Retirement planning

How a retiree’s health affects spending and asset allocation

October 20, 2022

Vanguard’s latest research suggests that the investment needs of most retirees in moderate to good health can be met by an asset allocation similar to that of a typical target-date fund (TDF). That’s good news for the millions of investors in TDFs and the thousands of employers who offer them as default investments in defined contribution (DC) retirement plans.

But the study also highlights that those in poor health could benefit from personalized asset allocation and advice. And all investors are well-served by factoring health care spending into their financial plans.

In one of the paper’s case studies, a hypothetical retiree—an upper-middle-class woman in poor health—would potentially be better off with an equity landing point closer to 42% to meet higher projected health expenses, especially if she lives beyond her life expectancy. However, the needs for the same retiree in moderate or good health would have been met by a more conventional asset allocation, such as a TDF that eventually falls to 30% equities. According to the comprehensive Health and Retirement Study,¹ less than 5% of the U.S. population have severe long-term care conditions that we would designate as poor health. Roughly 24% of the population are in moderate health and 72% are in good health.

Health affects optimal asset allocation in retirement

Glide paths

Notes: The case study assumes a married woman who, at age 60, has a $131,000 salary and $1.1 million in wealth. She retires at age 65 with an income of $117,000 in her final work year, $40,100 in Social Security benefits, and $92,000 as her target annual consumption amount. All figures are in 2021 dollars. Poor health means needing help with three or more activities of daily living and falling in the 95^th percentile for out-of-pocket health expenses. Those in moderate health and good health have no need for assistance in daily activities and have out-of-pocket health expenses that are in the 50^th percentile for moderate health and the 25^th percentile for good health. Our “baseline” investor has expenses similar to those for the investor in moderate health but with a longer lifespan, like the investor in good health. Average life expectancies are 69, 79, and 89 for those in poor, moderate, and good health, respectively. Please refer to the paper for more details.

Source: Vanguard calculations, based on the Vanguard Life-Cycle Investing Model.

However, the outcomes in the case studies also change depending on any of the other assumptions made for the retiree—gender, pre-retirement income, wealth, income replacement ratio, risk tolerance, and so on, said Fu Tan, investment research analyst in Vanguard’s Investment Strategy Group (ISG) and a co-author of the paper.

“As an example, if our case study subject was male and everything else was the same, the optimal equity allocation would be lower just because of the shorter life expectancy,” Tan said. “Our scenarios are built on the Vanguard Life-Cycle Investing Model [19-page research paper]. They’re examples in testing the rigors of Vanguard models and building out new features.”

Although the study focuses on the impact of health on asset allocation, for individual investors that should not be the only takeaway. Focusing on asset allocation over saving and spending might be putting the cart before the horse, said Nathan Zahm, head of goals-based investing research in ISG.

“Asset allocation is not a panacea,” Zahm said. “Especially with issues that are better addressed by saving and spending plans. Those are the primary drivers of financial solvency. Furthermore, a more aggressive asset allocation would be introducing more volatility to the portfolio—not something we would want for retirees who may not have the temperament or the finances to bear severe ups and downs in their assets. All of these should be considered in conjunction with a broader financial plan.”

For most investors, it would be premature to dismiss TDFs’ default glide paths, Zahm added. “TDFs still meet the needs of the majority of the population, but the population is heterogenous. The outliers—those who have more complex financial goals or circumstances—may need tailored advice.”

That complexity might be in the form of unique demographics, extended families, bequests, or more ambitious spending goals, not just health status.

The extra guidance could come from digital spending tools or calculators that plan sponsors could provide with TDFs in DC plans, Zahm said. Those with more complex needs might want to seek the more personalized guidance that a human advisor can provide.

“TDFs are great vehicles,” Tan said. “But the bottom line is that a default glide path alone may not be enough for some investors, and it can be well-complemented by financial planning.”

1 Based on data released in 2021 from 2010–2018 surveys conducted by the HRS, an ongoing longitudinal study that began in 1992. The HRS is sponsored by the National Institute on Aging and conducted by the University of Michigan.

Notes:

All investing is subject to risk, including the possible loss of the money you invest. Be aware that fluctuations in the financial markets and other factors may cause declines in the value of your account. There is no guarantee that any particular asset allocation or mix of funds will meet your investment objectives or provide you with a given level of income. Diversification does not ensure a profit or protect against a loss

Target date investments are subject to the risks of their underlying funds. The year in the investment's name refers to the approximate year (the target date) when an investor would retire and leave the workforce. The investment will gradually shift its emphasis from more aggressive investments to more conservative ones based on its target date. A target date investment is not guaranteed at any time, including on or after the target date.

The Vanguard Lifecycle Investing Model (VLCM) is designed to identify the product design that represents the best investment solution for a theoretical, representative investor who uses the target-date funds to accumulate wealth for retirement. The VLCM generates an optimal custom glide path for a participant population by assessing the trade-offs between the expected (median) wealth accumulation and the uncertainty about that wealth outcome, for thousands of potential glide paths. The VLCM does this by combining two set of inputs: the asset class return projections from the VCMM and the average characteristics of the participant population. Along with the optimal custom glide path, the VLCM generates a wide range of portfolio metrics such as a distribution of potential wealth accumulation outcomes, risk and return distributions for the asset allocation, and probability of ruin, such as the odds of participants depleting their wealth by age 95.

The VLCM inherits the distributional forecasting framework of the VCMM and applies to it the calculation of wealth outcomes from any given portfolio.

The most impactful drivers of glide path changes within the VLCM tend to be risk aversion, the presence of a defined benefit plan, retirement age, savings rate and starting compensation. The VLCM chooses among glide paths by scoring them according to the utility function described and choosing the one with the highest score. The VLCM does not optimize the levels of spending and contribution rates. Rather, the VLCM optimizes the glide path for a given customizable level of spending, growth rate of contributions and other plan sponsor characteristics.

A full dynamic stochastic life-cycle model, including optimization of a savings strategy and dynamic spending in retirement is beyond the scope of this framework.

Contributors