Value-at-Risk (VaR) is ubiquitous in risk management — regulators require it, risk reports cite it, and portfolio managers use it as a quick sanity check. But for Indian equity portfolios, standard parametric VaR (which assumes normally distributed returns) systematically underestimates the probability and severity of large drawdowns.
Fat Tails in Indian Equity Returns
A study of Nifty 50 daily returns from 2000–2025 shows excess kurtosis of approximately 6.8 — far above the normal distribution's kurtosis of 3. In plain English: large daily moves (>3%) occur roughly 4× more frequently than a normal distribution predicts. The March 2020 COVID crash (Nifty -13% in a single session) and the 2008 Lehman contagion both produced returns in the 8–10 sigma range under a normal assumption.
CVaR: A Better Tail Risk Measure
Conditional Value-at-Risk (CVaR), also called Expected Shortfall, addresses this by measuring the expected loss given that the loss exceeds VaR. For a 95% confidence level, CVaR asks: "On the worst 5% of days, what is the average loss?" This coherent risk measure is far more informative for Indian portfolios because it captures the severity of tail events, not just their threshold.
India-Specific Tail Events to Model
- RBI surprise rate decisions — Unexpected 50bps+ moves have historically caused Nifty Bank to drop 3–5% intraday.
- FII outflow cascades — When DXY spikes above 105 and US 10Y yields rise, FII selling creates correlated drawdowns across sectors.
- Earnings season gaps — Mid-cap stocks with thin float regularly open 15–25% lower on earnings misses.
- Policy shocks — Demonetisation (2016), F&O circular changes, and sudden sector regulations create regime breaks in historical return distributions.
Spectrum reports both 95% VaR and CVaR for your portfolio, computed using historical simulation over a rolling 3-year window. The historical method captures actual fat-tail behaviour without distributional assumptions.