It is 2008. A fund manager runs a $50M equity portfolio. His VaR model says there is a 1% chance of losing more than $2.1M in a single day. He sleeps well. Then the GFC arrives and the portfolio drops $11.4M in one session — a loss 5.4 times larger than VaR predicted. VaR told him where the cliff edge was. It said nothing about the drop.

This is the foundational problem with Value at Risk as a standalone metric. VaR defines a threshold loss at a given confidence level — say, 99% — but completely ignores the magnitude of losses beyond that threshold. It answers "how bad before things get bad?" It does not answer "how bad when things get bad?" For institutional portfolios with non-linear exposures, that silence is dangerous.

CONCEPTCVaR is the average of all losses that exceed the VaR threshold — it quantifies the tail, not just the edge of it.
WARNINGA portfolio optimised solely to minimise VaR can legally contain catastrophic tail risk that never appears in the headline number.
KEY IDEARegulators under Basel III now require CVaR-based stress metrics precisely because VaR failed during every major crisis since 1998.

Conditional Value at Risk — also called Expected Shortfall — fixes this by averaging all losses in the tail beyond the VaR cutoff. If 99% VaR on a $10M book is $400K, but the ten worst scenarios average $1.2M, CVaR is $1.2M. That $800K gap is the exposure VaR erases from the conversation. Portfolio managers sizing positions using only VaR are, structurally, underestimating required capital buffers by that entire margin.

VaR vs CVaR: Where the Gap LivesLoss Magnitude →FrequencyVaR 99%CVaR region(avg tail loss)$1.2M avgNormal range

The practical implementation follows a rules-based structure. First, calculate 99% VaR using historical simulation across at least 500 trading days. Second, isolate all scenarios breaching that threshold — typically the worst 1% of observations. Third, average those losses: that figure is CVaR. Position sizing then uses CVaR as the hard floor. If a single position contributes more than 15% of portfolio CVaR, reduce it until it does not. This is a mechanical rule, not a judgement call. For deeper reading on the mathematics of expected shortfall and its coherent risk properties, the academic literature is extensive. Traders wanting to understand how regulatory frameworks incorporate these measures can review conditional value at risk methodology as a starting point for building institutional-grade frameworks.

CVaR is not a crystal ball — no model survives contact with a genuine black swan intact. But it is a structurally superior tool for quantifying the exposure VaR deliberately ignores.

If your risk model only tells you where the pain starts, you are not managing risk — you are managing the illusion of it.

This content is for educational purposes only and does not constitute financial product advice. Past performance is not indicative of future results. Profit Logic Ltd (ACN 688 669 936) accepts no responsibility for errors or omissions in this content or anywhere on this website. Always seek advice from a licensed financial adviser before making investment decisions.