Updated longevity calculator - from the actuaries

Non-investing personal finance issues including insurance, credit, real estate, taxes, employment and legal issues such as trusts and wills
Post Reply
User avatar
Topic Author
Stinky
Posts: 2849
Joined: Mon Jun 12, 2017 11:38 am
Location: Sweet Home Alabama

Updated longevity calculator - from the actuaries

Post by Stinky » Fri Oct 11, 2019 4:27 pm

Below is a link to the updated longevity calculator, prepared by the Social of Actuaries and the American Academy of Actuaries (two of the professional actuaries bodies in the US. It allows calculations for either single or joint lives.

This calculator is more useful, for most folks, than using information from the Social Security Administration. The SS tables use "population" mortality, which is based upon the broad population, while this table allows users to specify their general health and whether or not they smoke. Smoking is huge contributor to early deaths.

https://www.longevityillustrator.org/
It's a GREAT day to be alive - Travis Tritt

User avatar
Cyclesafe
Posts: 1006
Joined: Wed Dec 31, 2014 1:03 pm

Re: Updated longevity calculator - from the actuaries

Post by Cyclesafe » Sat Oct 12, 2019 9:41 am

I found this calculator very helpful. Thanks for posting!
"Plans are useless; planning is indispensable.” - Dwight Eisenhower

User avatar
2pedals
Posts: 989
Joined: Wed Dec 31, 2014 12:31 pm

Re: Updated longevity calculator - from the actuaries

Post by 2pedals » Sat Oct 12, 2019 9:53 am

Very useful, thanks for posting.

Ferdinand2014
Posts: 1014
Joined: Mon Dec 17, 2018 6:49 pm

Re: Updated longevity calculator - from the actuaries

Post by Ferdinand2014 » Sat Oct 12, 2019 10:15 am

As a patient once said to me when I asked how he was doing, “Well, I get up every day and read the newspaper. If my name isn’t in the obituary, I get to work.” He is still reading the newspaper and he’s over 90.
“You only find out who is swimming naked when the tide goes out.“ — Warren Buffett

User avatar
Hayden
Posts: 1266
Joined: Tue Apr 16, 2013 5:13 pm

Re: Updated longevity calculator - from the actuaries

Post by Hayden » Sat Oct 12, 2019 10:50 am

Thanks for posting. That was helpful

peter_s
Posts: 28
Joined: Mon Jul 23, 2007 3:56 pm

Re: Updated longevity calculator - from the actuaries

Post by peter_s » Sat Oct 12, 2019 11:13 am

Very clear and concise. Should be very useful for focusing discussion on difficult topics.
Thank you.

User avatar
Wildebeest
Posts: 1204
Joined: Fri Dec 27, 2013 2:36 pm

Re: Updated longevity calculator - from the actuaries

Post by Wildebeest » Sun Oct 13, 2019 7:29 am

Thanks for posting.

I especially like the feature that both of a couple are still expected to be alive in the future.

We were celebrating our wedding anniversary a couple of weeks ago and I decided to google the chance that we would be celebrating our anniversary at age 80, 90 and 95. There was no easy calculator to be found and I ended up having to work it out myself the next day while being confident but not sure I had figured it out correctly.

This calculator would have given me the answer in less than a minute.
The Golden Rule: One should treat others as one would like others to treat oneself.

HoosierJim
Posts: 711
Joined: Wed Mar 24, 2010 7:11 pm

Re: Updated longevity calculator - from the actuaries

Post by HoosierJim » Sun Oct 13, 2019 7:47 am

What if you retire before age 60? Warning does not match real limit.
Image

User avatar
Topic Author
Stinky
Posts: 2849
Joined: Mon Jun 12, 2017 11:38 am
Location: Sweet Home Alabama

Re: Updated longevity calculator - from the actuaries

Post by Stinky » Sun Oct 13, 2019 8:00 am

HoosierJim wrote:
Sun Oct 13, 2019 7:47 am
What if you retire before age 60? Warning does not match real limit.
Image
It looks to me like your "retirement age" must be equal to or greater than your current age.

If, for example, you are 65 and you retired at 60, I would just put "65" for retirement age. The results should be reliable.
It's a GREAT day to be alive - Travis Tritt

User avatar
Topic Author
Stinky
Posts: 2849
Joined: Mon Jun 12, 2017 11:38 am
Location: Sweet Home Alabama

Re: Updated longevity calculator - from the actuaries

Post by Stinky » Sun Oct 13, 2019 8:01 am

Wildebeest wrote:
Sun Oct 13, 2019 7:29 am
Thanks for posting.

