Editor’s Letter

We are pleased to publish five articles focused on asset accumulation and drawdown over the lifecycle. Reflecting the goal of The Journal of Retirement, the articles contain actionable recommendations for policy advisors and household financial advisors.

The groundbreaking Chilean pension reform of 1981 has attracted much attention and academic research. However, research has focused on the economics of the system, not on the interplay between the politics and the economics. A funded system creates a pool of assets that is a tempting target for politicians to loot. Whereas most countries responded to the COVID crisis by borrowing to provide support to the unemployed, the Chilean government took the route of allowing participants to access their individual pension accounts. And the Chilean government has come under pressure recently to backtrack on the 1981 reform altogether.

Marcela Parada-Contzen’s article “The Characterization of Population Preferences and Assessments of Retirement Systems: Evidence from Chile” analyzes the attitudes of Chileans toward their pension system. It is significant because it offers an explanation of why the Chilean government acted as it did and why other governments—for example Poland and Argentina—have been able to backtrack on reforms. The shocking feature of the Parada-Contzen analysis is the extremely low level of support for individual accounts. The lesson is that the system will remain politically vulnerable unless leaders take steps to improve its effectiveness—for example by including workers in Argentina’s large informal sector—and fostering a sense of ownership.

The article “Optimal Claiming of Social Security Benefits” by Steven Diamond, Stephen Boyd, David Greenberg, Mykel Kochenderfer, and Andrew Ang investigates the optimal age to claim Social Security. Although many workers appear to adopt a default position of claiming Social Security immediately on retirement, those with financial assets have the option of delaying claiming and using financial assets to bridge the gap between retirement and delayed claiming of an enhanced benefit. This decision is often framed in terms of a breakeven age—how long do I have to live to recoup the benefits foregone? But breakeven analysis is flawed because it ignores the value of the additional longevity insurance acquired because of delay. The Diamond article analyzes the decision, taking account of the value of this insurance, and concludes that almost all single individuals should delay claiming until at least their late 60s. The article defers an analysis of married couples to future research. Married couples face different considerations, and it will often be optimal for the lower earner to claim as early as possible.

The journal always welcomes articles that are both provocative and well-reasoned, and we are particularly pleased to publish “The Life-Cycle Model Implies that Most Young People Should Not Save for Retirement” by Jason S. Scott, John B. Shoven, Sita N. Slavov, and John G. Watson. The article argues, based on a life-cycle model, that most young people should delay saving for retirement. Folk wisdom argues that young people should save for retirement because retirement plan contributions made at younger ages will have more years to accumulate. But this folk wisdom ignores the fact that plausible parameterizations of the life cycle model assume that individuals have a positive rate of time preference, i.e., they prefer consumption now to consumption later. If the rate of interest equals the rate of time preference, the benefit of more years of investment returns is precisely offset by the cost of more years of time-discounting.

At their most basic, life-cycle models posit that individuals should shift consumption from periods when incomes are higher and the marginal utility of consumption (the satisfaction created by an additional unit of consumption) is correspondingly lower (typically when working) to periods when incomes are lower and the marginal utility of consumption is correspondingly higher (typically in retirement). In other words, nobody wants to live high on the hog when working and then be poor in retirement. However, workers at the start of their careers often have low earnings, are burdened by student loans, and may be saving for a house deposit. Thus, they have higher marginal utility of consumption than commonly believed and should not be saving for retirement. As we know, life-cycle models are computationally challenging and, of necessity, involve simplifications. The Scott model abstracts from financial and labor market risk that might incent precautionary saving and potentially overturn their finding. We hope that, in a future article, the authors will investigate whether their conclusions are robust to the incorporation of such risk.

In another use of a life-cycle model, Ankul Daga, Timothy Smart, and David C. Pakula take on the computational complexity of addressing multiple investor goals, something they point out that practical applications have previously not attempted. Adopting the concept of a goal fulfillment gap, they use stochastic dynamic programming to quantify the value added by a strategy optimized across multiple goals to illustrate both optimal consumption and asset allocation in comparison with previous solutions. Like the Scott article, this represents a step forward in practitioners’ ability to apply systematic techniques to the fundamental retirement savings problem. In this case, it would be interesting to see them take on two additional challenges: (1) finding allocation solutions at the sub-asset class level (beyond equities and fixed income) and (2) incorporating utility functions that better reflect evidence on investors’ behavior in assessing risk and probability.

One of the most puzzling aspects of the annuity puzzle is the lack of demand for deferred income annuities, which are annuities purchased at retirement or even earlier, with payments commencing at some advanced age (thus insuring households against the risk of living unusually long). Annuities can offer higher income than similar unannuitized investments because of mortality credits, the reallocation of wealth from those who die to those who survive. Mortality credits increase with age because fewer survive to advanced ages, making deferred annuities particularly attractive. Deferred income annuities also have the practical advantage that only a small fraction of the household’s financial assets need be allocated to them, thus enabling the household to preserve liquidity.

But households need to set aside assets to finance consumption from retirement to the age at which the deferred income annuity payments commence. The article “Adaptive Retirement Planning, Sustainable Withdrawals, and Deferred Annunities” by Anran Chen, Steven Haberman, and Stephen Thomas investigates the optimal allocation of those assets. They find the optimal equity allocation is surprisingly high, sometimes close to 100 percent. We realize that currently available deferred income annuities are less than perfect—no currently available products offer inflation protection or exposure to equity returns. But we hope this paper will encourage financial advisors to explore whether they have a place in their clients’ financial plans.

Brett Hammond

Co-Editor

Anthony Webb

Co-Editor

• © 2023 Pageant Media Ltd