A Model Portfolio's Lifecycle

As wealth managers increasingly seek in-house expertise to create customized portfolios, As wealth managers increasingly seek in-house expertise to create customized portfolios, understanding the foundational principles of Mean-Variance Optimization (MVO) and the Capital Asset Pricing Model (CAPM) is the baseline from which all else follows, particularly if you have ambitions of delivering top quartile performance to your end investors. These concepts, pioneered by Harry Markowitz in 1952 and further developed into CAPM, provide a systematic approach to the benefits of taking on investment risk in one’s quest to deliver higher returns.

In this edition of the Model Portfolio Masterclass Series, we provide a high level overview of how MVO can help construct a Model Portfolio that aligns with a client’s risk tolerance and return objectives. In subsequent editions, the Masterclass Series will dig deeper into the topic and work through several ‘How To’ exercises to explain in more detail a wide range of approaches that a hands-on practitioner can follow.

What is a Model Portfolio?

A model portfolio is a pre-designed collection of investments, such as stocks, bonds, Mutual Funds and Exchange Traded Funds (ETFs) created to meet a specific investment goal or risk level. It serves as a template or guide that investors can follow to build their own portfolios, making it easier to align with objectives like growth, income, or capital preservation.

When it comes to designing Model Portfolios that include Structured Notes, though, such an inclusion makes the task of backtesting a strategy very difficult for those who do not have access to complex risk management systems. For that reason, while there is quite a broad range of investment products that are used to build Model Portfolios, for today’s discussion we shall restrict our attention to just using funds, revisiting this topic in later editions of the Masterclass Series.

Model Portfolio Lifecycle

A model portfolio is a pre-designed collection of investments, such as stocks, bonds, Mutual Before jumping in, it’s worth placing the crucial step of portfolio construction in context by first looking at the full lifecycle and key topics to address when running a Model Portfolio Service.

Defining Investment Objectives and Client Needs

• Identify Client Segments: Start by analysing the target client base. Consider demographic factors, risk tolerance levels, financial goals, and investment horizons.
• Set Clear Investment Objectives: Determine the core objectives for each portfolio, focusing on factors like growth, income, and capital preservation. This will inform the level of risk each portfolio will assume.
• Risk Targeting Framework: Define specific risk targets and tolerances for each portfolio, using standardized measures (e.g., volatility, maximum drawdown, Value-at-Risk) tailored to client needs.

Asset Allocation Strategy

• Design Strategic Asset Allocation (SAA): Develop a long-term asset mix based on the defined risk objectives. This could include equities, bonds, commodities, and alternatives depending on market expectations and risk tolerance.
• Incorporate Tactical Adjustments (TAA): Implement tactical shifts around the SAA to exploit shorter-term market opportunities, adhering to the portfolio's risk budget. Potentially add themes and sub-asset classes as tactical tilts.
• Factor Integration: For more precision, consider incorporating factors (e.g., value, momentum, quality) aligned with the portfolio’s objectives.

Investment Vehicle Selection

• Select Suitable Instruments: Choose between ETFs, Mutual Funds, and other investment vehicles that fit the portfolio's needs.
• Evaluate Cost and Liquidity: Balance expense ratios, bid-ask spreads, and other transaction costs with liquidity to ensure efficient access and scalability.
• Ensure Diversification: Use a diversified mix within each asset class to manage risk and achieve the targeted exposure across sectors and regions.

Portfolio Construction and Optimization

• Optimize Weightings: Use portfolio optimization techniques, such as mean-variance optimization or risk parity, to assign weights that align with the portfolio’s risk-return goals.
• Backtesting and Stress Testing: Assess historical performance and stress test the portfolio under various scenarios (e.g., interest rate shocks, market downturns).
• Implement Risk Controls: Define and apply risk controls, including maximum asset and sector exposures, to protect against concentration risks.

Platform Selection

• Portfolio Management Tools: Ensure the platform supports advanced tools for rebalancing, performance tracking, risk assessment, and reporting to meet your model portfolio needs.
• Platform Fees: Review the fee structure, including administrative, transaction, and custody fees, and consider how they will impact portfolio performance and client fees.
• Distribution Opportunities: Some platforms have extensive networks of financial advisors, which can help expand your reach. Assess the platform’s network size and type of advisors (e.g., wealth managers, IFAs) to ensure it aligns with your target audience.

Portfolio Implementation

• Execution of Trades: Implement the initial portfolio design, carefully managing transaction costs, slippage, and timing.
• Monitor for Drifts: Track any deviations from the target asset allocation due to market fluctuations or asset price movements.
• Rebalancing Protocols: Set periodic or threshold-based rebalancing protocols to maintain the portfolio’s allocation within the set risk bands.

Ongoing Monitoring and Rebalancing

• Regular Performance Review: Continuously review portfolio performance against benchmarks and reassess alignment with client objectives.
• Tactical Adjustments: Make tactical adjustments as necessary to adapt to changing market conditions without deviating from the portfolio’s risk profile.
• Rebalance as Needed: Conduct rebalancing in accordance with either time-based (e.g., quarterly) or drift-based (e.g., asset allocation deviation) protocols.

Implementing a Framework for Testing Model Portfolio Strategies

• If you want to become a hands-on practitioner, it will become apparent that while the theory of fund selection is easy to grasp, the vast number of available funds very quickly presents its own problems.
• When thinking about testing how a particular asset allocation model performs, one needs to differentiate between the model that determines the weight of the allocation to each asset class and sub-asset class, from the individual investments that you will allocate to those buckets?

