Welcome to Principles of Asset Allocation!

Welcome to one of the most critical parts of the CFA Level III journey! Think of Asset Allocation as the "skeleton" of a portfolio. It’s the foundation upon which everything else is built. While picking individual stocks is exciting, research shows that the vast majority of a portfolio's return variability comes from the asset allocation decision. In this chapter, we will learn how to build that foundation using mathematical tools, economic logic, and a bit of common sense. Don’t worry if the math looks intimidating at first—we’ll break it down piece by piece!

1. Defining Asset Classes

Before we can allocate money, we need to know what "buckets" we are putting it into. These buckets are our Asset Classes. But you can't just group things randomly. To be effective for Strategic Asset Allocation (SAA), asset classes should follow certain rules.

Criteria for Defining Asset Classes:

  • Homogeneous: The assets within the class should be similar. Example: All stocks in a "Large Cap US Equity" class should behave similarly to economic changes.
  • Mutually Exclusive: An asset shouldn't belong to two classes at once. You don't want to double-count!
  • Exhaustive: The classes should cover the majority of investable wealth globally.
  • Diversifying: Classes should not be highly correlated with each other. If everything moves in lockstep, there's no point in having different buckets.
  • Appropriate size: The class should be large enough to represent a significant portion of a portfolio.

Quick Tip: Think of your asset classes like ingredients in a kitchen. You keep spices in one spot, vegetables in another, and meats in the fridge. You wouldn't put a carrot in the spice rack because it doesn't "behave" like a spice!

Summary Takeaway: Asset classes must be distinct, similar within themselves, and together they should cover the investment universe without overlapping.

2. Mean-Variance Optimization (MVO)

Mean-Variance Optimization (MVO) is the "classic" way to build a portfolio. It was pioneered by Harry Markowitz. The goal is simple: For a given level of risk, we want the highest possible return. Or, for a given return, we want the lowest possible risk.

The Inputs (The "Ingredients"):

To run an MVO, you need three things for every asset class:

  1. Expected Returns: What we think the asset will earn.
  2. Standard Deviations: The measure of risk (volatility).
  3. Correlations: How the assets move in relation to each other.

The Output: The Efficient Frontier. This is a curve showing the set of portfolios that offer the best "bang for your buck." Any portfolio on this line is considered "efficient."

Common Mistakes to Avoid: A major trap in the exam is forgetting that MVO is highly sensitive to inputs. If you change your expected return for Gold by just 0.5%, the MVO might tell you to put 20% of your money in Gold instead of 2%! This is often called "Garbage In, Garbage Out."

Summary Takeaway: MVO helps find the Efficient Frontier, but it is extremely sensitive to the numbers you plug into it.

3. Addressing the Weaknesses of MVO

Because MVO has flaws (like being too sensitive or producing "corner solutions" where it tells you to put 0% in some assets), practitioners use a few tricks to fix it.

A. Reverse Optimization

Instead of starting with expected returns to find weights, we start with the Market Weights (how much the world actually invests in these assets) and "work backward" to see what returns the market is implying. This gives us a much more stable starting point.

B. Black-Litterman Model

This is a favorite for the CFA exam! It starts with Reverse Optimization (the market's view) and then allows the manager to add their own Subjective Views.
Analogy: Imagine a GPS. The "Reverse Optimization" is the standard map. Your "Subjective View" is you telling the GPS, "I know a shortcut on Main Street." The Black-Litterman model combines both to give you the best route.

C. Adding Constraints

Managers often add constraints to MVO to make it more "realistic," such as:

  • Budget Constraint: Weights must sum to 100%.
  • Non-negativity Constraint: No short selling (weights must be \(\ge 0\)).
  • Linear Constraints: "No more than 10% in Emerging Markets."

Quick Review: Which model combines market equilibrium with investor views? The Black-Litterman Model.

4. Risk Budgeting and Risk Contributions

In modern finance, we don't just care about how many dollars are in an asset; we care about how much risk that asset adds to the portfolio. This is Risk Budgeting.

Key Terms to Remember:

  • Marginal Contribution to Risk (MCTR): How much the total portfolio risk changes if we add a tiny bit more of a specific asset.
  • Absolute Contribution to Risk (ACTR): The actual amount of total risk (in percentage points) contributed by an asset class.
    Calculation: \(ACTR_i = Weight_i \times MCTR_i\)

The Goal: In an optimal portfolio, the Excess Return-to-MCTR Ratio should be the same for all asset classes.
Formula-ish concept: \(\frac{Expected Return_i - Risk Free Rate}{MCTR_i}\) should be equal for all \(i\).

Did you know? If one asset has a much higher ratio than others, you should add more of it! It’s giving you more "reward per unit of marginal risk."

Summary Takeaway: Risk budgeting looks at the sources of risk, not just the dollar amounts. The "optimal" point is where every asset provides the same marginal return per unit of marginal risk.

5. Factor-Based Asset Allocation

Sometimes, looking at "Equity" vs. "Bonds" isn't enough. Many assets share the same underlying risks. Factor-Based Allocation looks at things like:

  • Growth: Sensitivity to GDP.
  • Inflation: How the asset reacts to rising prices.
  • Liquidity: The risk of not being able to sell quickly.
  • Style Factors: Value, Momentum, Small-cap.

Analogy: Instead of looking at "Pizza" and "Burgers" (Asset Classes), you look at "Carbs," "Protein," and "Fat" (Factors). You realize that both pizza and burgers are high in fat, so you aren't as diversified as you thought!

Common Pitfall: Students often think Factor-Based is "better" than Asset-Class based. It’s just different. Factors are harder to invest in directly—you usually still have to buy an asset class (like a Value ETF) to get the factor exposure.

6. Rebalancing: Staying on Track

Once you set your Strategic Asset Allocation (SAA), the market will move, and your weights will change. If stocks go up, you might end up with 70% stocks when you only wanted 60%. You need to rebalance.

Rebalancing Strategies:

  1. Calendar Rebalancing: Rebalance every month or quarter. Simple, but you might rebalance when you don't need to.
  2. Percentage-of-Portfolio (Corridor) Rebalancing: Rebalance only when an asset moves outside a set range (e.g., \(60\% \pm 5\%\)).

How wide should the corridor be?

  • Transaction Costs: High costs = Wider corridors (don't trade as often).
  • Risk Tolerance: High risk tolerance = Wider corridors.
  • Correlation: High correlation between assets = Wider corridors (they move together, so the weights stay relatively stable).
  • Volatility: High volatility of the asset itself = Narrower corridors (to keep the risk from exploding).

Encouraging Note: Rebalancing is a "buy low, sell high" discipline. It forces you to sell what has done well and buy what has done poorly, which is psychologically hard but financially smart!

Summary Takeaway: Corridor widths depend on a tradeoff between the cost of trading and the risk of being away from your target. High costs and high risk tolerance lead to wider corridors.

Final Checklist for Success

  • Can you list the criteria for a good asset class?
  • Do you understand why Black-Litterman is often preferred over simple MVO?
  • Can you explain the difference between MCTR and ACTR?
  • Do you know which factors lead to wider or narrower rebalancing corridors?

Keep pushing forward! Asset allocation is the heart of the Level III curriculum. Master these principles, and the rest of the portfolio management topics will fall into place.