Editor’s Letter

In recent editor’s letters, I encouraged contributions to the journal that might form the basis for future topic-oriented issues. One of those topics is gender and race, on the grounds that much of what we read about retirement evaluates the aggregate, which is still well-represented by the older, male, white workforce of the past. We have seen research on the circumstances and behavior of women and a smaller amount on minorities, but we need additional insights and implications. I also continue to encourage articles on income/wealth on the grounds that we tend to focus on the middle to high end of the income/wealth spectrum and we need additional insights about lower-income and wealthy individuals. For example, new state-level pension plans and proposed US pension legislation would expand pension coverage for more current workers.

Of course, pension design is always a topic of compelling interest to the journal’s readers. In this issue, you will find four articles that directly address aspects of pension design and two that focus on issues—the use of active mutual funds and modeling mortality—that can affect it. Marion Laboure and Juergen Braunstein, in “A Migrant Flow Pension Model for Small Open Economies,” place pension design decisions in the context of a country’s macro forces. They develop a model that forecasts pension cohorts for an aging native population, cross-border workers, and migrants using three macro scenarios for the case of Switzerland. This kind of approach is applicable to pension planning in any country, large or small, with relatively open borders, changing demographics, and future macroeconomic uncertainty.

Three other articles focus on pension design as it relates to asset allocation. One of the most pressing themes running through practitioner research is how to improve how we characterize risk for purposes of portfolio allocation in both accumulation and decumulation. Thanks to Markowitz, the tradeoff between risk (sigma) and return (mu) has long taken center stage in portfolio selection, but we know that individuals legitimately see risk, not just as volatility, but also other dimensions associated with income shortfall, asset loss, inflation, and longevity, to name a few. In their article, “Shortfall Risk and Shortfall Duration for Portfolio Choice in Decumulation,” Ganlin Xu, Harry Markowitz, and John B. Guerard, Jr. advance our understanding of the portfolio allocation problem by looking at the effects of shortfall risk versus shortfall duration on allocations. Based on differences among retirees in habit formation, they conclude that a retiree who is unable to adjust to a change in spending should choose a portfolio that reduces the risk of a shortfall while the retiree who can more easily adjust spending should choose a portfolio that reduces the duration of a shortfall.

Like Xu, Markowitz, and Guerard, Javier Estrada and Mark Kritzman are interested in improving asset allocation for retirees. In “Toward Determining the Optimal Investment Strategy for Retirement,” they show that the sustainable spending “failure rate”—the most common metric used to evaluate post-retirement consumption—itself fails to account for strategies that fail early versus those that fail later in retirement. They propose instead the “coverage ratio,” which takes into account the differential utility associated with time of failure, to determine the optimal allocation to stocks and bonds using both historical and simulated returns. Similarly, Michael W. Crook, in “Liabilities Matter: Improving Target Date Glidepath Construction through Liability Adaptive Asset Allocations,” takes on the widely used human-capital-based utility approach to designing asset allocation glidepaths. While intuitively attractive, human capital does not necessarily behave in practice as it does in theory (e.g., it is assumed to be at its maximum on the first day of an individual’s career and then declines smoothly to zero at the point of retirement). Instead, a utility approach that accounts for projected future retirement liabilities improves outcomes and the ability to better customize portfolios for groups and individuals.

While not directly focused on pension design asset allocation, active versus passive fund selection has become a hot topic due to defined contribution plan sponsor concerns about fiduciary responsibility and the possibility of legal action. While no actual legal or regulatory decision prevents the use of active funds, many have assumed that passive fund expenses and benchmark-oriented performance are more easily defended. While acknowledging the well-known finding that the average active fund trails its comparable passive peers, Atanu Saha and Heather Roberts, in “Are Actively Managed Mutual Funds Per Se Imprudent Choices for 401(k) Plans?” show that simple screens, such as expense ratio, turnover, and Sharpe Ratio, can be used to significantly improve the odds of identifying actively managed mutual funds that do outperform. This is consistent with a large academic literature that identifies fund, fund manager, and fund family characteristics that are associated with persistent superior active management.

Finally, this issue contains an article addressing the challenge of modeling mortality to support pension design in those developing countries where a lack of historical data inhibits mortality projections. Samuel E. Assabil and Don L. Mcleish, in “Mortality Modeling Using Covariates with Ghana Census Data,” develop a model that takes into account the effects of healthcare, religion, reporting rates, and other factors that can affect current mortality estimates in one country, Ghana. They then borrow mortality data from neighboring South Africa to forecast Ghanian mortality changes and show that a similar approach can be used to improve forecasts in other developing countries.

Brett Hammond

Editor

• © 2019 Pageant Media Ltd