Quora Answer: What is the Most Overlooked Metric When Evaluating Portfolio Risk across Different Market Regimes?

The following is my answer to a Quora question: “What is the most overlooked metric when evaluating portfolio risk across different market regimes?”

The most overlooked portfolio risk metric is the one that kills you when everything falls simultaneously.  Most investors evaluate portfolio risk through volatility — the standard deviation of returns.  Their adviser shows them a number.  The number looks manageable.  The portfolio looks diversified.  Then a genuine crisis arrives, correlations converge to one, and everything falls together at precisely the moment diversification was supposed to prevent exactly that.  The most overlooked metric in portfolio risk evaluation is correlation instability across market regimes — specifically, the behaviour of asset correlations during stress periods versus normal periods.

Why Standard Volatility Misleads

Volatility measures how much a portfolio’s value fluctuates.  It is a useful metric in normal market conditions.  It is almost entirely useless as a crisis predictor because it is backwards-looking, calculated from historical price data, and assumes that the relationships between assets observed in calm markets will persist in turbulent ones.  They do not.

The 60/40 portfolio — 60% equities, 40% bonds — was constructed on the premise that equities and bonds move inversely.  When equities fall, bonds rise.  The correlation between them is negative.  The bond allocation provides ballast when the equity allocation loses value.  In 2022, both fell simultaneously. The Bloomberg Global Aggregate Bond Index lost approximately 16% — its worst year in recorded history.  The S&P 500 fell 18.1%.  The 60/40 portfolio lost approximately 16% — its worst performance since 2008.  The ballast failed when ballast was needed most.

This was not a black swan event.  It was a regime change — a shift from a deflationary environment where bonds and equities moved inversely to an inflationary environment where rising rates punished both simultaneously.  The correlation assumption that underpinned the portfolio’s construction was regime-specific.  The regime changed.  The portfolio was not designed for the new one.

The Metric: Conditional Correlation

Conditional correlation measures how the relationships between assets change depending on market conditions.  It answers a question that standard correlation ignores entirely: Does the diversification hold when I need it most?  The mathematics are not complex, but the implications are profound.  Research published in the Journal of Financial Economics found that equity correlations across international markets increase substantially during bear markets.  Assets that appear uncorrelated in normal conditions — US equities and European equities, for example — move much more closely together during global stress events.  The 2008 financial crisis, the COVID crash of March 2020, and the 2022 rate shock all produced the same pattern: correlations that were comfortably low in calm markets converged sharply during the crisis.

This phenomenon is sometimes called correlation breakdown — but that framing is backwards.  The correlations are not breaking down.  They are revealing themselves.  The low correlation observed in normal markets was always conditional on those conditions persisting.  The moment conditions changed, the true relationship between assets became visible.  A 2021 study by Kritzman, Page, and Turkington at State Street found that asset class correlations during turbulent periods were, on average, 0.35 higher than during quiet periods.  A portfolio constructed to hold assets with a 0.2 correlation in normal conditions may find those same assets correlating at 0.55 during the drawdown it was designed to withstand.

The Sequence of Returns Problem

Conditional correlation instability is compounded by a second overlooked metric: sequence of returns risk.  This is the risk that the timing of losses, not merely their magnitude, determines the outcome.  Two investors with identical average annual returns over twenty years can end with dramatically different wealth if one experiences large losses early in the sequence and the other experiences them late.  The investor who loses 30% in year one and recovers over years two through twenty ends with materially less wealth than the investor who gains strongly in years one through nineteen and loses 30% in year twenty — even if the average annual return is identical.

This matters because retirement savings, insurance cash values, and long-term investment portfolios are not evaluated on average returns.  They are evaluated on terminal wealth.  The sequence in which returns arrive is as important as the returns themselves — and standard volatility metrics do not capture it.  The IUL’s zero-per-cent floor directly addresses sequence of returns risk.  In a year of severe market decline, the policy credits zero rather than the index's negative return.  The cash value does not decrease.  The compounding base is preserved intact.  The recovery begins from the undamaged previous peak rather than from a depleted base.

Over a twenty-year period containing two significant drawdowns, the difference between a portfolio that absorbs the drawdowns and one that floors at zero is not merely the magnitude of the losses avoided.  It is the entire compounding trajectory from the point of each loss onward.  The portfolio that fell 38% and recovered over seven years compounded from a depleted base for seven years.  The IUL that floored at zero compounded from its previous peak for the same seven years.  The terminal wealth difference is substantial.

The Liquidity Dimension

The third overlooked metric is liquidity-adjusted risk — the recognition that an asset’s risk profile changes materially when the holder needs to sell it in a stressed market.  Listed equities appear liquid in normal conditions. During the COVID crash of March 2020, bid-ask spreads on listed equities widened dramatically. High-yield bond ETFs traded at discounts to their net asset values. Real estate investment trusts fell 40% before recovering. The assets were technically liquid — they could be sold — but the price of liquidity in a stressed market was a haircut that standard risk metrics had not incorporated.

Private credit — as the BlackRock HLEND situation demonstrated — gates redemptions when withdrawal requests exceed 5% of outstanding shares quarterly.  The investor who allocated to private credit for yield without understanding the liquidity mechanics discovered the true risk profile of the asset class, not from a prospectus but from a rejection letter.  Liquidity-adjusted risk asks: what is this asset actually worth when I need to sell it under pressure?  The answer is always less than the calm-market valuation suggests — and the discount is largest precisely when the need to sell is most acute.

What Actually Protects a Portfolio Across Regimes

The instruments that survive regime changes share three characteristics.  They maintain their value independently of market correlation structures.  They provide liquidity on the holder’s terms rather than the market’s terms.  And they protect the compounding base during drawdown periods.

Physical gold maintains its value during equity and bond correlation convergence events because it is not a financial asset in the conventional sense.  It has no counterparty.  Its value does not depend on corporate earnings, interest rate policy, or the solvency of an issuing institution.

The IUL’s floor protection preserves the compounding base during equity drawdowns regardless of correlation behaviour.  It does not outperform in bull markets.  It does not pretend to.  It eliminates the sequence of returns damage that destroys long-term wealth accumulation when crashes arrive at the wrong moment.

Singapore Government Securities provide AAA-rated sovereign exposure in a currency with structural appreciation bias, insulated from the US fiscal deterioration that is gradually eroding the real value of Treasury holdings.

Cash — unfashionable, yield-free, and perpetually underweighted by investors anxious to be fully deployed — provides the one thing no stressed market can destroy: optionality.  The investor who holds 10% to 15% in cash during a crisis does not need to sell assets at distressed prices to meet obligations.  They buy when others are forced to sell.

The Summary

Standard portfolio risk evaluation measures volatility in normal conditions, assumes correlations are stable, ignores the sequence of returns timing, and treats liquidity as binary. Every one of these assumptions fails in the market environments that actually matter.

The most overlooked metric is conditional correlation — the behaviour of asset relationships during stress — because it reveals that most diversification is regime-specific rather than structural.  The portfolio that looks diversified in calm markets often is not diversified in the markets that require it.  Addressing this requires instruments whose risk profiles do not depend on correlation stability: floor-protected accumulation vehicles, genuinely uncorrelated assets, and liquidity reserves that provide optionality when everything else is falling simultaneously.  The investor who evaluates portfolio risk only in calm conditions is buying an umbrella that closes in the rain.

Terence Nunis | Executive Chairman, Equinox Zenith & Red Sycamore | Author, The 1% Playbook: The Billionaire Cheat Code