Duration as an Equity Hedge

Ask a portfolio manager in the 2000s or 2010s how to hedge an equity book cheaply and the answer, most of the time, was: buy long-dated government bonds. When stocks fell sharply, investors fleeing risk typically ran toward the safety of government debt, pushing bond prices up and yields down at exactly the moment equities were falling — a hedge that, unlike buying puts, actually paid you (in coupons) to hold it.

Why the relationship isn't fixed

Whether bonds rally or fall alongside a stock selloff depends on why stocks are falling. If the scare is about growth — a recession fear, a credit event, a liquidity crunch — investors want safety and lower expected future rate cuts push bond yields down, so bonds rally exactly as stocks fall: negative correlation, a good hedge. If the scare is about inflation instead — prices rising faster than expected, forcing the central bank to hike more than markets priced in — bond yields rise (bond prices fall) at the same time stocks fall on tighter financial conditions: positive correlation, a hedge that actively hurts you. The two decades before 2022 were dominated by growth-type shocks and low, stable inflation, which is exactly the environment where the negative stock-bond correlation held reliably. 2022 was dominated by an inflation shock, and stocks and bonds sold off together for the first time in a generation.

The scatter above, with a negative correlation dialed in, is the shape a growth-scare regime produces: bond returns and equity returns moving in opposite directions. Push the correlation slider to a positive value instead to see the 2022-style regime, where the same duration position would have added to equity losses rather than offsetting them.

Worked example

A portfolio holds $100 million of equities with an assumed equity beta to a broad market selloff, and overlays a long position in 10-year Treasury futures, notional sized so its duration-adjusted price sensitivity offsets a chosen fraction of the equity risk. Suppose a growth scare pushes equities down 10% and, historically in that regime, the 10-year yield tends to fall by about 60 basis points. With a 10-year duration near 8.5:

$$\Delta P_{bond} \approx -8.5 \times (-0.0060) = 0.051$$

A roughly 5.1% gain on the bond position, which, sized appropriately relative to the equity notional, can offset a meaningful share of the 10% equity drawdown — a hedge that also earns coupon income while waiting for the scare to arrive, unlike a put option that decays.

What this means in practice

Multi-asset risk managers watch the sign of realized and implied stock-bond correlation as a regime indicator in its own right, not just a input to portfolio variance: a persistently positive correlation is a signal that the classic "60/40" playbook and duration overlays need rethinking, and that inflation-linked assets or shorter-duration hedges may do a better job.