I especially like the feature that both of a couple are still expected to be alive in the future.

We were celebrating our wedding anniversary a couple of weeks ago and I decided to google the chance that we would be celebrating our anniversary at age 80, 90 and 95. There was no easy calculator to be found and I ended up having to work it out myself the next day while being confident but not sure I had figured it out correctly.

This calculator would have given me the answer in less than a minute.
Sometimes actuaries can produce useful things! :D
It's a GREAT day to be alive - Travis Tritt

User avatar
WoodSpinner
Posts: 1013
Joined: Mon Feb 27, 2017 1:15 pm

Re: Updated longevity calculator - from the actuaries

Post by WoodSpinner » Sun Oct 13, 2019 8:16 am

It’s a great tool ...

BUT would love some more guidance in figuring how the overall Health questions. A bit subjective with significant longevity implications.

WoodSpinner

EnjoyIt
Posts: 3211
Joined: Sun Dec 29, 2013 8:06 pm

Re: Updated longevity calculator - from the actuaries

Post by EnjoyIt » Sun Oct 13, 2019 8:54 am

Nice link. Thanks.
My risk of living past 95 is 12%
My risk of a 4% withdrawal rate failing in 5%
That means my risk of living past 95 and running out of many is 0.6%

I like those odds.

User avatar
Topic Author
Stinky
Posts: 2849
Joined: Mon Jun 12, 2017 11:38 am
Location: Sweet Home Alabama

Re: Updated longevity calculator - from the actuaries

Post by Stinky » Sun Oct 13, 2019 9:49 am

WoodSpinner wrote:
Sun Oct 13, 2019 8:16 am
It’s a great tool ...

BUT would love some more guidance in figuring how the overall Health questions. A bit subjective with significant longevity implications.

WoodSpinner
Absent other guidance on the site, think of your 10 best friends.

Is your health in the top third, middle third, or bottom third of those 10? Fill out your status accordingly.
It's a GREAT day to be alive - Travis Tritt

User avatar
nisiprius
Advisory Board
Posts: 39764
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Updated longevity calculator - from the actuaries

Post by nisiprius » Sun Oct 13, 2019 11:01 am

In the search for certainty, it is easy to pay too much attention to averages and not enough to dispersion. Alas (or perhaps fortunately????) you do not know when you will die, and all of the longevity calculators are giving you relatively small differences in averages, which are on top of relatively large dispersions.

This isn't to say you should ignore them, or that smoking doesn't matter. But don't take them literally, either. Don't think of your life expectancy as "24 years" when the truth is probably more like "24 ± 12."

Yes, smoking is a powerful predictor of longevity, probably the most powerful of all the factors used by longevity calcuators. And, yes, each of my parents smoked three packs a day. I don't smoke. And I have currently outlived each of them by more than ten years, so, sure it matters. Don't smoke.

Now, imagine John Doe and Richard Roe, born 4/1/1956, retiring at age 65, male, in excellent health. The only difference is that John is a smoker and Richard is not. The statistics say that on the average, we expect Richard to live eight years longer than John.

But before continuing, I would like you to guess. Despite smoking, what is the probability that John will outlive Richard?

Here are the curves from the calculator.

Image

Yes, there is a huge difference. It is a median difference of about 8 years (green and grey curves, read at the 50% probability).

However, each of them is facing an uncertainty of--depending on how you decide to figure the range--around 25 years in their number of years in retirement. "Smoking" determines maybe eight years' worth. "The breaks," apart from smoking, twice as much.

Smoker John, despite smoking, could easily (18% probability) live over 25 years in retirement. Nonsmoker Richard, despite not smoking, could easily (20% probability) die within 15 years.

Now, I had to do a little spreadsheet work on this, but, assuming independence, by my calculations...

...there is a 30% chance of smoker John outliving nonsmoker Richard.




