Life Expectancy vs. Lifetime Planning Period

The inspiration for this post was a recent conversation I had with my buddies at our weekly R.O.M.E.O (Rossmoor Old Men Eating Out) lunch and gab session. The exciting topics we discussed included:

• How much longer can we expect to live?
• How much longer should we plan on living? and
• What is the likelihood that the President of the United States will die in office during the term starting January 20, 2025 given the two likely candidates at this time.

This post will briefly discuss these topics and, as usual, we will use the Actuaries Longevity Illustrator to quantify longevity expectations. For more geeky discussion of the Actuaries Longevity Illustrator and how to use it in your retirement planning, you may want to revisit our following posts:

November 17, 2021—Using the Actuaries Longevity Illustrator in Your Retirement Planning

September 27, 2021—Don’t Know How Long You’ll Live in Retirement? Another Good Reason to Build a Robust Floor Portfolio to Fund Your Essential Expenses

September 14, 2021—How Long Should You Plan to Live?

The numbers provided in this post are based on the current version of the Actuaries Longevity Illustrator (ALI), which is due to be changed any day to reflect revised mortality assumptions used in this year’s (2023) Social Security Trustees Report. As noted in prior posts, the results shown in the ALI are rounded to the nearest year, and this rounding can create minor discontinuities. 

How much longer can we expect to live?

Table 1 below show the probability of survival (in years) for male and female non-smokers of various ages who are assumed to be in excellent health. The 50% probability of survival is also known as remaining life expectancy (in years for Table 1 and age for Table 2). The 75% probability of survival means that out of 100 individuals of that current age today, 75 are expected to survive the period shown, and the 25% probability of survival means that out of 100 individuals that age today, only 25 are expected to survive the period shown.

Table 1--Probability of Survival (in years) for Male and Female Non-Smokers in Excellent Health

	Male				Female
Current Age	75% Prob.	50% Prob.	25% Prob.		75% Prob.	50% Prob.	25% Prob.
60	20	28	34		23	30	36
65	16	23	29		19	25	31
70	12	18	24		14	20	26
75	8	14	19		10	15	21
80	5	9	14		6	11	16
85	3	6	10		4	7	11
90	1	4	7		2	4	8

Source: Actuaries Longevity Illustrator 7/14/2023

For those who expect to live longer (or shorter) than the average U.S. person in excellent health, you can adjust the results of Table 1 by subtracting (or adding) years to your current age. For example, an aged 65-year-old male who expects to live as long as a 63-year-old male may conclude that his remaining life expectancy (50% probability) is 25 years rather than 23. This is approximately the same effect as if he assumed he had the same remaining life expectancy as a female his own age.

Table 2 below adds the years in Table 1 to the current age and shows expected age at death for the same probabilities of survival.

Table 2—Expected Age at Death for Various Probabilities of Survival for Non-Smokers in Excellent Health

	Male				Female
Current Age	75% Prob.	50% Prob.	25% Prob.		75% Prob.	50% Prob.	25% Prob.
60	80	88	94		83	90	96
65	81	88	94		84	90	96
70	82	88	94		84	90	96
75	83	89	94		85	90	96
80	85	89	94		86	91	96
85	88	91	95		89	92	96
90	91	94	97		92	94	98

Source: Actuaries Longevity Illustrator 7/14/2023

The key take-aways from these tables include:

• For any given individual in excellent health, there is a wide range of years that you may actually live into the future. For example, if you are currently an age 70 male, you have a 25% probability that you will die prior to reaching age 82, a 50% probability that you will die after attaining age 82 and before reaching age 94 and a 25% probability that you will die after attaining age 94. 
• This wide range makes it difficult to plan for retirement. There are non-zero probabilities of death (some of them bigger than others) in each of the next 30+ years for a 70-year-old male.
• As we age, our expected age at death increases somewhat, but not very much until we get quite old (more so for the 50% probability than for the 75% probability of survival).
• When planning, it is a mistake, in our opinion, to focus too heavily on life expectancy. 50% of the relevant population can be expected to live longer than their life expectancy.

This leads us to the question of

How much longer should we plan on living?

At How Much Can I Afford to Spend in Retirement, we recommend, as the default assumption in our spreadsheets, that users assume a lifetime planning period based on the 25% probability of survival for non-smokers in excellent health from the ALI. The default assumption (and other default assumptions about the future) may be overridden if desired. This assumption is more conservative (by about 5 or 6 years) than assuming a lifetime planning period based on life expectancy, but less conservative than assuming a lifetime planning period based on a 10% (or less) probability of survival. Using the 25% probability may result in relatively minor experience losses when users reach their mid-80s, but these losses can be offset by gains from other sources (such as reductions in planned spending at those later ages). 

Of course, users can easily see the impact on their current funded status (ratio of their assets to their spending liabilities) of assuming shorter or longer longevity planning periods. As discussed in earlier posts, increasing the assumed lifetime planning period may have almost no impact on the household funded status if the household has invested sufficient assets in protected lifetime income such as Social Security, pensions and life annuities to cover their essential expenses.

We also note that if you consider yourself extremely healthy and favor using lifetime planning periods in excess of the 25% probability of survival years default assumption, you definitely should be considering purchasing a lifetime income annuity, as they are priced by insurance companies assuming closer to 50% probability of survival.

What is the likelihood that either Biden or Trump will die in office if reelected for the next term?

We attempt to answer this question using the ALI and by making the following additional assumptions:

• They both survive until January 20, 2025 and one of them remains or takes over again as President
• Their mortality follows the current ALI mortality for non-smoking males in excellent health—no adjustments for their current physical conditions, extra medical attention they would get as presidents, extra stress they might be under as presidents, etc., etc., etc.

Biden’s date of birth is November 20, 1942 and Trump’s is June 14, 1946. So, as of January 20, 2025, Biden would be 82 years and two months old and Trump would be 78 and seven months old.

Eyeballing the “Probability of Living for a Specified Number of Years” graph in the ALI for a current 82-year-old shows that Biden would have about an 80% probability of surviving four years, or a 20% probability of not surviving for four years from his inauguration.

Similarly, we estimate Trump would have approximately an 86% probability of surviving four years, or a 14% probability of not surviving for four years from his inauguration.

Summary

Determining probabilities of survival is what actuaries do for a living, but it is important to note that pieces of you won’t die every year. Living is a binary process that involves either living or dying each year. Determining the year or range of years you might die is a difficult (if not impossible) exercise that makes planning for retirement difficult. As with other assumptions about the future, you aren’t likely to get this one exactly right. That is why it is important to have a process like the actuarial process discussed in our post of June 11, 2023 for keeping your plan on track when actual experience deviates from assumptions you make.