NextFin News - The bond market's worst stretch in nearly two decades still has room to run, according to Deirdre Dunn, head of global rates at Citigroup and chair of the Treasury Borrowing Advisory Committee, in an interview aired September 24. Her warning lands as the benchmark 10-year Treasury yield trades around 5.16%, its highest level since 2007, and the 30-year yield pushes to roughly 5.45%, a mark not seen since 2004. The message from one of Wall Street's largest fixed-income trading desks is that investors should not treat this selloff as a tactical overshoot that mean-reverts after the next inflation print.
The Selloff in Numbers: A Multi-Decade Reset, Not a Blip
The scale of the move is what separates this episode from the routine volatility that has defined the post-pandemic bond market. The 10-year Treasury note yield, the reference price for mortgages, corporate debt, and equity discount rates across the global financial system, touched 5.158% intraday on September 24 before settling near 5.11% - levels last seen in the run-up to the global financial crisis. The 30-year bond, which carries the purest exposure to long-duration risk, jumped 5.1 basis points in a single session to 5.45%, its strongest level in 22 years. The two-year yield, more sensitive to Federal Reserve policy, rose toward 4.9%.
This is not an isolated US story. German 10-year Bund yields have reached their highest since 2011, while UK gilt yields have climbed to levels unseen since 2007 for the 10-year and 1998 for the 30-year. Japan's benchmark 10-year government bond yield struck 3% for the first time since 1996, and Australian benchmark yields hit their highest since 2011. A gauge of global sovereign yields has touched its highest since 2007. When every major bond market sells off at once, the driver is not a country-specific quirk - it is a repricing of the global term premium.
The policy backdrop shifted decisively on September 16, when the Federal Reserve raised its benchmark rate by 25 basis points to a range of 3.75% to 4%, the first increase since 2023. The decision was unanimous, 12-0, and Fed Chair Kevin Warsh left no ambiguity about the committee's priorities. "Inflation is too high and has been for too long," Warsh said at his post-meeting news conference. The Fed's own median projection penciled in one additional 25-basis-point hike for 2026, and rate markets have moved to price it in: traders were assigning roughly a 69% probability to another quarter-point increase at the next meeting, up from close to 50% earlier in the week.
The long end has already done much of the tightening for the Fed. The 30-year yield has settled above 5% on 55 days since the start of the year - the most in any year since 2006 - meaning borrowing costs for mortgages and corporate investment have been rising regardless of what policymakers do next.
Why This Time Is Different: Four Drivers, Not One
Most bond selloffs have a single dominant narrative - a hot inflation print, a surprise from the central bank, a supply shock. This one has four, and they reinforce rather than offset each other. That combination is what makes the call that the selloff has further to go credible even after yields have already moved this far.
First, the monetary regime has flipped. For three years, the market operated on the assumption that the Fed's next move was more likely a cut than a hike. That assumption is now dead. The September increase was not a one-off defensive gesture; it was the opening move of a renewed tightening cycle, backed by a policy projection that still sees further hikes ahead. When the direction of travel reverses, duration assets do not reprice once - they reprice in stages, as each new data point confirms the regime change.
Second, an energy shock is feeding back into inflation expectations. Escalating hostilities between the United States and Iran have pushed Brent crude to around $100 a barrel, after US West Texas Intermediate crude surged more than 5% to $90.22 earlier in the month. Energy is the most liquid and immediate channel through which geopolitical risk becomes inflation risk: higher oil prices flow into gasoline, freight, fertilizer, and plastics within weeks, not quarters. Dunn has described oil as the purest real-time signal of the inflation path for market participants, and has noted that clients are "starting to settle more in for the long haul" on the lasting implications of an energy shock.
Third, supply is competing for a finite pool of capital. The US Treasury market, worth roughly $29 trillion, is absorbing record issuance against a backdrop of a federal deficit running above 6% of GDP and gross government debt that has surpassed $40 trillion. At the same time, AI hyperscalers have launched a historic borrowing spree - a combined $75 billion of investment-grade bond issuance from Nvidia, SpaceX, and Amazon in recent weeks - forcing corporate debt to compete directly with sovereign debt for the same fixed-income allocations. When both the government and the world's most valuable companies are tapping the bond market simultaneously, the price of duration rises for everyone.
Fourth, the global anchor is gone. For two decades, Japan's yield-curve control and Europe's asset purchases exported low yields to the rest of the world. With the Bank of Japan allowing its 10-year yield to reach 3% and European central banks no longer acting as buyers of last resort, that anchor has lifted. A global bond market without a low-yield anchor is a bond market where the equilibrium level of yields is structurally higher.
The Term Premium Has Come Back to Stay
Beneath the headline yields, the more important move is in real rates. The 10-year Treasury inflation-protected yield sits at 2.62%, and the 30-year real yield at 3.07%. Positive real yields of this magnitude mean that government bonds now compete with equities on income alone - a condition that did not exist during the zero-interest-rate era. An investor can earn a positive real return without taking equity risk, which lowers the hurdle rate for every asset class and forces a repricing of risk premiums across the board.
