Editor’s Introduction for the 2023 Special Issue on Multi-Asset Strategies and Asset Allocation

INTRODUCTION

This is the eighth special issue on multi-asset strategies and asset allocation. The opening article is a synopsis of PMR’s “Multi-Asset Strategies” webinar presented in a Q&A format that summarizes the latest thinking about multi-asset strategies. The four expert panelists were Stefano Cavaglia (State of Wisconsin Investment Board), Stephen A. Gorman (Wellington Management Company), Brian Jacobsen (Allspring Global Investments), and Eugene Podkaminer (Franklin Templeton Investment Solutions).

The next two articles are perspective pieces by two prominent practitioners. Antti Ilmanen, a principal at AQR Capital Management, explains what happened to major asset prices prior to 2022 and what changed in 2022 in “Investing in Interesting Times.” In “Multi-Asset Portfolios in the New Order,” Wai Lee of Allspring Global Investments explores the potential challenges of investing in a multi-asset portfolio in the new order of heightened risks, framing the discussion in the context of strategic and tactical allocation, capital and risk, and assets and factors.

A major challenge faced by strategic asset allocators is inflation risk. This risk has forced allocators to reassess the analytical tools they employ, including standard simulation methods. In “Strategic Asset Allocation and Inflation Resilience,” Wesley K. Phoa explains why standard simulation methods fail to do an effective job in modeling inflation dynamics. In particular, they do not generate uncertainty in long-term average inflation. Another drawback is that despite low-frequency regime switching between negative and positive stock–bond correlation regimes linked to inflation effects, standard simulation methods fail to capture this dynamic. The author describes how it is not necessary for asset allocators to build a joint model of asset class returns, inflation, cash yields, and correlation regimes to overcome the limitations of simulation models. Instead, they can retrofit an existing simulation engine. Phoa argues that this method is efficient and scalable to many asset classes.

In traditional portfolios, a fundamental determinant of risk is the relationship between stock and bond returns. For the first two decades of the 21st century, the stock–bond correlation was consistently negative, and investors could depend on their bond allocation to provide protection during periods when equities declined. In “A Changing Stock–Bond Correlation: Drivers and Implications,” Alfie Brixton, Jordan Brooks, Pete Hecht, Antti Ilmanen, Thomas Maloney, and Nicholas McQuinn explain why this was not the case in the previous century. A simple macroeconomic model— supported by international empirical evidence—posits that equity and bond market returns exhibit opposite-sign sensitivities to growth news and same-sign sensitivities to inflation news. As a result, periods of relatively higher inflation (growth) uncertainty tend to coincide with positive (negative) stock–bond correlation. The recent increase in inflation uncertainty could thus bring back the positive correlation that existed toward the end of the 20th century. The authors explain the broad implications for investors, either increasing portfolio risk or requiring allocation changes that may reduce expected returns. The potential for alternatives to make up the diversification deficit in a positive stock–bond correlation world is then explored.

There is a growing body of empirical research showing that carry, value, and momentum factors exist in all asset classes, suggesting that these three factors may be the robust styles across asset classes and history. In “Multi-Asset Style Factors Have Their Shining Moments,” Philippe Declerck, Benoit Bellone, Mounir Nordine, and Thomas Vy examine the following questions empirically: (1) How do multi-asset styles perform across time and across different market regimes? (2) How should multi-asset styles be expected to behave during alternative phases of the stock market cycle? (3) Are cross-asset styles sensitive to volatility conditions? (4) Are there different responses to changes in bond yields? and (5) Is any style more likely to be structurally more cyclical or defensive? The authors explore these questions first by describing how single asset class factors behave and then by assessing the current debate opposing style rotation to diversification to determine if a case can be made for more time-varying and concentrated multi-asset style portfolio constructions.

As market conditions vary over the course of the business cycle, Alessio de Longis and Dianne Ellis investigate whether investors are compensated to take risk and the type of risk in “Tactical Asset Allocation, Risk Premia, and the Business Cycle: A Macro Regime Approach.” A practical regime-based framework for tactical asset allocation (TAA) is proposed by the authors. The framework involves combining leading economic indicators and global risk appetite for the purpose of identifying the following four macro regimes: recovery, expansion, slowdown, and contraction. Focusing on the term premium, credit premium, and equity premium, de Longis and Ellis report distinct performance characteristics across regimes for traditional asset classes and their underlying risk factors. Simple and practical examples of TAA strategies for long-only multi-asset and fixed-income portfolios are provided.

Building income-oriented portfolios comes with a complex set of unique tradeoffs. Investor types either approaching or in retirement need to be concerned about the income properties of investment portfolios. In “Asset Allocation for Retirement Income: A Framework for Income-Oriented Investors,” Steve Sapra, Sean Klein, and Rene Martel address the income characteristics of equities and bonds and propose a framework for building multi-asset portfolios with varying degrees of income orientation. A staple of income-oriented strategies, fixed-income funds, tend to trade off volatility in income for volatility in wealth. The authors demonstrate how an income-oriented investor can use equity and fixed-income assets to trade off the level of risks to their near-term income, long-term income, and their wealth. They conclude that despite the long-term growth potential of equity income and the prospect for higher bond yields over the next several years, a meaningful allocation to bonds is likely optimal for investors with an orientation toward organic portfolio income.

In “Regret and Optimal Portfolio Allocations,” David Blanchett introduces an objective function to incorporate regret aversion into portfolio optimizations, providing an extension of the more traditional portfolio optimization framework. Regret aversion in the proposed framework is a parameter distinct from risk aversion. The author then explores the implications of regret on an individual stock portfolio. Considering regret—which is an emotion that can be experienced by both retail and institutional investors—can produce notable changes in optimal portfolio weights. Blanchett finds that incorporating regret aversion results in higher allocations to relatively inefficient and potentially risky assets. The portfolio impact, however, varies depending on investor preferences and modeling assumptions.

Although the Markowitz mean–variance portfolio optimization model remains the workhorse method for most asset allocators, it has several well-known shortcomings. One of the framework’s major shortcomings is that it is an asset-only framework, failing to recognize liabilities or goals and how they may change over time. Peter Mladina, in “An ICAPM Framework for Asset Allocation,” argues that portfolio theory based on the intertemporal capital asset pricing model (ICAPM) largely resolves key issues with modern portfolio theory and standard CAPM portfolio theory, offering a unified framework for liability-relative, goals-based, and asset-only asset allocation. Documenting the application of ICAPM portfolio theory to practice, Mladina addresses key implementation and technical issues related to the liability hedge, risky-asset portfolio optimization and constraints, portfolio selection and Monte Carlo simulation, and extensions to goals-based and asset-only asset allocation.