Yield Curve Disinversion and the Convexity Trap in Fixed Income

November 17, 2026 – For decades, fixed-income allocators treated yield curve disinversion as a straightforward green light to extend duration. In textbook cycles, a negative 2-year/10-year spread normalizes through a “bull steepening”: growth slows, the Federal Reserve slashes short-term policy rates, front-end yields plunge, and long-duration bonds rally cleanly.

That playbook has broken down in late 2026. Rather than disinverting into an easing cycle, the U.S. Treasury curve is navigating a hawkish, volatility-charged regime in which the Federal Reserve raised its target range by 25 basis points to 3.75%–4.00% in September 2026,1 long-end yields have pushed above 5.10%, and rate volatility has structurally repriced higher.2, 3

For institutional portfolios, insurance general accounts, and private wealth allocators, the central hazard in this environment is not simple duration exposure—it is negative convexity. When yield curves un-invert amid elevated volatility and ongoing central bank balance-sheet runoff, mortgage-backed securities (MBS) and callable credit instruments expose investors to asymmetric extension risk precisely when bond-market hedges are most expensive.4, 5

TLDR: Key Findings at a Glance

  • A Hawkish Disinversion: After plunging to -108 bps in July 2023 and exiting inversion in September 2024,6 the 2s10s spread stood at +29 bps on November 16, 2026,7 compressed from +48 bps at year-end 2025 as the 10-year Treasury climbed to 5.15%.2, 8
  • Volatility Drives the Spread: The ICE BofA MOVE Index closed at 98.79 on November 13, 2026—well above its 56 January low—keeping the value of borrowers’ embedded prepayment options elevated.3, 8, 9
  • Private Balance Sheets Absorb Convexity: Federal Reserve MBS holdings have declined to $1,884 billion,10 and under the October 28, 2026 FOMC directive, agency principal payments are reinvested into Treasury bills rather than mortgages.11
  • Portfolio Playbook: Anchor core duration in the 2-year to 5-year belly and 10-year TIPS (2.91% real yield),2 and pair selective Agency MBS carry with high-quality non-QM and second-lien RMBS to mitigate negative convexity.12, 13

From Prolonged Inversion to a Hawkish Disinversion Regime

To understand why current curve dynamics catch so many fixed-income models off guard, it helps to trace the arc of the past four years. Between March 2022 and July 2023, the Federal Reserve delivered 525 basis points of tightening, driving the 2-year/10-year Treasury spread into the deepest and longest inversion in more than four decades—troughing at -108 bps in July 2023 before finally un-inverting in September 2024.6

By late 2025, the 2s10s slope had widened to +48 bps as markets priced in gradual policy normalization.8 Yet 2026 disrupted that consensus. Persistent inflation above the Fed’s 2% objective—aggravated by energy supply shocks and technology-goods price pressures—pushed the FOMC to hike rates by 25 bps in September 2026 and hold the target range at 3.75%–4.00% in October, while Fidelity notes that September’s median FOMC dot plot still projects another quarter-point increase before year-end.1, 14, 15

Instead of front-end yields collapsing, the entire curve shifted upward and bear-flattened during the third quarter, with the 10-year yield climbing more than 80 bps in Q3 from roughly 4.42% to 5.29%16 and settling at 5.15% in the Federal Reserve’s November 13 H.15 release.2

U.S. Rates & Mortgage Market Pulse (Mid-November 2026)
2s10s Curve Spread (Nov. 16)
+29 bps
Positive slope; compressed from +48 bps at YE 2025
10-Year Treasury Yield (Nov. 13)
5.15%
Up +27 bps on the week; 10Y TIPS real yield at 2.91%
ICE BofA MOVE Index (Nov. 13)
98.79
30-day realized vol at 101.00; Jan. 2026 low was 56
30-Year Fixed Mortgage (Nov. 12)
7.11%
Freddie Mac PMMS; 196 bps above Nov. 13 10Y UST
Sources: Federal Reserve Board H.15 (Nov. 16 release), U.S. Treasury / YCharts, StreetStats, Freddie Mac PMMS.2, 3, 7, 17

The Convexity Trap: Why MBS Behave Like a Short Option

A non-callable U.S. Treasury note exhibits positive convexity: as yields decline, its price rises at an accelerating pace, and as yields climb, its price falls at a decelerating pace. Duration works in the investor’s favor at both tails.

