and a gilt market already carrying significant fiscal and supply concerns. The same oil shock therefore produces three different monetary-policy problems. That matters when comparing curves.

What exactly has been repriced?

The front end is relatively straightforward. If markets believe central banks need to raise rates further, two- and three-year yields rise. But that does not fully explain what is happening further along the curve.

A 10-year Treasury at 5% does not mean that investors expect the Fed funds rate to average 5% for the next decade. A long-term yield can be thought of, somewhat simplistically, as:

expected future short rates + term premium

The first is monetary policy. The second is the compensation investors require for holding duration: inflation uncertainty, fiscal uncertainty, supply, volatility and the possibility that the bond will subsequently have to be sold at a materially higher yield.

For much of the post-financial-crisis period, that premium was artificially compressed. Central banks were enormous, price-insensitive buyers of government bonds. Today, governments are issuing substantial quantities of debt while central banks are no longer providing the same structural bid. The private sector increasingly has to absorb the duration.

At what yield is it willing to do so?

That is why the current sell-off can continue even after markets become comfortable with the eventual level of policy rates.

The curve should tell us what happens next

The outright 10-year yield gets the headlines. I would spend at least as much time watching the curve.

If two-year yields continue rising faster than tens and thirties, the market is primarily repricing monetary policy. That is a relatively conventional tightening trade: central banks tighten, financial conditions become more restrictive, growth eventually slows and the long end anticipates the consequences.

But suppose the front end stabilises while ten- and thirty-year yields continue rising. That is much more interesting. It tells us that the sell-off is migrating from monetary-policy expectations towards term premium, fiscal concerns and duration supply. And that can be considerably more persistent.

In that environment, being right about the next central-bank meeting may not be enough to be right about the bond market.

Then look at swaps

Government bonds should not be analysed in isolation. If Bund yields rise because EUR swap rates are rising by approximately the same amount, we are principally looking at a rates move. If Bunds begin materially underperforming swaps, the message changes.

The market may be demanding additional compensation specifically for holding government bonds because of supply, balance-sheet constraints or a changing investor base. The same logic applies to Treasuries and gilts.

Outright yields tell us that bonds are selling off. Swap spreads can help tell us why.

If long-end sovereign bonds continue cheapening against swaps even after the front-end policy repricing stabilises, I would take that seriously. It would suggest that the market is not simply repricing central banks. It is repricing government duration itself.

Cross-market relative value matters too

A global sell-off does not mean every market should sell by the same amount. Treasuries combine Fed expectations, US inflation, growth and enormous Treasury supply. Bunds combine ECB policy with a changing German fiscal position and a European market adjusting to substantially more duration. Gilts combine global rates with particularly sensitive domestic inflation and fiscal dynamics. Within Europe, France adds another dimension: fiscal credibility and sovereign risk relative to Germany.

The next phase of this move may therefore become increasingly differentiated. The interesting question gradually changes from “Are global yields going higher?” to “Which market has already priced the new regime, and which has not?” That is where outright macro begins to become relative value.

Where do we go from here? Three scenarios

Rather than attach false precision to a single yield target, I would frame the next move conditionally. The ranges below are a trading map, not a forecast: they describe what would have to happen for the market to move into each regime.

The most important distinction is between the base and bear cases. Another 25–50bp does not necessarily require a new macro regime if oil remains elevated and central banks continue to reprice. A move materially beyond that becomes more significant if it occurs after the front end has stabilised. That would be evidence that term premium itself is breaking higher.

What would make me start buying?

Not simply a yield level. I would want to see a change in the reaction function of the market.

• Oil rises, but yields no longer make new highs.
• Inflation surprises on the upside, but the long end refuses to sell off.
• A hawkish central-bank announcement pushes two-year rates higher, but tens and thirties rally.
• Heavy sovereign issuance arrives and is absorbed without progressively larger concessions.
• Swap spreads stabilise and cross-market volatility begins to fall.

That would tell us something much more important than an arbitrary yield target. It would tell us that the marginal seller of duration is becoming exhausted and that real-money demand is returning.

The trader's view

Multi-year highs are not, by themselves, a reason to buy bonds. But neither is the current sell-off evidence that yields can rise indefinitely. The market is trying to find a new equilibrium after a remarkable period in which central-bank balance sheets distorted the price of duration.

The question is whether today's yields already compensate investors sufficiently for inflation, fiscal supply and volatility, or whether the term premium still needs to rise.

Has the front end finished repricing monetary policy?

Has the long end finished rebuilding term premium?

Is the market willing to absorb sovereign supply without demanding progressively larger concessions?

The answer will not necessarily come from the next inflation print or central-bank statement. It may come from the market itself.

Markets rarely turn because a number looks expensive or cheap. They turn when new information stops producing the expected price reaction.

When the next piece of bearish news fails to push yields materially higher, that will be the first indication that the sell-off is running out of sellers.

Until then, 5% is an important milestone. But it is not necessarily the destination.

This article presents a general professional perspective on fixed-income markets. It does not describe a current portfolio, disclose proprietary positions or constitute investment advice or a recommendation concerning any instrument or strategy. Market levels are a snapshot as at 14 September 2026.