Chapter 17: Risk
17.1 Systematic vs. Unsystematic Risk and Process of Enterprise Risk Management
Deconstructing Total Portfolio Risk
Total portfolio risk consists of two distinct components:
$$\sigma_{\text{total}}^2 = \sigma_{\text{systematic}}^2 + \sigma_{\text{unsystematic}}^2$$
Portfolio Variance
^
| \
| \ Total Risk
| \
| +------------------------- Unsystematic / Idiosyncratic Risk (Diversifiable)
| \
| +----------------------- Systematic / Market Risk (Non-diversifiable)
|
+-------------------------------------------> Number of Portfolio Stocks (N)
N = 30-40 stocks
- Systematic Risk (Market Risk): Arises from macroeconomic and geopolitical forces affecting all businesses simultaneously (inflation spikes, interest rate hikes, currency devaluations, wars). Systematic risk cannot be eliminated through diversification.
- Unsystematic Risk (Idiosyncratic / Specific Risk): Specific to an individual company or industry (labor strikes, corporate governance fraud, product recall, loss of key patent). Through proper diversification across 30–40 non-correlated stocks, unsystematic risk approaches zero.
The Enterprise Risk Management (ERM) Process in PMS
Managing an institutional PMS involves a structured, continuous 5-step risk management lifecycle:
+---------------------------------------------------------------------------------------------------+
| THE ERM PROCESS IN PORTFOLIO MANAGEMENT |
+-------------------+-------------------------------------------------------------------------------+
| 1. Identification | Catalog all sources of risk (market, credit, liquidity, operational, model) |
+-------------------+-------------------------------------------------------------------------------+
| 2. Measurement | Quantify risk using VaR, standard deviation, duration, credit spreads |
+-------------------+-------------------------------------------------------------------------------+
| 3. Evaluation | Compare measured risk metrics against mandated IPS tolerance limits |
+-------------------+-------------------------------------------------------------------------------+
| 4. Mitigation | Implement hedging (index futures, puts), rebalancing, diversification limits |
+-------------------+-------------------------------------------------------------------------------+
| 5. Monitoring | Daily MTM stress tests, limit monitoring, and compliance oversight |
+-------------------+-------------------------------------------------------------------------------+
17.2 Quantitative Risk Measurement: Value-at-Risk (VaR), Conditional VaR (CVaR), and Tracking Error
Value-at-Risk (VaR)
Value-at-Risk (VaR) is the primary quantitative metric used by financial institutions to measure downside tail risk. VaR answers the question: "What is the maximum monetary loss expected over a specified holding period ($T$) at a given statistical confidence level ($1 - \alpha$)?"
Normal Distribution of Portfolio Returns
Probability
^
| Mean Return
| |
| |
| +--+--+
| / \
Tail| / \
Loss| / \
<---|----------------+ \
+-------------------------------------------> Portfolio Return
5% Cutoff (VaR)
95% Confidence Interval
Methodologies for Calculating VaR:
- Parametric (Variance-Covariance) VaR: Assumes portfolio returns follow a normal distribution: $$\text{VaR}{(\alpha, T)} = V_p \times \left[ Z\alpha \times \sigma_p \times \sqrt{T} - \mu_p \times T \right]$$ Where $V_p$ is portfolio value and $Z_\alpha$ is the standard normal critical value ($Z_{0.05} = 1.645$ for 95% confidence, $Z_{0.01} = 2.326$ for 99% confidence).
- Historical Simulation VaR: Revalues the current portfolio against actual historical daily market price shifts over the past 500 or 1,000 trading days, ranking the outcomes to identify the percentile cutoff.
- Monte Carlo Simulation VaR: Generates tens of thousands of simulated market paths using geometric Brownian motion algorithms.
Conditional VaR (CVaR) / Expected Shortfall
VaR has a critical structural limitation: it only identifies the threshold loss, revealing nothing about the magnitude of losses beyond that point. Conditional VaR (CVaR) or Expected Shortfall (ES) quantifies the expected average loss given that the loss exceeds the VaR cutoff: $$\text{CVaR}\alpha = E[L \mid L > \text{VaR}\alpha]$$ CVaR is a coherent risk measure satisfying sub-additivity, making it mathematically superior for managing tail risk.
Tracking Error (TE)
For portfolios managed relative to a designated benchmark index, Tracking Error measures the volatility of excess returns: $$\text{Tracking Error} = \sigma_{(R_p - R_b)} = \sqrt{\frac{\sum_{t=1}^T (e_t - \bar{e})^2}{T - 1}}$$ Where $e_t = R_{pt} - R_{bt}$ is the excess return (active return) at period $t$.
- Passive index funds strive for a tracking error approaching zero.
- Active PMS managers accept higher tracking error (e.g., 4% to 8%) in pursuit of positive active returns (Alpha).
17.3 Stress Testing, Scenario Analysis, and Mitigation Frameworks in PMS
Stress Testing and Scenario Analysis
Standard risk models (such as VaR) often underestimate risk during extreme market panics because asset correlations tend to spike toward +1.0 and liquidity evaporates. To evaluate portfolio resilience under extreme tail-risk conditions, portfolio managers deploy Stress Testing and Scenario Analysis:
+---------------------------------------------------------------------------------------------------+
| STRESS TESTING & SCENARIO METHODOLOGIES |
+-------------------+-----------------------------------+-------------------------------------------+
| Approach | Operating Mechanism | Historical / Hypothetical Scenarios |
+-------------------+-----------------------------------+-------------------------------------------+
| Historical | Re-run current portfolio holdings | - 2008 Global Financial Crisis |
| Scenario Analysis | through exact price shifts of past| - March 2020 COVID-19 Liquidity Shock |
| | market crises | - 2013 Taper Tantrum Currency Collapse |
+-------------------+-----------------------------------+-------------------------------------------+
| Hypothetical | Model severe simultaneous shocks | - Sudden 250 bps interest rate hike |
| Scenario Analysis | to macro variables without past | - Crude oil spike to $150/barrel |
| | precedent | - Sovereign credit downgrade of major debt|
+-------------------+-----------------------------------+-------------------------------------------+
| Sensitivity | Shock an individual parameter | - 20% decline in Nifty 50 |
| Analysis | holding other variables constant | - 100 bps widening in credit spreads |
+-------------------+-----------------------------------+-------------------------------------------+
Risk Mitigation Protocols in PMS Accounts
A professional Portfolio Manager applies tiered risk controls at multiple levels:
+---------------------------------------------------------------------------------------------------+
| TIERED RISK MITIGATION MATRIX |
+-------------------+-------------------------------------------------------------------------------+
| Security Level | - Hard stop-loss guidelines |
| | - Single stock exposure cap (e.g., maximum 8%–10% of portfolio NAV) |
+-------------------+-------------------------------------------------------------------------------+
| Sector Level | - Sector exposure limit (e.g., maximum 25%–30% in any single industry) |
+-------------------+-------------------------------------------------------------------------------+
| Market Cap Level | - Defined minimum allocations to Large Cap, Mid Cap, and Small Cap buckets |
+-------------------+-------------------------------------------------------------------------------+
| Liquidity Level | - Mandated minimum allocation to T-Bills, Liquid Funds, or high-volume stocks |
| | to ensure client withdrawal liquidity without market impact penalties |
+-------------------+-------------------------------------------------------------------------------+
| Derivative Level | - Deploying index futures or protective puts during high-volatility regimes |
| | within SEBI PMS hedging regulations |
+-------------------+-------------------------------------------------------------------------------+