Harry Markowitz’s dissertation on portfolio selection in 1952 focused on the value of combining two risky investments that do not move in lockstep with one another. Markowitz’s cutting-edge research focused on the combination of these investments and concluded that the right combination could in fact produce higher returns at lower risk. He famously stated that “diversification is the only free lunch in investing.” We serve the free lunch by properly diversifying investments that show low correlation to one another.
This is the basis of modern portfolio theory (MPT). Markowitz is typically referred to as the father of modern portfolio theory, but other pioneers built on his original work, including Bill Sharpe and Eugene Fama. Sharpe developed the capital asset pricing model (CAPM) and the Sharpe ratio, which measures risk-adjusted results. Fama is often referred to as the father of the efficient market hypothesis (EMH). Markowitz and Sharpe won Nobel Prizes in economics in 1990; Fama won in 2013.
For the past 20 years or so, advisors have incorporated Markowitz’s work into mean-variance optimization (MVO) models, which help advisors build diversified portfolios. Markowitz’s work helped transform the way advisors engage their clients, focusing their attention on portfolio outcomes rather than picking individual stocks. Back in 1952, Markowitz did not have access to the powerful computers that advisors now have at their fingertips. Now wealth advisors can optimize a myriad of asset classes, considering their respective return, risk, and correlation statistics in a matter of seconds.
Modern Portfolio Theory assumes that investors are risk averse, and that a rational investor will not invest in a portfolio if a second portfolio offers less risk and the same or better return.
MPT has inherent limitations: investors are not always rational, and they do not always select the less-risky portfolio. Investors often chase returns, gravitating to a hot manager or asset class, especially during bull markets. Markets are not always efficient, and they are prone to boom-and-bust periods, where emotions shift from euphoria to fear. MVO models use long-term capital market assumptions, but returns, risks, and correlations are not stable over the long run. This article addresses some limitations of MPT and evaluates alternative techniques for allocating capital. It delves into the following issues:
• What asset allocation approaches can wealth advisors use?
• What limitations does each approach have?
• How should wealth advisors evolve their approach?
• What is the appeal of a goals-based approach?
Alternatives To MPT
Post-MPT theory came along in 1991, in response to some of the limitations with MPT. This portfolio optimization methodology uses the downside risk of returns instead of the mean variance of investment returns that MPT uses. Both theories describe how to value risky assets and how rational investors should use diversification to achieve portfolio optimization. The difference lies in each theory’s definition of risk and how that risk influences expected returns.
Black-Litterman came on the scene in 1992 when Fischer Black and Robert Litterman tried to address some of MPT’s limitations, most notably that future returns may be different from historical results. The Black-Litterman model starts with an asset allocation that is based on the equilibrium assumption (assets will perform in the future as they have in the past), and then modifies that allocation by considering projections of future asset class performance.
Liability-driven investing is a common approach used when dealing with defined-benefit pension plans, because the liabilities are defined and predictable. It is designed to match a plan’s income needs with the appropriate asset allocation. Alternatively, it can work to align HNW investors’ yearly cash-flow needs.
Risk parity is a portfolio allocation strategy that uses risk to determine allocations across various components of an investment portfolio. It builds on MPT, making allocations based on predetermined risk and return targets. Risk parity is used primarily by hedge funds and institutions. The strategy helps limit risk, but this approach will likely lag in rising markets, where investors are rewarded for taking on risk.
Goals-based investing combines attributes of MPT and behavioral finance by solving for an HNW investor’s goals, rather than maximizing returns or minimizing risks. Goals-based investing moves the discussion from beating the market to funding family needs. It focuses on making ongoing progress toward agreed-upon goals.
Critics point to MPT’s limitations, but many of the alternative asset allocation methodologies have drawbacks as well. For MPT and post-MPT, the biggest limitation is the robustness and accuracy of the data used to optimize. Using only long-term historical averages of the underlying asset classes may be a flawed approach if future results are dramatically different than the long-term historical data. Black-Litterman seeks to address these limitations by using projections of future results, but what if they are also flawed?
Flawed capital market assumptions hurt all the approaches above, as well as the results they produce. Merely using long-term historical averages may lead to higher expectations for returns and income than may be achievable today. Plus, with elevated correlations, investors may not achieve the desired diversification benefits.
Although MPT, post-MPT, and Black-Litterman are somewhat similar in approach, the nuances of each lead to different portfolios. Liability-driven investing and risk parity are more institutional approaches and may not be the best approach for HNW families. Goals-based investing has become increasingly popular because it aligns with investors’ goals and objectives, although consistently reinforcing progress relative to these goals can be challenging, given investors’ fixation on short-term results.