A post making the rounds on retirement social media sites makes the claim that retiring in your late 50s or early 60s means you’ll live a lot longer than those who work into their late 60s.
Some noted the implausibility of the data, but many took the comments at face value. Most of the comments are from those who searched their memories for friends or relatives who worked too long at a job they hated and died before they could enjoy their retirement. People, especially those who retired early, love the idea that the research confirmed the wisdom of their decision.
Despite being wrong, the popularity of the findings presents an opportunity to discuss what science says about retirement age and longevity, and perhaps most importantly about the natural human tendency to embrace information that makes us feel good.
Diving Into the Data
When the table popped on the Facebook feed of a popular financial planning podcaster a week ago, I asked David Blanchett, head of retirement research at PGIM, to run the data on longevity and age at retirement. These types of questions can best be answered by using a big, nationally representative panel data set such as the Health and Retirement Study.
Blanchett focused on retirement status in 2012 and whether the retiree was still alive in 2022. Since the mortality distribution is like a bell curve, a higher survival rate over 10 years implies a greater average longevity in the long run. Here is how many 65-year-olds in 2012 who were working, semi-retired or fully retired were still alive in 2022.
Percent of 65-Year-Olds Still Alive 10 Years Later
The actual data paints a very different story about whether working into old age will kill you. Only 73% of men who are retired by age 65 are still alive 10 years later, and 83% of retired women are still alive.
“The differences in life expectancy are relatively staggering," Blanchett notes. "A 90% survival probability to age 75, when age 65, implies a life expectancy of about age 90 (or 25 years) versus a life expectancy of only age 80 (or about 15 years) for men who are fully retired.”
So a 65-year-old retired man can on average expect to live to about 80 while a 65-year-old who is still working can expect to live to age 90.
This may sound like bad news for the worker who decided to retire at age 60. But this isn’t the whole story. Many who retire early do so because their health won’t allow them to keep working. A heart attack or cancer diagnosis may force a worker into early retirement, and these unfortunate early retirees will reduce the average longevity of all early retirees.
Blanchett and I have found in past research that self-assessed health status is a surprisingly good predictor of how long people actually live in the HRS. According to the HRS, 55% of men who are still working at the age of 65 consider themselves to be in “excellent” or “very good” health compared to just 40% of men who are retired at age 65. This confirms our suspicion that early retirees tend to be significantly less healthy than those who continue working.
We can sort people into groups based on health status in 2012 and see how many are still alive in 2022 by their age in 2012. This will allow us to see whether a healthy worker who decides to retire early will live about as long as a healthy worker who delays retirement.
If we focus on the difference in 10-year longevity between retirees and workers, we see that those who continue working live longer than those who retire within each health status category.
The difference in longevity between workers and early retirees is pretty modest (6%) among those in great (very good or excellent) health, while the difference is significantly larger for those who were in good health in 2012 (13% longer for workers) and those who considered themselves in not good (fair or poor) health in 2012 (24% longer).
The likely explanation is that the retired workers who considered themselves to be in “fair” or “poor” health in 2012 were in worse shape on average than those who could physically still work but rated themselves in the same “not good” category. In other words, deciding not to retire early isn’t going to reduce your expected longevity but it probably isn’t going to extend it by that much either.
Why Lies Go Viral
The idea that waiting to retire led to much lower lifespans was a powerful message to many, especially in online communities of retirees. People like it when data confirms their beliefs, and many who retire early want to believe that they made the right choice.
This is especially true in a society that elevates hard work and productivity — those who retire early may feel some insecurity about their decision to leave the workforce. They were eager to believe data that explained why they made the right choice.
Confirmation bias is one of the strongest and most consistent biases in psychology. Because changing our beliefs takes so much mental energy and may be unpleasant, we tend to be most receptive to information consistent with our own view of the world and skeptical of information that challenges it.
One of the most interesting aspects of the social media discussions on the fake longevity table was the discussion among financial professionals. More than a few financial planners reached into their memories for examples of clients who waited too long to retire and died before they could enjoy their wealth, or retirees who decided to leave a good-paying job to enjoy life in their 60s.
Part of our job as professionals is to recognize our own susceptibility to confirmation bias. When a patient hears medical misinformation, their doctor needs to be able to provide an unbiased viewpoint based on medical science.
Likewise, it is a planner’s job to steer clients away from the abundant financial misinformation peddled on social media toward information that is accurate and consistent with the science. Clients are counting on professionals to protect them from appealing but incorrect information that could lead to bad decisions.
Michael Finke, Ph.D., CFP, is a professor of wealth management and research fellow at the Retirement Income Institute.
© Arc, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to TMSalesOperations@arc-network.com. For more information visit Asset & Logo Licensing.