Risk Budgeting: Allocating Risk Across a Portfolio
Risk budgeting is a framework for allocating total portfolio risk across individual positions, asset classes, or managers. Rather than focusing solely on capital weights, risk budgeting ensures that each component contributes an intentional share of portfolio risk. This approach is widely used by institutional investors, pension funds, and asset managers to maintain diversification benefits and prevent any single position from dominating portfolio risk.
What Is Risk Budgeting?
Risk budgeting is the process of setting a total portfolio risk limit and then allocating that risk systematically across portfolio components. Unlike traditional asset allocation, which focuses on how much capital to invest in each asset class, risk budgeting focuses on how much risk each component should contribute.
Risk budgeting separates the allocation decision (how much risk each position should contribute) from the measurement decision (how to calculate those risk contributions). This allows portfolio managers to think about risk in terms of explicit budgets rather than implicit consequences of capital weights.
The framework originated in institutional asset management, where pension funds and endowments needed a systematic way to allocate risk across multiple external managers. The Ontario Teachers’ Pension Plan pioneered this approach in the 1990s, setting plan-wide risk limits and parceling out “risk units” to individual managers while accounting for cross-manager correlations.
While risk budgeting is most commonly implemented using Value at Risk (VaR) or tracking error, the framework itself is flexible. Active equity managers often use tracking error budgets, while pension funds may use surplus risk (the risk of funded status declining) as their risk measure. The core principle remains the same: set a total risk limit, then allocate it intentionally.
For a related concept, see how risk-adjusted performance metrics like RAROC use similar risk allocation principles to evaluate business unit performance.
How to Build a Risk Budget
Building a risk budget involves four key steps, each requiring explicit decisions about risk measurement and allocation:
Step 1: Define the Risk Metric
Before setting any limits, specify exactly how risk will be measured. For a VaR-based risk budget, this means defining:
- Time horizon (e.g., 1-day, 10-day, or 1-month VaR)
- Confidence level (e.g., 95% or 99%)
- VaR methodology (historical simulation, parametric, or Monte Carlo)
Step 2: Set the Total Portfolio Risk Limit
Establish the maximum acceptable portfolio VaR. For an institutional fund, this might be expressed as a dollar amount ($5 million at 95% confidence over a one-month horizon) or as a percentage of assets (5% monthly VaR). This total limit becomes the “budget” to be allocated.
Step 3: Calculate Component VaR for Each Position
Decompose the total portfolio VaR into component VaRs that sum to the total. Component VaR captures each position’s contribution to portfolio risk, accounting for correlations with other holdings. For parametric VaR, this requires the covariance matrix; for simulation-based methods, component VaRs are derived from scenario-level P&L contributions. For detailed mechanics, see Portfolio VaR Risk Decomposition.
Step 4: Allocate Risk Budgets
Assign target risk contributions to each asset class or manager. These allocations can be proportional to capital weights, based on expected return-to-risk ratios, or set using other criteria. The key insight from Jorion’s research is that the sum of unit-level VaR limits can exceed the group-level limit because of diversification benefits — for example, individual manager limits might sum to $75 million while the fund-level limit is $60 million.
Risk budgeting enables delegation without micromanagement. Once a manager receives their risk budget, they have autonomy to transact freely within that limit without requiring senior approval for individual trades. This is why pension funds favor this approach for managing multiple external managers.
How to Calculate Component VaR for Risk Budgeting
Component VaR measures each position’s contribution to total portfolio risk. The formula relates a position’s weight, its marginal impact on portfolio risk, and the total portfolio VaR:
Where:
- wi — the weight of position i in the portfolio
- MVARi — the marginal VaR of position i, defined as ∂VaR/∂wi (the partial derivative of portfolio VaR with respect to the position’s weight)
A critical property of component VaR is that all component VaRs sum exactly to total portfolio VaR. This additive property makes it ideal for risk budgeting — you can allocate the total risk budget across positions without any “unallocated” risk.
Marginal VaR depends on the position’s volatility, its correlation with the rest of the portfolio, and the portfolio’s overall volatility. Positions that are highly correlated with the portfolio have higher marginal VaRs, while positions that provide diversification (low or negative correlation) have lower marginal VaRs.
For a detailed walkthrough of component VaR calculation mechanics, see our article on Portfolio VaR Risk Decomposition.
Risk Budgeting Example
Consider an institutional portfolio allocating $100 million across three asset classes. The investment committee has set a total VaR limit of $5 million (5% at 95% confidence over a one-month horizon).
An institutional fund with a $100 million portfolio and a 95% one-month VaR limit of $4.8 million allocates capital across three asset classes. Using historical volatilities and correlations from the past five years, the risk management team calculates the following component VaR decomposition:
| Asset Class | Capital Weight | Component VaR | Risk Contribution |
|---|---|---|---|
| U.S. Equities | 55% | $3.36M | 70% |
| Investment-Grade Bonds | 35% | $0.96M | 20% |
| Real Assets (REITs) | 10% | $0.48M | 10% |
| Total | 100% | $4.80M | 100% |
The equity allocation consumes 70% of the risk budget despite representing only 55% of capital. This occurs because equities have higher volatility than bonds and are moderately correlated with the real assets allocation. Bonds, despite their 35% capital weight, contribute only 20% of portfolio risk due to their lower volatility and diversification benefits relative to the equity-dominated portfolio.
