Other Asset Allocation Approaches

Welcome to Other Asset Allocation Approaches!

In your CAIA studies, you have seen that traditional Mean-Variance Optimization (MVO) has practical drawbacks, including sensitivity to inputs, error maximisation, and an over-reliance on expected returns. To address these limitations, institutional investors employ several alternative asset allocation frameworks. In this chapter, we cover the core alternative approaches outlined in the CAIA curriculum: the core-satellite approach; top-down, bottom-up, and mixed approaches; risk budgeting and risk buckets; risk parity; and heuristic quantitative weighting schemes.

1. The Core-Satellite Approach

The core-satellite approach divides a portfolio into two distinct segments designed to balance broad market exposure and outperformance:

• The Core: Represents the largest share of the portfolio (often 70% to 80%). It is typically allocated to passive or low-cost index products designed to capture broad market beta efficiently.
• The Satellites: Represent smaller, focused allocations (20% to 30%) made to active managers, alternative investment strategies, or tactical ideas aiming to generate alpha and improve diversification.

Benefits of Core-Satellite:
• Cost Efficiency: High active management fees are confined to the satellite investments rather than paid across the total portfolio.
• Clear Risk Separation: Market beta risk is anchored in the core, while active tracking risk is ring-fenced within the satellites.

2. Top-Down, Bottom-Up, and Mixed Approaches

Investors can determine portfolio weights through different hierarchical processes:

• Top-Down Approach: Starts at the macroeconomic level. The allocator evaluates global economic indicators, inflation, interest rates, and geopolitical trends to establish allocations across asset classes, regions, or sectors before selecting individual funds or securities.
• Bottom-Up Approach: Focuses primarily on individual securities or single-manager opportunities without predefined top-down targets. The aggregate portfolio allocation is simply the cumulative result of individual asset selections.
• Mixed Approach: Combines top-down strategic and tactical guidance with bottom-up manager and security selection. For example, top-down policy bounds (caps and floors) establish the allowable asset-class ranges, while bottom-up fundamental analysis determines specific security or fund investments within those bounds.

3. Risk Budgeting and Risk Buckets

Traditional asset allocation allocates capital (dollars or percentages). However, capital allocations do not equal risk allocations: a portfolio with 60% equities and 40% fixed income often has over 90% of its total volatility driven by equities. Risk budgeting resolves this by allocating units of total portfolio risk (such as variance or Value at Risk) across investment categories or factors.

Risk Buckets

Allocators frequently group assets into strategic risk buckets based on economic function rather than legal asset class (e.g., Growth/Capital Appreciation, Capital Preservation/Defensive, and Inflation-Hedging).

Factor-Based Risk Budgeting

In a factor-based framework, portfolio variance is decomposed into systematic risk factor exposures (such as equity risk, term/duration risk, credit spread risk, and liquidity risk). Rather than asking "How much should we invest in private equity vs. public stocks?", the allocator determines the target marginal contribution to risk ( ext{MCR}) from each underlying factor, ensuring that the portfolio is not inadvertently concentrated in a single macroeconomic risk driver.

4. Risk Parity

Risk parity is an asset allocation method where each asset class or risk bucket contributes an equal share to the total portfolio risk. In a two-asset portfolio of equities and bonds, a standard 60/40 mix is heavily dominated by equity risk. Risk parity rebalances the portfolio so that the percentage contribution to risk is identical across all components.

Mechanics and the Use of Leverage

Because lower-risk assets like government bonds exhibit much lower volatility than equities, an un-levered equal-risk portfolio would require a massive capital allocation to bonds, resulting in an expected return that may fail to meet investor return hurdles. Risk parity addresses this by applying leverage to the lower-risk asset class (or to the entire balanced risk-parity portfolio) to scale up total expected return while maintaining equal risk contributions.

The Volatility Anomaly

Risk parity is grounded partly in the low-volatility anomaly (or volatility anomaly). Empirical research demonstrates that low-risk, low-beta assets often deliver higher risk-adjusted returns (higher Sharpe ratios) than high-beta assets. By leveraging a high-Sharpe-ratio, low-volatility asset base, risk-parity portfolios aim to outperform un-levered traditional portfolios across full market cycles.

5. Other Quantitative Approaches

When investors seek rules-based allocations without the parameter estimation error of MVO, they turn to simpler quantitative models:

• Equal-Weighting ( ext{1/N} Approach): Capital is divided equally across all available assets, setting weight \(w_i = \frac{1}{N}\). This heuristic requires zero estimation of expected returns, volatilities, or correlations, completely eliminating optimizer estimation error and providing an empirical benchmark that is surprisingly difficult for complex models to beat out-of-sample.
• Inverse-Volatility Weighting: Weights are assigned inversely proportional to each asset's individual volatility (standard deviation, \(\sigma_i\)), ignoring correlations:
\(w_i = \frac{1/\sigma_i}{\sum_{j=1}^{N} (1/\sigma_j)}\)
Assets with lower volatility receive higher weights, dampening total portfolio volatility without requiring correlation matrix estimation.