VaR and Risk Budgeting in Investment Management

Welcome to the World of Risk Budgeting!

Hello there, future FRM! In your journey through Part II, you’ve already learned how to calculate Value-at-Risk (VaR). In this chapter, we take that knowledge and put it to work in the real world of Investment Management. Think of risk as a "currency" that a fund manager has to spend. How much risk should we take on? Where should we spend it to get the best return? That’s what Risk Budgeting is all about. Don’t worry if the math looks a bit scary at first—we’ll break it down step-by-step using simple analogies.

1. What exactly is Risk Budgeting?

In simple terms, Risk Budgeting is the process of deciding how much risk to take and how to allocate that risk across different assets or managers. Just like a household budget tracks where every dollar goes, a risk budget tracks where every unit of "volatility" or "VaR" is being used.

Why do we do it?
• To ensure we don't take more risk than the client is comfortable with.
• To make sure we are getting "paid" (return) for the risks we take.
• To identify which specific assets are driving the overall risk of the portfolio.

Key Takeaway

Risk budgeting transforms risk management from a "defensive" tool (preventing losses) into an "offensive" tool (efficiently allocating capital to maximize returns).

2. The Three Pillars of Portfolio VaR: MVaR, IVaR, and CVaR

This is the "meat and potatoes" of this chapter. Students often get these three confused, but they are quite different. Let's look at them through the lens of a Chef making a Stew.

A. Marginal VaR (MVaR)

The Definition: MVaR is the change in the total portfolio VaR resulting from a small (marginal) change in a specific asset's weight.

The Math: \( \text{MVaR}_i = \frac{\partial \text{VaR}_p}{\partial w_i} \)

The Analogy: Imagine you are adding a tiny pinch of salt to your stew. MVaR tells you how much the "saltiness" of the entire pot changes with that tiny extra pinch. It represents the sensitivity or the rate of change.

B. Incremental VaR (IVaR)

The Definition: IVaR measures the impact on total VaR when you make a significant change, such as adding a completely new asset or selling an entire position.

The Math: \( \text{IVaR} = \text{VaR}_{\text{new}} - \text{VaR}_{\text{old}} \)

The Analogy: This is like deciding to dump a whole bucket of potatoes into your stew. It’s not a "tiny pinch"; it’s a before-and-after comparison of the total risk.

C. Component VaR (CVaR)

The Definition: CVaR tells us how much the total VaR would decrease if a specific position were removed from the portfolio, accounting for its correlation with other assets.

The Math: \( \text{CVaR}_i = \text{MVaR}_i \times w_i \)

The Analogy: If you have a bowl of stew, CVaR tells you exactly how many "risk calories" are coming specifically from the meat, the carrots, or the broth. If you add up all the CVaRs, you get the Total Portfolio VaR.

Quick Review Box

• MVaR: The slope (rate of change).
• IVaR: The difference (before vs. after).
• CVaR: The piece of the pie (total contribution).
Pro-Tip: If you sum up all CVaRs (\( \sum \text{CVaR}_i \)), you must get the total Portfolio VaR. This is a common exam trick!

3. Using VaR to Allocate Risk

Now that we know how to measure risk, how do we use it to manage a portfolio? We use the Risk-to-Revenue ratio. In a perfectly optimized portfolio, the ratio of Expected Return to Marginal VaR should be the same for all assets.

\( \frac{E(R_i) - R_f}{\text{MVaR}_i} = \text{Constant for all assets} \)

Why? Because if one asset gives you more return per "unit of marginal risk" than another, you should move your money into that asset until the ratios even out. This is the "Equimarginal Principle."

Did You Know?

Most students think you should just look at the asset's individual volatility. But in a portfolio, correlation is king. An asset might be very volatile on its own, but if it moves in the opposite direction of your other assets (low or negative correlation), its MVaR might actually be very low!

4. Risk Budgeting with Active Management

In active management, we aren't just looking at total VaR; we are looking at Tracking Error (the risk of deviating from a benchmark). Risk budgeting here involves two parts:

1. Strategic Asset Allocation: Deciding the long-term weights of asset classes (e.g., 60% stocks, 40% bonds).
2. Active Management Budget: Giving individual fund managers a "budget" of how much they can deviate from those weights to try and beat the market.

Common Pitfalls to Avoid

• Ignoring Correlations: Portfolio VaR is not just the sum of individual VaRs. You must account for how assets move together.
• The "Fat Tail" Problem: VaR assumes a certain distribution (usually normal). In the real world, extreme events happen more often than the model predicts. Never rely on VaR alone!
• Treating MVaR as CVaR: Remember, MVaR is a derivative (a rate), while CVaR is an amount.

5. Performance Measurement: Sharpe vs. VaR-based measures

In FRM Part I, you learned the Sharpe Ratio. However, in Part II, we sometimes use VaR-based measures for performance evaluation because they focus on "downside" risk rather than "total" volatility.

Risk-Adjusted Return on Capital (RAROC):
\( \text{RAROC} = \frac{\text{Expected Return}}{\text{VaR}} \)

If a manager has a high RAROC, they are generating a lot of profit for every dollar of "tail risk" they are taking. This is a favorite tool for banks and large investment firms.

Summary and Key Takeaways

• Risk Budgeting is the systematic allocation of risk (usually VaR) across a portfolio.
• Marginal VaR measures how much the next dollar invested changes total risk.
• Component VaR shows exactly how much each asset contributes to the total portfolio VaR.
• Optimization happens when the Return/MVaR ratio is equal across all positions.
• RAROC is a key metric for evaluating whether a manager is using their risk budget wisely.

Don't worry if this seems tricky at first! The math of partial derivatives in MVaR can be intimidating, but if you focus on the concept—that every asset "contributes" to risk based on its size and its relationship with others—you are well on your way to mastering this chapter. Keep practicing the CVaR and MVaR calculations, and you'll do great!