VIX vs 10Y Treasury Yield

VIXCLS is calculated from S&P 500 SPX options expiring 23 to 37 days out using a model-free strip integration that CBOE published in 2003 (the post-2003 methodology; the pre-2003 series was VXO, based on at-the-money OEX options). DGS10 is the constant-maturity 10-year Treasury yield from the U.S. Treasury H.15 release, which the Treasury constructs from the daily yield curve fit to all outstanding nominal Treasuries. Both are end-of-day series on FRED, both published with one business day lag, and both have intraday tickers (VIX on Cboe, 10Y futures on CME).

The two series capture different parts of the same Fed-policy expectation set. VIX prices the variance of equity returns; DGS10 prices the average nominal short rate over the next decade plus term premium. Adrian, Crump, and Moench (2013, New York Fed) decompose DGS10 into expected average short rate and term premium components; the term-premium component is what links most cleanly to VIX during stress episodes because forced selling in the long end produces both higher yields and higher equity vol simultaneously.

From 1990 through 2021, the rolling 60-day correlation between daily VIX changes and daily DGS10 changes was negative on average, around -0.25 to -0.40 in calm regimes and reaching -0.55 during the September 2008 Lehman window. The mechanism is flight-to-quality: when equity stress accelerates, real-money allocators sell stocks and buy long-duration Treasuries, pushing VIX up and DGS10 down. October 10, 2008 produced the cleanest example: VIX closed at 69.95, DGS10 fell from 3.86% on October 9 to 3.86% on October 10, then to 3.86% on October 14, with substantial intraday flight-to-quality flattening visible on the curve.

The 2022 episode broke the relationship. From January 3, 2022 to October 21, 2022, DGS10 rose from 1.63% to 4.25% while VIX averaged 25.4 (peaking at 36.45 on March 7, 2022 and 33.63 on October 12, 2022). The S&P 500 fell 25.4% peak-to-trough. The 60-day rolling correlation between VIX and DGS10 turned positive for sustained windows for the first time since the 1994 Greenspan tightening. The proximate cause was Fed Funds rising from 0.08% to 3.83% over the same nine months, the fastest tightening cycle since 1980. When rates are themselves the source of equity stress, both legs rise together and the textbook hedge breaks.

The MOVE index, ICE BofA's measure of expected Treasury volatility, is the cleanest cross-check on whether VIX-DGS10 co-movement is driven by genuine cross-asset stress or by rates-only stress. When MOVE rises faster than VIX, the source of stress is concentrated in the rates market (the September 2022 UK gilt episode is the textbook case: MOVE peaked at 159 on October 18, 2022 while VIX held in the high 20s). When VIX rises faster than MOVE, the source is concentrated in equities (October 2018 sold off with VIX at 36 while MOVE held below 90).

Dealer balance sheet capacity has structurally tightened since the 2014 Supplementary Leverage Ratio (SLR) implementation, which the Federal Reserve Board has publicly acknowledged contributes to thin Treasury liquidity during stress. SLR-constrained dealer behavior is what produced the September 2019 repo spike (overnight Secured Overnight Financing Rate jumped to 5.25% from 2.43% on September 17, 2019) and the March 2020 Treasury market dislocation that required Fed intervention via $1.6 trillion of Treasury purchases over three weeks. Both episodes produced VIX spikes alongside disorderly DGS10 moves rather than the textbook flight-to-quality flattening.

Convex Risk Appetite Index (CRAI) blends VIX, ICE BofA HY OAS (BAMLH0A0HYM2), and equity put-call ratios. When CRAI is in its lower quartile, the equity-stress channel is the dominant driver and the VIX-DGS10 relationship is most likely to be inverse: the textbook flight-to-quality regime applies. When CRAI is in its upper quartile and DGS10 is rising, the pair is in the rates-driven equity stress regime that broke the relationship in 2022.

Convex Net Liquidity Impulse (CNLI), based on Federal Reserve H.4.1 weekly data (reserves plus reverse repo plus Treasury General Account), provides the slower regime conditioner. CNLI turned sharply negative in mid-2022 as Quantitative Tightening at $95 billion per month combined with rising Treasury General Account balances drained system reserves. That CNLI contraction is what pulled both VIX higher and DGS10 higher simultaneously, because liquidity withdrawal hits all asset classes at once. The June 2022 to October 2022 window had the deepest CNLI drawdown since the indicator was first computed and produced the most pronounced VIX-DGS10 co-movement of the post-1990 history.

The cleanest example of the joint pair signaling regime change in the same direction across both legs is the April 2020 to February 2021 reflation. Federal Reserve QE running at $120 billion per month plus the CARES Act fiscal impulse drove VIX from 82.69 on March 16, 2020 to 19.40 by August 11, 2020 while DGS10 first compressed to 0.52% on August 4, 2020 then rallied to 1.74% by March 19, 2021. VIX falling and DGS10 rising together is the reflation regime: liquidity provision suppresses equity vol while reflation expectations push the long end higher.

This is the configuration most macro desks watched in 2024 to early 2026 as the disinflation trade matured. DGS10 traded between 3.7% and 4.7% over the trailing 12 months while VIX averaged in the high teens. That joint reading is consistent with the late-cycle steady-state regime: the Fed has finished the cutting cycle, term premium has re-normalized to roughly 50 to 75bp per Adrian-Crump-Moench measures, and equity vol is anchored by buyback flows and the systematic vol-selling complex that dominates short-dated SPX options.

Configuration one: rising VIX, falling DGS10 (textbook risk-off). Reduce equity beta, lengthen Treasury duration, and treat the pair as confirming a flight-to-quality regime. Historical analogs: October 2008, March 2020, December 2018. Median equity drawdown in this regime since 1990 is roughly 12% over 6 to 12 weeks before central-bank intervention or fiscal response.

Configuration two: rising VIX, rising DGS10 (rates-driven equity stress). The hedge is broken. Reduce both equity and duration; cash and short-dated TIPS are the cleanest defensive holds. Historical analogs: 2022 full year, 1994 Greenspan tightening. Configuration three: falling VIX, falling DGS10 (goldilocks easing). Add equity beta and lengthen duration. Historical analogs: April 2019, late 2024. Configuration four: falling VIX, rising DGS10 (reflation). Equity rotation toward cyclicals and value, short long-duration Treasuries. Historical analog: April 2020 to March 2021. The current April 2026 configuration is closest to mid-2024 with VIX low and DGS10 range-bound.