Mortgage hedging returns as 10-year yield tests 4.69%

Mortgage hedging is creeping back into the Treasury market as the 10-year yield tests the upper end of its recent range, reviving a feedback loop that can make a bond selloff feed on itself.

For bond investors, the risk is mechanical rather than mysterious: higher yields make mortgage-backed securities last longer because fewer borrowers refinance, and portfolio managers who suddenly own too much duration often cut it by selling Treasuries or Treasury futures. Selling then pushes yields higher again.

At 4.69 per cent on the 10-year Treasury, the pressure point is no longer theoretical. The selloff created about $40 billion of 10-year Treasury exposure equivalent, according to a Goldman Sachs estimate cited in Bloomberg’s report on the return of mortgage hedging.

Morgan Stanley Investment Management portfolio manager Vishal Khanduja described a force that had been largely dormant.

“It’s a big deal. We didn’t talk about it for a long period of time.”
— Vishal Khanduja, Morgan Stanley Investment Management

Khanduja’s vantage is the trading-desk version: convexity hedging has moved from background plumbing to the screen. Barclays strategist Amrut Nashikkar sees a wider problem, because the mortgage channel is reappearing while investors test the Treasury market’s tolerance for higher real rates, oil-driven inflation risk and a Federal Reserve path that no longer points cleanly toward cuts.

How the loop works

Mortgage-backed securities do not behave like plain Treasury bonds, because their cash flows depend on the homeowner’s option to refinance. Falling mortgage rates shorten MBS duration as borrowers refinance faster; rising rates slow refinancing, extend the securities and leave holders with more duration than they intended.

Typical hedges include selling Treasuries, selling futures or paying fixed in swaps. Those trades do not cause every yield backup, yet they can add fuel once rates have already moved, and Reuters reported in May that the Treasury selloff was being exacerbated as mortgage investors hedged against rising yields.

Negative convexity is the old phrase for it. In practice, a bond investor can become a forced seller into weakness, and the New York Fed described that risk in a 2014 note on convexity event risks in a rising-rate environment, warning that MBS-linked hedging can amplify rate moves when markets are already strained.

Speed matters more than a single yield level. Coupon distribution, volatility, dealer balance-sheet capacity and the share of the MBS universe trading close to par all shape when duration extension becomes urgent. By those measures, the 2026 market has more securities sitting in the danger zone than it did a few years ago.

Why 2026 looks different

Coupon math is the first reason. Bloomberg said more than $2 trillion of mortgage-backed securities now carry coupons of 5 per cent or higher, roughly four times the amount three years ago, and about a third of outstanding mortgage securities trade near par value, where convexity is most sensitive.

Today’s mortgage market therefore differs from the frozen, low-coupon book of the early pandemic era. Millions of borrowers with cheap loans remain locked in and are not refinancing at current mortgage rates, but the higher-coupon universe has grown enough that a yield move can change duration exposure across a much larger block of securities.

Ownership is the second difference. During the pandemic, the Federal Reserve accumulated $2.7 trillion of mortgage securities, helping mute some private-sector hedging pressure. Runoff has shifted more of the marginal response back to investors that manage duration actively, leaving a market once cushioned by a price-insensitive official buyer to absorb flows from managers who cannot ignore risk budgets.

Skeptics still have a case. Current flows are not automatically a replay of 2003 or 2013, when convexity hedging became a defining feature of violent rate moves. The Fed still owns a large MBS portfolio, banks model the risk more carefully, and Treasury stress has several causes beyond mortgages.

Nashikkar’s warning is that investors may be setting the bar for concern too high.

“The extent to which it can destabilize rates now, I think, has been under-appreciated.”
— Amrut Nashikkar, Barclays

Harley Bassman, managing partner at Simplify Asset Management and a longtime observer of mortgage convexity, was also cited in the Bloomberg piece. His presence in the debate is useful because it keeps the analysis grounded in market structure rather than metaphor: the issue is not a monster suddenly loose in Treasuries, but a familiar hedge that becomes more powerful when coupon distribution, volatility and liquidity line up.

For traders, the issue is not whether mortgage hedging explains the entire selloff. It is whether the flow acts as an accelerant: inflation, oil prices or fiscal supply push yields up first, then convexity hedging makes the next leg sharper and less orderly.

Policy risk raises the stakes

Macro pressure is already heavy. CNBC’s Treasury-yield coverage linked recent rate moves to the US-Iran oil shock, with the 10-year yield at 4.4729 per cent in early trading at the start of the week. The Financial Times reported that hopes of a Strait of Hormuz reopening later pulled oil lower and drove the 10-year yield down 0.13 percentage point to 4.81 per cent.

Oil headlines are not mortgage headlines, but they can wake the mortgage book up. Costlier energy feeds inflation anxiety, inflation anxiety lifts term-premium demands, and higher term premium pushes yields toward levels that extend MBS duration. After that, the hedge arrives.

MarketWatch added a separate policy wrinkle, arguing in its analysis of what could force the Fed to hike that a Kevin Warsh-led central bank may need to prepare for tighter policy if inflation fails to cool. Whatever probability traders assign to that path, the bond market has to price the tail, and convexity hedging is most dangerous when tails are already moving day to day.

“Higher yields create a lot of negative feedback dynamics. They raise the cost of funding for the Treasury and worsen an already concerning fiscal outlook.”
— Amrut Nashikkar, Barclays

That fiscal link makes the story broader than mortgage desks. Washington is selling large amounts of debt into a market that is less certain about inflation, less certain about the Fed and less cushioned by central-bank MBS buying. A self-reinforcing hedging flow does not need to be dominant in that setting; it only needs to arrive when liquidity is poor.

What traders should watch

Volatility is the first signal. A slow climb in the 10-year yield gives mortgage investors time to rebalance, while a fast move through a widely watched range forces hedges to catch up. Futures volumes, swap spreads and agency MBS spreads will reveal more about stress than the headline yield alone.

Coupon-level price action is the second signal. If securities closest to par start cheapening together, duration extension is becoming concentrated, and that is where the hedging loop has teeth.

Real-money demand is the third check. Pension funds, insurers and overseas reserve managers can blunt a convexity shock if higher yields attract them quickly. A delayed response lets mortgage hedgers dominate the marginal price for longer than the broader market expects.

A Treasury accident is not inevitable. Still, a technical force many investors stopped watching is relevant again. In a calmer market, mortgage hedging is plumbing; in a selloff testing 4.69 per cent on the 10-year, it can become price action.