From 2000 to 2021, when the S&P 500 had a month down more than 3%, the 10-year Treasury yield fell, and bond prices rose, in 62% of cases. Since January 2022 that has happened in 1 of 11 cases. The rolling three-year correlation between stocks and Treasuries, negative for two decades, has been positive since 2022 and stood at +0.38 in October 2026. If your portfolio was built on the assumption that bonds cushion stock losses, the assumption needs checking. For the basics of setting a mix, see strategic vs tactical asset allocation.
What is diversification supposed to do?
Diversification is supposed to reduce a portfolio’s swings by combining assets whose returns don’t move in lockstep, so a loss in one is partly offset by gains or smaller losses in others.
The math depends on correlation. Two assets with the same volatility and a correlation of zero, held half and half, have about 29% less volatility than either alone. At a correlation of +0.5 the reduction is about 13%. At +1 it is zero. That is why the correlation of your holdings in bad markets matters far more than their correlation on average: an average of zero made up of +0.8 in crashes and −0.4 in rallies doesn’t protect you when it counts.
Why does diversification fail in a crisis?
Diversification fails in a crisis because investors sell what they can, not what they want to, and correlations between risky assets jump. Assets that looked independent turn out to share the same buyers, the same leverage and the same fear.
- Forced selling. Margin calls and redemptions make funds sell their most liquid holdings, which spreads losses into assets with no fundamental link to the original problem.
- A common factor. In a panic, the market factor dominates. US large caps, small caps, international stocks and credit all trade as one bet on risk appetite.
- Volatility targeting. Risk-parity and volatility-targeted strategies cut exposure across asset classes when volatility rises, selling everything at once.
- Concentration. A portfolio that looks diversified across 500 stocks can still be heavily exposed to a handful of mega-cap tech names that make up a large share of the index.
These effects are why the S&P 500 has fat tails: the worst days cluster, and they hit most of the market at once.
Why did bonds stop hedging stocks?
Bonds stopped hedging stocks because inflation came back. When the main risk to markets is weak growth, bad news pushes stocks down and bonds up. When the main risk is inflation and rising rates, bad news pushes both down.
The chart covers three regimes. In the 1990s, with inflation still falling from high levels, the correlation was mildly positive. From 2001 to 2021, a long period of low and stable inflation, it was negative, reaching about −0.70 in 2012, and bonds were the best diversifier available. In 2022, with core inflation above 5% and the Fed raising rates by 4.25 percentage points in a year, it turned sharply positive. Our stock-bond correlation guide covers the mechanism and what to watch.
How did bonds behave in the months that mattered?
In the months that mattered most, bonds went from a reliable hedge to an unreliable one. The full-period correlation understates the change; the bad months tell the story.
From 2000 to 2021 there were 55 months in which the Nasdaq Composite fell more than 3%; Treasuries rallied in 69% of them. Since 2022 there have been 14, and Treasuries rallied in only 21%. The S&P 500 shows the same pattern more sharply: 1 rally in 11 bad months. In 2022 the S&P 500 fell 19.4%, the 10-year Treasury lost an estimated 16.4%, and a 60/40 portfolio lost about 18%, about as much as the stock market itself.
Many portfolios still size bonds as if the correlation were −0.3. If it is +0.3, a 40% bond allocation adds far less protection than the risk model says, and the portfolio is riskier than it looks.
What still diversifies a stock portfolio?
Nothing works in every crisis, but explicit hedges and cash work in the most kinds. The trick is to know which diversifier covers which shock.
| Diversifier | Helps most in | Weak spot | Running cost |
|---|---|---|---|
| Long Treasuries | Recessions, panics | Inflation and rate shocks | Pays yield (5.31% on the 10-year) |
| T-bills / cash | Every selloff | No upside, re-entry timing | Pays yield (4.22% on 3-month bills) |
| Trend following | Long bear markets (2008, 2022) | Sudden crashes, choppy markets | Fund fees |
| Index puts | Any decline past the strike | Costs premium every year | About 1.5%–3% a year |
| Gold | Some crises, currency stress | Unreliable in sharp selloffs | None, but no yield |
The upside of 2026 is that bonds pay far more than they did in 2021: 5.31% on the 10-year against 1.52% then. They are a better income asset and still a decent hedge against a recession. They just can’t be the only hedge. For how to combine them with options, see tail risk hedging strategies.