The damage to long-duration holders illustrates the magnitude of the repricing. The iShares 20+ Year Treasury Bond ETF (TLT), the most widely held vehicle for long-end exposure, closed at $80.71 on September 15 when the 10-year yield first pierced 5% - down roughly 36% over five years. This is not a cyclical drawdown that a single rate cut repairs. It is the unwinding of a decade-long compression in the term premium, the extra yield investors demand for bearing the risk of holding long-dated debt. Once investors learn that duration carries real risk again, they do not unlearn it quickly.
From her position atop one of Wall Street's largest fixed-income businesses, Dunn sees the behavioral shift firsthand. The question traders are now asking, in her telling, is not "when does the Fed cut?" but "who pays for all this?" - the war, the rebuilding, the deficit. That reframing matters because it moves the debate from the cyclical, the next inflation print, to the structural, the fiscal and geopolitical cost of the next decade.
"Clients are starting to settle more in for the long haul, and the lasting implications of an energy shock," Dunn said in a recent interview, describing how investors at her firm are positioning for a multi-year repricing rather than a temporary spike.
The Counter-Thesis: Why the Selloff Could Be Overdone
The strongest case against the view that the selloff has further to run is straightforward: the bond market may be pricing a hawkish fantasy that the real economy will not support. Some major banks have warned that markets remain too hawkish, arguing that the tightening already delivered through higher yields is doing much of the central bank's work and that another rate hike could tip an economy already under pressure from energy prices into recession. If growth breaks, the Fed's hiking cycle ends abruptly - and bond yields fall faster than they rose.
There is also a valuation argument. Some strategists contend that a 5% yield on the 10-year note acts as a psychological magnet for buyers who had been waiting on the sidelines, and that reaching that threshold could draw in demand from investors unwilling to move further out the curve at lower levels. At 5.16%, that demand should begin to appear, capping how much further yields can climb without a fresh catalyst.
Finally, the recession channel is real. Higher energy prices, rising borrowing costs, and a sliding stock market squeeze businesses and consumers simultaneously. Dunn herself has acknowledged that the pendulum can swing back toward growth concerns, and that fears about artificial intelligence's impact on labor markets could return to the rates market. A sharp deterioration in employment data would force the Fed to reverse course, and the bond market would rally violently in anticipation.
These counter-arguments are serious, but they share a common weakness: they require the economy to break before the bond market is vindicated. The base case is not a soft landing that brings yields back to 4%, nor a hard landing that sends them there. It is a slow grind higher as the market prices a world of persistent deficits, deglobalized supply chains, and elevated defense spending - none of which reverse on their own.
What Would Prove the Selloff Call Wrong
A judgment without a falsifying signal is a wish. The specific signal that would break the "selloff has further to go" thesis is a sustained combination of cooling core inflation and retreating energy prices: if core PCE prints at or below 0.2% month-over-month for two consecutive months while Brent crude falls back below $80 a barrel, the inflation-risk premium embedded in long yields would evaporate, and the Fed's hiking cycle would lose its rationale. A second falsifying signal would be a rapid deterioration in the labor market - nonfarm payrolls contracting for two consecutive months - which would force the Fed to pivot back to cuts regardless of the inflation path.
Who Benefits, Who Is Exposed
The implications split cleanly by time horizon and by asset class. In the short term, the beneficiaries are holders of cash and short-duration instruments: money-market funds, Treasury bills, and floating-rate notes all earn more as yields rise, with the 13-week Treasury bill yield already near 3.91%. Banks' net interest margins get a second wind as the front end of the curve stays elevated.
The exposed are equally clear. Long-duration assets - long-term Treasury bonds, growth and technology equities whose valuations depend on distant cash flows, and commercial real estate refinancing against higher mortgage rates - face continued pressure. The 30-year mortgage rate, which tracks the 10-year yield, is already at its highest level in over a year, a politically sensitive metric ahead of the midterm elections. Homebuyers, corporate treasurers rolling debt, and pension funds with duration-matching obligations all pay more.
Across time horizons, the picture diverges. In the short term, volatility will dominate as traders position around each inflation print and Fed speaker. In the medium term, the path depends on the data: a string of hot prints pushes the 10-year toward 5.5%, while a growth scare pulls it back toward 4.5%. In the long term, the structural forces - deficits, deglobalization, defense spending, and the end of foreign central-bank demand as a reliable buyer - point to a higher equilibrium yield than the one the market lived with from 2009 to 2021.
Scenarios frame the range. The base case - a 10-year yield holding between 5% and 5.5% through year-end, with one more Fed hike priced and delivered - is what current market pricing implies. The upside case for yields, argued publicly by technology investor Dan Niles, who said on September 15 that the 10-year is headed to 6%, a level not seen since the dot-com bubble, would require inflation to re-accelerate alongside continued heavy issuance. The downside case - a recession forcing the Fed to cut - would send the 10-year back toward 4% and restore the value of the duration that has been punished this year.
The bond market is no longer pricing a temporary inflation overshoot. It is pricing a regime in which deficits, deglobalization, and defense spending keep the term premium bid - and that is a story that does not end with one Fed meeting. For investors, the lesson from Dunn's call is that the rally they are waiting for may be the one that never comes, and that the cost of duration is no longer free.
Explore more exclusive insights at nextfin.ai.