The framework that we have developed at Algo-Chain is to take an ‘ETF first’ approach. Many ETFs are index trackers providing end investors with access to sections of the market, and in that sense, act as good proxies for parts of the market. ETFs are well suited for testing an asset allocation model as there invariably is an ETF that provides the exposure you are looking for. Once you have a good handle on what your allocation model looks like, it’s your choice if you allocate using an index tracker or instead prefer to allocate to a tried and tested active manager.

There will be many sceptics who frown when you show them your latest backtested portfolio, ignore them, investing is a science, and it is not feasible to launch an investment service without having first tested one’s strategy. How you interpret your test results is a different matter altogether, and is best left for another discussion, but test you must and for that you need data.

Mean-Variance Analysis and the Efficient Frontier

In preparation for subsequent editions of the Masterclass Series where we will drill into the topics of fund selection and portfolio construction, let’s establish the theoretical baseline by which one’s effort will be directed.

At the core of Markowitz’s portfolio theory is the idea that risk and return can be optimized through diversification. By combining assets that vary in returns and risk profiles, investors can create portfolios that aim to minimize risk for a given level of return. The result is the efficient frontier, a curve representing portfolios that offer the highest return for each level of risk.

To construct a portfolio on the efficient frontier, a mean-variance optimizer uses historical data to estimate expected returns, variances, and covariances among assets. For an ETF model portfolio, this might involve selecting a mix of ETFs representing different asset classes or sectors, each with unique return patterns and correlations. By finding the optimal weights for each ETF, wealth managers can build portfolios that balance the trade-off between risk and return in line with client preferences.

Incorporating Risk-Free Assets and Other Model Portfolios

Occasionally it might prove practical to introduce a risk-free asset - typically represented by Government bonds or Treasury bills – but in doing so this will modify the efficient frontier. When this happens, the efficient frontier simplifies to a straight line known as the capital market line, which connects the risk-free asset to what is often referred to as the tangency portfolio that offers the highest Sharpe ratio, or risk-adjusted return.

In practical terms, the tangency portfolio serves as the ‘core’ or base in a mean-variance optimized portfolio. By adjusting the allocation between this portfolio and the risk-free asset, wealth managers can create portfolios tailored to various levels of risk tolerance.

What more commonly happens is that a portfolio manager will have constructed both a low risk and high risk portfolio and will blend these two portfolios with a set of different weights to once again produce a suite of portfolios across a range of risk tolerances. Doing so will no longer ensure that the resulting portfolios live on their appropriate efficient frontiers, but it is often felt the operational benefits of simplicity outweigh the impact of sub-optimal risk adjusted returns.

Capital Asset Pricing Model (CAPM)

Derived from mean-variance optimization, the CAPM introduces the concept that an asset’s risk is not merely its volatility but its systematic risk, or beta – a measure of its correlation with the overall market. CAPM posits that in equilibrium, all investors hold the market portfolio, a weighted portfolio of all risky assets. An asset’s expected return, according to CAPM, depends on its beta relative to the market portfolio, not on its individual volatility.

For wealth managers, CAPM provides insights into risk management by emphasizing assets’ correlations with the market. When constructing an ETF portfolio, selecting ETFs with different betas enables wealth managers to manage exposure to systematic risk effectively, tailoring the portfolio’s overall beta to align with client preferences.

Limitations of Mean-Variance Optimization and CAPM

While mean-variance optimization and CAPM are powerful frameworks, they come with limitations. Both approaches are highly sensitive to estimates of expected returns and covariances. Small changes in these estimates can lead to significant shifts in portfolio allocations, making the process vulnerable to estimation error.

Additionally, traditional mean-variance optimization assumes that past performance will predict future returns, an assumption that may not hold in volatile markets. CAPM’s reliance on beta as a sole risk measure can also be limiting, as it doesn’t account for other risk factors, such as value or momentum.

Addressing Limitations with Robust Estimation Techniques

To address these limitations, wealth managers can employ robust estimation techniques. For example, Bayesian approaches adjust return estimates by incorporating prior beliefs, helping to reduce extreme portfolio allocations. Similarly, shrinkage techniques can refine covariance estimates by ‘shrinking’ them toward a more stable structure, reducing sensitivity to estimation errors.

Factor models also provide a way to enhance mean-variance optimization. By incorporating multiple risk factors - like size, value, and momentum - factor models can capture risks not accounted for by CAPM alone. Using ETFs that target specific factors allows wealth managers to achieve diversification beyond market beta, aligning portfolio construction with modern portfolio theory’s best practices.

Next Edition of The Model Portfolio Masterclass Series

In this edition Using mean-variance optimization and CAPM in ETF model portfolios equips wealth managers with robust tools to construct tailored portfolios. By understanding the efficient frontier and the impact of systematic risk, wealth managers can bring in-house expertise that resonates with clients’ goals. Addressing the limitations of traditional methods with robust estimation techniques ensures a more resilient portfolio, positioning wealth managers to meet client needs in a dynamic market environment.

Disclaimer

*The podcast provided by Allan Lane & Irene Bauer has been converted from their own original content, into a podcast using Generative AI tools and the voices used in the podcast are not their own. All information provided has been fact checked.

The investments referred to in this podcast is targeted at professional Wealth Managers & Financial Advisors and may not be suitable for all investors. Twenty20 Solutions Ltd does not provide, and nothing in this podcast should be construed as, investment or other advice. It is not intended that anything stated in this podcast should be construed as an offer, or invitation to treat, or inducement for you to engage in any investment activity. The information in this podcast relating to model portfolios & individual funds suggested by Algo-Chain is purely for research and educational purposes only.