November 23, 2021
Planning for retirement can be challenging, in part because no one knows with certainty how long they will live or what their health care costs will be.
Financial advisors and financial planning tools tend to rely on conservative longevity assumptions and a fixed amount for out-of-pocket health care spending, but there might be a better approach.
In a recently published paper, On Track for Success? Insights from Stochastic Health Care Modeling, Vanguard retirement research analysts Fu Tan and Timothy Smart show how a personalized, realistic assessment of these two key components can provide a better foundation for retirement planning.
“While longevity and health costs aren’t topics that we want to think about often,” Ms. Tan said, “the more we can do to understand their uncertainty, the greater confidence we will have to deal with any surprises that arise during retirement.”
Longevity: It’s possible to be too cautious
Erring on the conservative side with your retirement calculations might sound prudent, especially as you don’t get a second chance to get it right. That approach, though, can result in an unrealistically modest assessment of whether your savings will last through retirement. You might feel compelled to save more or resign yourself to a more frugal lifestyle.
Take the question of longevity. Some financial advisors and financial planning tools base their projections on a life expectancy of 100 years. A hypothetical example in the On Track for Success? paper used that projection in a hypothetical example for a healthy 65-year-old woman, funding her retirement through a 6% fixed dollar withdrawal rate on her savings of $500,000. (The 6 percent rate was chosen to illustrate the sensitivity of the models used and does not represent a recommendation for a sustainable spending strategy.) That projection resulted in a success rate—the probability that the woman’s savings would last her through retirement—of just 33.6%. But when the analysts used Social Security Administration data that project an average life expectancy of 86.8 years for a woman her age, the success rate for the same spending strategy increased to 71.3%. (The SSA data show that men tend to have shorter life expectancies and higher success rates than women of comparable age.)
The difference in those success rates suggests that using population-based mortality tables like the SSA’s can provide a more personalized projection of a retiree’s future and a better estimate of their success rate.
Health care costs and the problem with averages
Common retirement planning practice uses estimates for out-of-pocket health care costs that are fixed over time. The paper acknowledges that this approach provides a reasonable input for gauging the success rate of a retirement portfolio. At the same time, that average figure obscures the challenges that fluctuations in out-of-pocket health care costs can pose to a retiree using a fixed spending strategy.
Starting with the same parameters for the hypothetical example mentioned above, the estimated health care costs in retirement for a healthy 65-year-old woman would be calculated at an annual fixed amount of $3,000. Research described in the paper, however, estimates that her out-of-pocket health care costs could range from $1,605 to $18,046 in a given year. There’s a chance she could spend a little less than the fixed annual average of $3,000, but she could also end up having to spend as much as $15,000 more than that—a sizable chunk of her annual spending target that would require her to cut spending significantly in other areas to compensate.
This exercise underscores the need to factor into retirement planning the reality that health care costs might average a certain amount over the long haul but will vary significantly from year to year.
Better projections should lead to better planning
Many investors and advisors now test their retirement plans’ viability using simulations that reflect asset-return uncertainty instead of simply counting on a fixed average annual return from the financial markets.
Richer models that can incorporate sources of uncertainty such as one’s health status and longevity in a personalized way mark another advance in retirement planning. Researchers and advisors developing these capabilities further will help provide retirees and those planning for retirement with deeper insights into longevity risks and health care costs and the strategies they can use to manage them.