Agency mortgage-backed securities invert that geometry. Because U.S. residential borrowers hold a penalty-free right to prepay their mortgage at par at any time, an MBS investor is effectively long a fixed-rate bond and short an American call option on interest rates.5, 8 Think of it as an asymmetric tollbooth: when mortgage rates drop, homeowners refinance into cheaper loans, handing principal back to the MBS holder precisely when reinvestment yields are lower; when rates rise, borrowers cling to their existing below-market coupons, slowing prepayments and locking the investor into lagging cash flows for longer.8

Beneath the surface, this extension risk creates a dangerous feedback loop during rapid yield repricings. As documented by Federal Reserve Bank of New York researchers, when medium- to long-term Treasury yields jump, MBS effective duration lengthens.4 Active hedgers—including mortgage real estate investment trusts (REITs), mortgage servicers, and dealers—must dynamically shed duration by selling Treasury notes or paying fixed in interest-rate swaps. In extreme episodes such as 1994 and 2003, synchronized convexity hedging amplified the initial rate sell-off into a self-reinforcing “convexity event.”4

Just how fast can the curve shift in the current regime? Over four trading sessions between November 9 and November 13, 2026, yields jumped 21 to 32 basis points across the 2-year to 30-year curve, pushing real 10-year TIPS yields up 25 basis points to 2.91%.2

Treasury Constant Maturity Nov. 9, 2026 Nov. 13, 2026 4-Day Yield Change
3-Month Nominal 4.24% 4.32% +8 bps
2-Year Nominal 4.56% 4.88% +32 bps
5-Year Nominal 4.67% 4.98% +31 bps
10-Year Nominal 4.88% 5.15% +27 bps
30-Year Nominal 5.26% 5.47% +21 bps
10-Year TIPS (Real) 2.66% 2.91% +25 bps
2s10s Curve Slope +32 bps +27 bps -5 bps (Flatter)
Source: Board of Governors of the Federal Reserve System, H.15 Selected Interest Rates (Nov. 16, 2026 release). Red indicates higher yields (bond price headwind).2

Dissecting the Mortgage Spread: Policy Runoff Meets Rate Volatility

Why do 30-year mortgage rates sit at 7.11%17—nearly two full percentage points above the 10-year Treasury—even when corporate credit spreads remain tight? In a May 2026 Federal Reserve Bank of Boston policy brief, economist Paul S. Willen strips out cash-flow timing, Fannie Mae and Freddie Mac guarantee fees (roughly 42 bps), and originator intermediation markups to isolate the pure coupon spread: the compensation investors demand for bearing prepayment risk.8

Willen’s rolling regression demonstrates that three observable variables—the 2s10s yield curve slope, swaption implied volatility, and refinancing intermediation costs—explain roughly 80% of coupon spread variation since 2006.8 Crucially, curve slope and volatility pull in opposite directions. A 100 bps steepening of the 2s10s curve trims the coupon spread by approximately 40 bps because a steeper curve signals higher future rates and fewer refinancings. Conversely, a mere 10 bps rise in swaption implied volatility widens the coupon spread by roughly 15 bps.8

That dynamic explains why mortgage spreads have stayed stubbornly wide in late 2026 despite the curve no longer being inverted. With the 2s10s spread compressed back down to +27 to +29 bps and the ICE BofA MOVE Index hovering near 99 (after peaking at 113.60 on October 5),3, 18 embedded option costs remain elevated. At the same time, the Federal Reserve is steadily stepping back as an unhedged buyer: Fed MBS holdings have fallen to $1,884 billion as of November 11, 2026,10 and under the October 28, 2026 FOMC implementation directive, principal payments from agency securities are rolled into Treasury bills rather than reinvested into MBS.11, 14

