Estimating the length of retirement is one of the most important aspects in financial planning, yet many investors underestimate their average lifespan, which can have a detrimental effect on their ability to retire successfully, according to new research.
Morningstar Investment Management’s Estimating ‘The End’ of Retirement explores various factors relating to how to estimate “the end” of retirement in a financial plan using data from the Health and Retirement Study (HRS) and the Survey of Consumer Finances (SCF). While there is an expansive amount of literature examining objective mortality factors, the study notes that there is no consensus approach to estimating the length of retirement. Moreover, it’s not clear to what extent subjective estimates are reliable and how and if objective factors are correctly considered, the paper emphasizes.
Morningstar found that retirement periods based on personalized mortality factors can vary more than 15 years, resulting in significantly different estimates for required savings and/or optimal spending levels in retirement.
The study notes, for example, that the average person who overestimates his or her life expectancy would be expected to save 38% more than required, and the average person who underestimates his or her life expectancy would be expected to save 30% less than required.
In theory, retirement periods should be personalized for each client, based on that client’s facts and circumstances, considering attributes such as years until retirement, income, health status and smoker status to name a few, the study explains. Yet, Morningstar’s review of the key assumptions in 31,211 financial plans suggests financial planners are likely not personalizing mortality assumptions, given the predominant use of certain ages and the focus on using periods in multiples of five. In fact, the study found that approximately 70% of assumed retirement end ages are age 90 and 20% of end ages are age 95.
Subjective and Objective
While subjective mortality estimates are relatively accurate, on average, and households appear to do a relatively good job of considering various objective factors, such as health status, there are significant errors in individual estimates and households do not appear to correctly consider all the relevant objective factors (such as income), Morningstar notes.
For example, the study found that individuals who said they had a 0% probability of surviving to a given age (75) actually had about a 50% chance and those who said they had a 100% probability had about an 80% chance.
Morningstar notes that overall, the results imply that relying on objective factors will result in far more accurate survival estimates than subjective estimates alone. As such, financial planners should educate themselves on how to better model and personalize mortality assumptions into financial plans.
“Therefore, retirement periods should be determined using objective criteria—that is, don’t ask the client how long he/she thinks he/she will live, rather, base the estimate off objective information about the client,” writes David Blanchett, head of Retirement Research at Morningstar Investment Management and author of the report.
The assumptions used by financial advisors also do not appear to incorporate the additional “tail risk” associated with the longer potential retirement period for a married couple, the study further observes. As such, Morningstar suggests that the length of retirement should be determined considering the shortfall aversion metric (for example, probability of success) to ensure recommendations are not overly conservative.
Modeling the End
For a single household, Morningstar suggests that adding five years to a personalized life expectancy estimate would be a way to estimate an appropriate retirement period. For joint households, the study recommends adding eight years to the longest life expectancy of the two members, or three years if the retirement period considers the actual joint survival probabilities. According to Blanchett, this approach suggests a retirement period of 30 years (to age 95) is a reasonable assumption for the average 65-year-old male/female couple retiring today.
The retirement end age assumptions should also be reviewed regularly to ensure they are timely and reflect all available information about both the client and mortality tables, the study further emphasizes.
“In summary, forecasting ‘the end’ of retirement is a complex decision,” says Blanchett. “There is no one single ‘right’ way to do it; however, it is important that estimates be personalized to household attributes and that the assumed length is determined in light of other modeling assumptions, such as the probability of success, to ensure an appropriate value is selected.”