Appendix:

I used the first table on their results page. I took differences to get the probability of dying within each five-year bucket; for example, a 20% chance John lives 70-75 years. a 5% chance Richard lives 65-70 years. I created 64 categories for each possible combination; for example, John 70-75 and Richard 65-70 is one of them. I multiplied the numbers to get 64 probabilities. For example, a chance of 20% x 5% = 1.0% that John lives 70-75 years and Richard lives 65-70 years. I then classified each of the 64 probabilities as to whether it is one in which John outlives Richard, Richard outlives John, or one in which they both die within the same five-year period. If they die within the same period, I figured it was a 50% chance of either outliving the other. I then counted all the categories in which John outlives Richard, plus half the sum of the categoreis in which they die within the same 5-year period.

Image
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

TheNightsToCome
Posts: 467
Joined: Fri Jun 30, 2017 11:48 pm

Re: Updated longevity calculator - from the actuaries

Post by TheNightsToCome » Sun Oct 13, 2019 11:29 am

EnjoyIt wrote:
Sun Oct 13, 2019 8:54 am
Nice link. Thanks.
My risk of living past 95 is 12%
My risk of a 4% withdrawal rate failing in 5%
That means my risk of living past 95 and running out of many is 0.6%

I like those odds.
"My risk of a 4% withdrawal rate failing in 5%"

That might be true if current stock and bond valuations were equal to historical averages. However, they are much higher, and thus prospective returns are much lower (than the historical average) and SOR risk is higher.

If we had 100+ data points conditional on today's valuations, it's highly likely that the 4% withdrawal rate would have failed (much) more often than 5%.

JackoC
Posts: 1009
Joined: Sun Aug 12, 2018 11:14 am

Re: Updated longevity calculator - from the actuaries

Post by JackoC » Sun Oct 13, 2019 12:13 pm

I agree it's not that directly meaningful for retirement financial planning to look at life *expectancy* 50% tile. That only directly allows you to judge if strategies are more or less likely than 50-50 to succeed.

You need the whole distribution, in which case if it's accurate you don't really have to worry that much about dispersion. What's the 95%-tile (or whatever you choose as acceptable bound for running out of money due to age)? That's what you actually want to know.

Raw SSA numbers aren't useful because the US population is too demographically stratified for those numbers to be useful for any given person. Same reason you wouldn't use a whole world longevity table, it includes too many people in too much different overall life situations than yours.

Smoking and general health (in one's 60's let's say) are two big factors to make the numbers more individually relevant. Because besides being important direct factors in longevity in a given demographic group, they are also proxies for which demographic group you're in (poorer and less educated people are more likely to be smokers though also tend to have more other risks of earlier death than higher socio-econ people, etc.). Although other online calculators which explicitly include long lists of questions about education, exercise, family history etc. give more varying results. The problem is most of those only give the 50%-tile. Some I've seen give 25%/75%, but constructing complete valid distributions including all those factors is probably difficult.

My own answers in the more detailed ones have yielded 50%-tile more like low-mid 90's (male in 60's), this gives around 87. Again that's not a big difference in LE compared to the dispersion, but the 50%-tile isn't by itself that relevant. More important for financial planning purposes would be the % likelihood changes of living to a given advanced age, say 95. This one says I have 15-22%* chance of reaching 95, but some of the more complicated ones give 95 as my 50%-tile. That's a pretty big difference.

I think one should probably look at annuities if the differences in calculators makes a big difference in likelihood of running out of money.

*'average' v. 'excellent' health. I find that question actually harder to answer than a series of what particular medical problems I or relatives have or had, level of exercise, diet, the components of health if you will.

User avatar
dm200
Posts: 22923
Joined: Mon Feb 26, 2007 2:21 pm
Location: Washington DC area

Re: Updated longevity calculator - from the actuaries

Post by dm200 » Sun Oct 13, 2019 1:34 pm

Stinky wrote:
Fri Oct 11, 2019 4:27 pm
Below is a link to the updated longevity calculator, prepared by the Social of Actuaries and the American Academy of Actuaries (two of the professional actuaries bodies in the US. It allows calculations for either single or joint lives.
This calculator is more useful, for most folks, than using information from the Social Security Administration. The SS tables use "population" mortality, which is based upon the broad population, while this table allows users to specify their general health and whether or not they smoke. Smoking is huge contributor to early deaths.
https://www.longevityillustrator.org/
This calculator shows my probability of living a fairly long life (nearest age is now 74) as lower than I expect.