This example illustrates why risk budgeting provides a different perspective than capital allocation. A portfolio manager focused only on capital weights (55% equities) might underestimate how much the equity allocation dominates portfolio risk (70% of VaR).
Risk Budgeting vs Mean-Variance Optimization
Risk budgeting and mean-variance optimization (MVO) are complementary frameworks, not substitutes. Understanding their differences helps clarify when to use each approach.
Risk Budgeting
- Objective: Allocate total risk across positions
- Primary inputs: Covariance matrix, total risk limit
- Output: Target risk contributions (component VaR)
- Focus: Risk management and control
- Advantage: Does not require return forecasts
Mean-Variance Optimization
- Objective: Maximize return for given risk level
- Primary inputs: Expected returns, covariance matrix
- Output: Optimal asset weights
- Focus: Portfolio construction
- Sensitivity: Highly sensitive to return estimates
The key distinction is that MVO outputs target weights, while risk budgeting outputs target risk contributions. Many institutional investors use both: MVO to determine initial portfolio weights, then risk budgeting to express those weights as risk allocations for ongoing monitoring and control.
Risk budgeting’s primary advantage is that it does not require expected return estimates — only covariances. Since return forecasts are notoriously unreliable and MVO is highly sensitive to estimation errors, risk budgeting offers a more robust framework for ongoing portfolio management.
Monitoring and Rebalancing
A risk budget is only useful if it’s actively monitored and enforced. Effective monitoring involves regular calculation of component VaRs and predefined triggers for rebalancing:
Daily or Weekly Monitoring
- Calculate component VaR for each position or asset class
- Compare actual risk contributions to budget allocations
- Flag positions exceeding their risk budget by a threshold (e.g., 120% of allocation)
Rebalancing Triggers
- Threshold breach: Rebalance when any position exceeds its risk budget by more than 20%
- Volatility regime change: Review all budgets when market volatility increases significantly (e.g., VIX doubles)
- Correlation breakdown: Reassess budgets when correlation assumptions are violated
Attribution Analysis
When a position exceeds its risk budget, determine whether the breach is due to:
- Position size changes (drift in capital weights)
- Volatility changes (the asset became more volatile)
- Correlation changes (the asset became more correlated with the portfolio)
Correlations tend to spike toward 1.0 during market stress, which can cause risk budgets set during calm periods to be inadequate. Monitor correlation changes actively and consider using stress-adjusted correlations when setting initial budgets. For active managers, see our Tracking Error Calculator for monitoring active risk budgets.
Common Risk Budgeting Mistakes
Risk budgeting is conceptually straightforward but easy to implement poorly. Here are the most common mistakes:
Mistake 1: Confusing Capital Weights with Risk Contributions
The most fundamental error is assuming that capital allocation equals risk allocation. As the example above shows, a 55% allocation to equities can represent 70% of portfolio risk. Portfolio managers who report only capital weights may be unaware of their true risk concentrations.
Fix: Always calculate and report both capital weights and risk contributions. Use risk budgeting to identify when a position’s risk contribution exceeds its capital weight.
Mistake 2: Ignoring Covariance Changes
Risk budgets calculated using historical correlations can become dangerously inaccurate when correlations shift. During the 2008 financial crisis, correlations between equities and credit spiked from approximately 0.3 to 0.8, causing positions that appeared well-diversified to move in lockstep.
Fix: Use stress-adjusted or conditional correlations when setting risk budgets. Regularly stress-test the budget using crisis-period correlation assumptions.
Mistake 3: Treating the Budget as Static
Setting an annual risk budget and ignoring it until year-end defeats the purpose. Volatility regimes change — the VIX can triple in a matter of weeks — and a budget appropriate for calm markets may be inadequate during stress.
Fix: Implement dynamic risk budgets that scale with market volatility, or at minimum, review budgets whenever volatility regimes shift significantly.
Limitations of Risk Budgeting
While risk budgeting is a powerful framework, it inherits the limitations of its underlying risk measures and models:
VaR-based limitations propagate: If your risk budget uses VaR, the limitations of your chosen VaR methodology apply. Parametric VaR assumes normally distributed returns; historical simulation depends on lookback period selection; all VaR methods underestimate tail risk by design (VaR says nothing about losses beyond the confidence threshold). A risk budget that appears balanced may still be exposed to extreme events.
Correlation Instability
Risk budgets depend critically on correlation estimates, which are notoriously unstable. Correlations measured during normal markets may not hold during stress periods. Diversification benefits assumed in the budget may evaporate precisely when they’re needed most.
Model Dependency
Component VaR depends on the VaR methodology used (historical simulation, parametric, or Monte Carlo). Different models can produce materially different risk allocations for the same portfolio, making it difficult to compare budgets across organizations.
Illiquid and Non-Mark-to-Market Assets
Risk budgeting is poorly suited for illiquid assets like private equity, venture capital, and real estate. These assets are often marked to model rather than marked to market, producing artificially smooth returns that understate true volatility. A VaR-based risk budget will systematically underallocate risk to these positions.
For a broader discussion of VaR limitations in portfolio management, see VaR in Investment Management.
Frequently Asked Questions
Disclaimer
This article is for educational and informational purposes only and does not constitute investment advice. Risk budgeting approaches and VaR calculations involve assumptions about volatility and correlations that may not hold in all market conditions. Always conduct your own analysis and consult a qualified financial advisor before making investment decisions.