Mortgage Spread Regime Shifts: Coupon Spread & Primary Mortgage Basis (bps)
Oct. 2021 Coupon Spread (2s10s +117 bps | Vol 74 bps) – 46 bps
Oct. 2022 Coupon Spread (2s10s -71 bps | Vol 119 bps) – 190 bps
Dec. 2025 Coupon Spread (2s10s +48 bps | Vol 81 bps) – 85 bps
Nov. 2026 Primary 30Y Mortgage vs. 10Y UST Spread (7.11% vs. 5.15%) – 196 bps
Sources: Federal Reserve Bank of Boston (2021–2025 option-adjusted coupon spread); Freddie Mac PMMS & Federal Reserve H.15 (Nov. 2026 primary spread, which includes g-fees and intermediation).2, 8, 17

Portfolio Positioning: Navigating Duration, Carry, and Optionality

In a flat, high-volatility curve regime, passive aggregate bond indexing leaves portfolios overexposed to long-end rate shocks and embedded negative convexity. Institutional allocators can improve risk-adjusted carry across three targeted levers:

1. Anchor Duration in the Belly and Real-Yield Ballast

With 2-year Treasuries yielding 4.88% and 10-year Treasuries at 5.15%, extending out to 30 years (5.47%) compensates investors with only 59 bps of additional yield over the 2-year point for taking on vastly greater duration sensitivity.2 Concentrating core Treasury exposure in the 2-year to 5-year belly captures nearly 90% of nominal curve yield without absorbing long-end fiscal and supply volatility. Where longer duration is required for liability matching, 10-year TIPS at a 2.91% real yield offer explicit protection if supply-side inflation remains sticky.2

2. Be Selective Across Agency MBS Coupons

As PIMCO’s November 2026 credit research notes, Agency MBS has proved markedly more resilient to rate volatility than during the 2023 regional banking stress: mortgage REIT balance sheets are healthier, allowing them to buy into weakness rather than act as forced sellers, and Agency MBS continues to offer positive excess carry over investment-grade corporates without corporate credit risk.12 In addition, a January 2026 government-sponsored enterprise (GSE) directive to purchase $200 billion in Agency MBS has established a soft technical backstop on spreads.13 Even so, elevated MOVE readings compress option-adjusted spreads (OAS), making coupon selection critical: balancing seasoned discount pools (minimal prepayment risk, lower carry) against production-coupon pools (higher nominal carry, higher callability) helps neutralize portfolio convexity.

3. Capture Structural Convexity Advantage in Non-Agency RMBS

Where Agency MBS forces investors to own borrowers’ unpenalized refinancing options, high-quality non-Agency residential MBS (RMBS) offers a cleaner structural profile. According to Guggenheim Investments’ structured credit analysis, U.S. homeowner equity has climbed to 2.3x outstanding mortgage debt—roughly triple the Global Financial Crisis trough—providing substantial loss remoteness across a $750 billion market.13 Investment-grade tranches backed by non-qualified mortgages (non-QM) and closed-end second liens (HELOCs/CES) yield roughly 5.5% with shorter weighted-average lives and meaningfully less negative convexity than Agency pass-throughs, because non-QM borrowers face higher GSE-ineligible refinancing friction and second-lien borrowers keep their low-rate first mortgages intact.13

Key Risk Alert – Extension & Step-Up Traps in Structured Credit: Not all non-Agency RMBS carries benign optionality. As PIMCO warns, earlier-2026 AAA non-QM deals face heightened extension risk: if slower prepayments keep deals outstanding past their call dates, senior coupons step up, squeezing excess spread.12 Favor newer-vintage structures with accelerated senior paydown waterfalls, and avoid deeply subordinated tranches or prime jumbo pools with high callability.12, 13

Optimistic vs. Pessimistic Views: Where Rates and Spreads Go Next

The Optimistic (Bull-Steepening) Case: Sequential monthly inflation cools as energy disruptions stabilize and prior tariff pass-throughs fade, allowing the Fed to pause after 2026 without delivering the additional 2027 hikes priced into parts of the futures market.13, 15 Under that scenario, the ICE BofA MOVE Index retreats from near 99 back toward the low-70s YTD average,9 compressing MBS coupon spreads by 30–50 bps and generating both price appreciation and excess carry for intermediate fixed income.8, 12