EnjoyIt
Posts: 3211
Joined: Sun Dec 29, 2013 8:06 pm

Re: Updated longevity calculator - from the actuaries

Post by EnjoyIt » Sun Oct 13, 2019 3:58 pm

TheNightsToCome wrote:
Sun Oct 13, 2019 11:29 am
EnjoyIt wrote:
Sun Oct 13, 2019 8:54 am
Nice link. Thanks.
My risk of living past 95 is 12%
My risk of a 4% withdrawal rate failing in 5%
That means my risk of living past 95 and running out of many is 0.6%

I like those odds.
"My risk of a 4% withdrawal rate failing in 5%"

That might be true if current stock and bond valuations were equal to historical averages. However, they are much higher, and thus prospective returns are much lower (than the historical average) and SOR risk is higher.

If we had 100+ data points conditional on today's valuations, it's highly likely that the 4% withdrawal rate would have failed (much) more often than 5%.
Valuations were much higher around the late 90s early 2000s and those folks appear to be doing just fine today. Valuations help, but they don't predict the future. If you had 100+ data points, you still could not predict future returns. Even if the odds of failure at 95 is 20% then my risk of running out of money and living longer than 95 is 2.4%. Guess what, I still like those odds.
If we add in that retirees spend less and less as they get older, my odds improve. If you add in that I am not a robot and will make some adjusts in my spending if the markets acts atrociously my odds of failure drop well below a fraction of a fraction of a percent. Guess what, I still like those odds.

OffGridder
Posts: 51
Joined: Thu Jul 23, 2015 8:03 am
Location: Eastern WA.

Re: Updated longevity calculator - from the actuaries

Post by OffGridder » Sun Oct 13, 2019 5:18 pm

nisiprius wrote:
Sun Oct 13, 2019 11:01 am
In the search for certainty, it is easy to pay too much attention to averages and not enough to dispersion. Alas (or perhaps fortunately????) you do not know when you will die, and all of the longevity calculators are giving you relatively small differences in averages, which are on top of relatively large dispersions.

This isn't to say you should ignore them, or that smoking doesn't matter. But don't take them literally, either. Don't think of your life expectancy as "24 years" when the truth is probably more like "24 ±12

Image
If you want to see the sobering reality of "dispersion" without the deep statistical and actuarial analysis, just start reading the daily obituaries in your newspaper.
"Goodness is the only investment that never fails." | H.D. Thoreau

TheNightsToCome
Posts: 467
Joined: Fri Jun 30, 2017 11:48 pm

Re: Updated longevity calculator - from the actuaries

Post by TheNightsToCome » Sun Oct 13, 2019 6:14 pm

EnjoyIt wrote:
Sun Oct 13, 2019 3:58 pm
TheNightsToCome wrote:
Sun Oct 13, 2019 11:29 am
EnjoyIt wrote:
Sun Oct 13, 2019 8:54 am
Nice link. Thanks.
My risk of living past 95 is 12%
My risk of a 4% withdrawal rate failing in 5%
That means my risk of living past 95 and running out of many is 0.6%

I like those odds.
"My risk of a 4% withdrawal rate failing in 5%"

That might be true if current stock and bond valuations were equal to historical averages. However, they are much higher, and thus prospective returns are much lower (than the historical average) and SOR risk is higher.

If we had 100+ data points conditional on today's valuations, it's highly likely that the 4% withdrawal rate would have failed (much) more often than 5%.
Valuations were much higher around the late 90s early 2000s and those folks appear to be doing just fine today. Valuations help, but they don't predict the future. If you had 100+ data points, you still could not predict future returns. Even if the odds of failure at 95 is 20% then my risk of running out of money and living longer than 95 is 2.4%. Guess what, I still like those odds.
If we add in that retirees spend less and less as they get older, my odds improve. If you add in that I am not a robot and will make some adjusts in my spending if the markets acts atrociously my odds of failure drop well below a fraction of a fraction of a percent. Guess what, I still like those odds.
"Even if the odds of failure at 95 is 20% then my risk of running out of money and living longer than 95 is 2.4%. Guess what, I still like those odds."

This is probably closer to our current circumstance.

"Valuations help, but they don't predict the future."

They do predict the future, just not with precision:

Larry Swedroe alluded to a paper by Cliff Asness (Swedroe here: https://www.etf.com/sections/index-inve ... nopaging=1 and Asness here: file:///C:/Users/Curt/Downloads/An%20Old%20Friend%20The%20Stock%20Markets%20Shiller%20PE%20(1).pdf ):

“In a November 2012 paper, “An Old Friend: The Stock Market’s Shiller P/E,” Asness, of AQR Capital Management, found that the Shiller CAPE 10 provides valuable information. Specifically, he found 10-year-forward average real returns drop nearly monotonically as starting Shiller P/Es increase.