What about concentrated stock positions?
For executives and tech employees, the biggest diversification failure is usually one stock, not one asset class. If a single company is 30% or more of your net worth, its earnings and its sector’s fortunes drive your outcomes more than the stock-bond mix.
A concentrated position often can’t be sold quickly, because of trading windows, lock-ups or a large embedded tax bill. That makes direct hedges, such as collars and protective puts on the stock itself, more useful than adding bonds elsewhere. We compare the choices in hedging vs going to cash.
How do you check whether your diversification will hold up?
- List your exposures by risk factor (equity market, single stock, rates, inflation), not by fund name.
- Check how each holding behaved in 2020 (a growth shock) and 2022 (an inflation shock).
- Recalculate portfolio risk with a positive stock-bond correlation, not the 2000–2021 average.
- Identify any position over 10% of net worth and plan a direct hedge for it.
- Decide which explicit hedge (puts, a put spread or a collar) covers the gap, and budget for it.
- Review the multi-asset mix once a year, and the correlations each quarter.
Frequently asked questions
Why does diversification fail in a market crash?
In a crash, investors sell whatever they can to raise cash or cut risk, so assets that normally move independently fall together. Correlations rise toward one, and the spread of returns that diversification relies on shrinks exactly when losses are largest.
Does diversification still work in 2026?
It still reduces risk over full market cycles, but it protects less in sharp selloffs than many investors assume. The main change since 2022 is that bonds have tended to fall with stocks: the 36-month stock-bond correlation was +0.38 in October 2026.
Is the 60/40 portfolio dead?
No, but its bond half is a weaker hedge than it was from 2000 to 2021. In 2022 a 60/40 mix lost about 18% because stocks and bonds fell together. Higher bond yields mean it now earns more income, but investors who want crash protection need explicit hedges as well.
What does correlations go to one mean?
It is shorthand for the tendency of most risky assets to fall at the same time in a crisis. Correlations rarely reach exactly one, but they rise sharply, so a portfolio of many different stocks, sectors and countries behaves more like a single bet.
What diversifies a stock portfolio when bonds don't?
Cash and T-bills hold their value in any selloff. Trend-following funds have tended to gain in long bear markets, including 2022. Index put options pay contractually in any decline past their strike. None is free, so most investors combine them.
Does owning more stocks improve diversification?
Owning more stocks cuts company-specific risk, but most of the benefit comes from the first 20 to 30 holdings. It does nothing for market risk, which is what drives losses in a crash. A concentrated position in one company adds risk that no amount of index diversification removes.
Key takeaways
- Diversification works on average but weakens in crises, when correlations rise and assets fall together.
- Bonds were a reliable stock hedge from 2000 to 2021; since 2022 Treasuries rallied in only 1 of 11 S&P 500 months down more than 3%.
- The rolling stock-bond correlation turned positive in 2022, peaked near +0.67 and was +0.38 in October 2026.
- Each diversifier covers one kind of crisis: Treasuries growth shocks, trend following slow bear markets, cash everything but with no upside.
- Explicit hedges such as index puts pay by contract in any kind of selloff, at a known cost.
- Concentrated stock is the biggest diversification gap for executives and tech employees; hedge it directly.
Sources and method
- S&P 500 via FRED (SP500) — monthly S&P 500 returns, 2016 to 2026
- Nasdaq Composite via FRED (NASDAQCOM) — monthly stock returns, 1990 to 2026
- 10-Year Treasury Constant Maturity Rate via FRED (DGS10) — monthly yield changes used as the bond return
- Investor.gov, Assessing Your Risk Tolerance — asset allocation and diversification basics
- SEC, Beginners’ Guide to Asset Allocation, Diversification, and Rebalancing — what diversification can and can’t do
Correlations are Proflex calculations from FRED data: monthly index returns (Nasdaq Composite from 1990; S&P 500 from October 2016, the start of FRED’s daily S&P series) against the negative change in the 10-year Treasury yield, over rolling 36-month windows through October 2026. Down-months are calendar months with an index decline of more than 3%. The 2022 Treasury return is estimated from yield changes and an assumed 8.5-year duration. Crisis behavior in the matrix is a qualitative summary.