Component VaR
Component VaR (CVaR) decomposes a portfolio's total Value at Risk into the contribution of each individual position or risk factor, representing the amount by which the portfolio's VaR would decrease if a given position were removed — accounting not just for that position's standalone volatility but also its correlations with all other positions in the portfolio.
Key takeaways
- Component VaR measures the risk contribution of a position in the context of the full portfolio, not in isolation.
- Unlike standalone VaR, Component VaR can be negative — a position may actually reduce portfolio VaR if it is sufficiently negatively correlated with the rest of the portfolio.
- The sum of all component VaRs equals the total portfolio VaR: this additivity property makes it the preferred decomposition method for risk attribution.
- Component VaR = Beta of asset to portfolio × Portfolio VaR, where beta is computed against the portfolio rather than the market.
- Risk managers use Component VaR to identify concentrated risk contributors and optimize diversification.
Explanation
Value at Risk (VaR) estimates the maximum loss a portfolio could incur over a given holding period at a specified confidence level. While portfolio VaR is a single number, it does not by itself reveal which positions are driving risk. Component VaR provides this decomposition.
Formally, for a portfolio with weight vector w and covariance matrix Σ, portfolio variance is:
σ²_p = w^T Σ w
The marginal contribution of position i to portfolio variance is:
∂σ²_p/∂w_i = 2(Σw)_i
Component VaR for position i is:
CVaR_i = w_i × (Σw)_i / σ_p × z × VaR_portfolio
where z is the confidence level multiplier (e.g., 1.645 for 95% one-tailed). Equivalently:
CVaR_i = ρ_{i,p} × σ_i × w_i × z
where ρ_{i,p} is the correlation of asset i with the total portfolio return, σ_i is asset i's volatility, and w_i is the weight. This formulation clarifies the economic intuition: a position contributes risk proportional to its weight, its standalone volatility, and its correlation with the portfolio.
A key property is that Component VaRs are additive: Σ CVaR_i = Portfolio VaR. This makes Component VaR ideal for risk budgeting — assigning each portfolio manager or strategy a VaR 'budget' — and for monitoring whether actual risk contributions align with intended portfolio construction. In contrast, standalone VaRs are not additive (due to diversification effects), and Incremental VaR (the change in portfolio VaR from adding a new position) is computationally intensive and path-dependent.
Component VaR is most useful in risk factor analysis. Rather than decomposing by individual securities, practitioners often decompose portfolio VaR into contributions from factor exposures (equity beta, duration, credit spread, FX beta, commodity beta), helping isolate whether risk is concentrated in intentional factor bets or unintended residuals.
Formula
CVaR_i = ρ_{i,p} × σ_i × w_i × z × Portfolio Value | Σ CVaR_i = Total Portfolio VaR
Example
A hedge fund holds three positions: $40M in SPY (equity ETF, daily vol 1.2%), $30M in TLT (Treasury ETF, daily vol 0.9%), and $30M in GLD (gold ETF, daily vol 0.8%). Correlations: SPY/TLT = −0.35, SPY/GLD = 0.05, TLT/GLD = 0.15. Portfolio daily vol = 0.74% (due to diversification benefit). At 99% VaR (z = 2.326), portfolio VaR = $100M × 0.74% × 2.326 = $1.72M. Computing Component VaR: SPY contributes $1.45M (84% of total VaR despite being 40% of the portfolio — because of its high vol and positive correlation structure); TLT contributes $0.11M (7%, well below its 30% weight — negative correlation with SPY provides diversification); GLD contributes $0.16M (9%). The risk manager flags SPY as a concentration risk, noting that its Component VaR share (84%) dramatically exceeds its portfolio weight (40%).
Related terms
Beta Concentration Risk Conditional Value At Risk Correlation Covariance Covariance Matrix Credit Spread Diversification Duration Equity Fat Tails Gold