The Pessimistic (Convexity-Stress) Case: Core inflation remains anchored above 3%, forcing the FOMC to hike again in December 2026 toward 4.00%–4.25% (with eight FOMC participants already projecting 4.25%–4.50% by end-2027).15 If 10-year Treasury yields break back above October’s 5.31% peak toward 5.50%+, private MBS holders absorbing the Fed’s ongoing runoff could be forced to dump duration into a thin market, reviving convexity-driven volatility across both rates and credit.4, 10

The Bottom Line: Yield curve disinversion in 2026 has not restored the classic bull-steepening playbook. With 2s10s sitting at just +27 to +29 bps and bond volatility elevated near 99 on the MOVE Index, fixed-income investors should resist reaching blindly for long-end duration or unhedged Agency MBS beta. Anchor core risk in the 2-year to 5-year belly and 2.91% real-yield TIPS, demand structural call protection in non-Agency RMBS, and treat negative convexity as a priced risk factor rather than a free lunch.

References

[1] Board of Governors of the Federal Reserve System. (2026, September 16). Federal Reserve issues FOMC statement. federalreserve.gov

[2] Board of Governors of the Federal Reserve System. (2026, November 16). H.15 Selected Interest Rates (Daily). federalreserve.gov

[3] StreetStats. (2026, November 13). VIX S&P 500 Volatility and MOVE Treasury Volatility. streetstats.finance

[4] Malz, A. M., Schaumburg, E., Shimonov, R., & Strzodka, A. (2014). Convexity Event Risks in a Rising Interest Rate Environment. Federal Reserve Bank of New York, Liberty Street Economics. libertystreeteconomics.newyorkfed.org

[5] Raymond James. (2026). Duration & Convexity – Fixed Income Bond Basics. raymondjames.com

[6] First Citizens Group. (2024, December 9). ‘Disinversion’ of the United States Yield Curve. firstcitizensgroup.com

[7] U.S. Department of the Treasury / YCharts. (2026, November 16). 10-2 Year Treasury Yield Spread (Daily). ycharts.com

[8] Willen, P. S. (2026, May 19). Why Mortgage Rates Exceed Treasury Yields. Federal Reserve Bank of Boston Current Policy Perspectives 26-3. bostonfed.org

[9] Columbia Threadneedle Investments. (2026). Navigating the rate repricing: European short-term high yield. columbiathreadneedle.com

[10] StreetStats. (2026, November 11). Fed Balance Sheet & Net Liquidity (H.4.1). streetstats.finance

[11] l0g.fr. (2026, November 11). How to Read H.4.1: the Fed Balance Sheet, Line by Line. l0g.fr

[12] Karoui, L. (2026, November 9). The Credit Market Lens: Where to Seek Diversification From the AI Investment Theme. PIMCO. pimco.com

[13] Guggenheim Investments. (2026, August 13). Non-Agency RMBS: Income, Convexity, and a Rebuilt Market (Third Quarter 2026 Structured Credit Outlook). guggenheiminvestments.com

[14] Board of Governors of the Federal Reserve System. (2026, October 28). Federal Reserve issues FOMC statement. federalreserve.gov

[15] Fidelity Investments. (2026). What’s next for interest rates. fidelity.com

[16] Nationwide Financial. (2026, October 29). Bond market signals: opportunity or risk? nationwide.com

[17] Freddie Mac. (2026, November 12). Primary Mortgage Market Survey (PMMS). freddiemac.com

[18] Chatham Financial. (2026, November 3). Inflation isn’t done, and rate volatility is rising. cf.com

Disclaimer: This article provides financial information for educational purposes only and should not be considered financial, investment, legal, or tax advice. The information presented is based on sources believed to be reliable as of the date of publication, but accuracy and completeness cannot be guaranteed. All investments involve risk, including the possible loss of principal, and past performance is not indicative of future results. Forward-looking statements and projections are inherently uncertain and subject to change without notice. The views expressed are those of New School Financials and do not necessarily reflect the official policy or position of any affiliated financial institution. Consult with a qualified financial advisor before making any investment decisions. Fixed-income securities, including Treasury obligations and mortgage-backed securities (MBS), are subject to interest rate risk, prepayment and extension risk, liquidity risk, and credit risk.

Comments

Popular NSF Posts

Neutral Rate Neutrality: Recalibrating r* in a Post-Quantitative Easing Regime