He also found that, as the starting Shiller CAPE 10 ratio increased, worst cases became worse, and best cases became weaker. Additionally, he found that, while the metric provided valuable insights, there were still very wide dispersions of returns. For instance:

When the CAPE 10 was below 9.6, 10-year-forward real returns averaged 10.3%. In relative terms, that is more than 50% above the historical average of 6.8% (9.8% nominal return less 3.0% inflation). The best 10-year-forward real return was 17.5%. The worst 10-year-forward real return was still a pretty good 4.8%, just 2.0 percentage points below the average and 29% below it in relative terms. The range between the best and worst outcomes was a 12.7 percentage point difference in real returns.

When the CAPE 10 was between 15.7 and 17.3 (about its long-term average of 16.5), the 10-year-forward real return averaged 5.6%. The best and worst 10-year-forward real returns were 15.1% and 2.3%, respectively. The range between the best and worst outcomes was a 12.8 percentage point difference in real returns.

When the CAPE 10 was between 21.1 and 25.1, the 10-year-forward real return averaged just 0.9%. The best 10-year-forward real return was still 8.3%, above the historical average of 6.8%. However, the worst 10-year-forward real return was now -4.4%. The range between the best and worst outcomes was a difference of 12.7 percentage points in real terms.

When the CAPE 10 was above 25.1, the real return over the following 10 years averaged just 0.5%—virtually the same as the long-term real return on the risk-free benchmark, one-month Treasury bills. The best 10-year-forward real return was 6.3%, just 0.5 percentage points below the historical average. But the worst 10-year-forward real return was now -6.1%. The range between the best and worst outcomes was a difference of 12.4 percentage points in real terms.”

randomguy
Posts: 8531
Joined: Wed Sep 17, 2014 9:00 am

Re: Updated longevity calculator - from the actuaries

Post by randomguy » Sun Oct 13, 2019 7:02 pm

TheNightsToCome wrote:
Sun Oct 13, 2019 6:14 pm

When the CAPE 10 was above 25.1, the real return over the following 10 years averaged just 0.5%—virtually the same as the long-term real return on the risk-free benchmark, one-month Treasury bills. The best 10-year-forward real return was 6.3%, just 0.5 percentage points below the historical average. But the worst 10-year-forward real return was now -6.1%. The range between the best and worst outcomes was a difference of 12.4 percentage points in real terms.”
And what does that data set look like? Lets look at the year where the CAPE10 was above 25 on Jan 1. What years did a 2012 paper have to look at?
1929
1997-2002

That isn't a ton of data to draw conclusions from. If we were writing the paper today we would add in 2004-2007
2002 CAGR .38%
2004 CAGR 4.88
2005 CAGR 5.41
2006 CAGR 5.33
2007 CAGR 5.03%

I am guessing that if you reran the numbers, the average return would jump significantly. The longer we go without a big correction, we will get more and more high PE10s with decent returns as 2010+ get added to the dataset.

EnjoyIt
Posts: 3211
Joined: Sun Dec 29, 2013 8:06 pm

Re: Updated longevity calculator - from the actuaries

Post by EnjoyIt » Sun Oct 13, 2019 10:30 pm

TheNightsToCome wrote:
Sun Oct 13, 2019 6:14 pm
EnjoyIt wrote:
Sun Oct 13, 2019 3:58 pm
TheNightsToCome wrote:
Sun Oct 13, 2019 11:29 am
EnjoyIt wrote:
Sun Oct 13, 2019 8:54 am
Nice link. Thanks.
My risk of living past 95 is 12%
My risk of a 4% withdrawal rate failing in 5%
That means my risk of living past 95 and running out of many is 0.6%

I like those odds.
"My risk of a 4% withdrawal rate failing in 5%"

That might be true if current stock and bond valuations were equal to historical averages. However, they are much higher, and thus prospective returns are much lower (than the historical average) and SOR risk is higher.

If we had 100+ data points conditional on today's valuations, it's highly likely that the 4% withdrawal rate would have failed (much) more often than 5%.
Valuations were much higher around the late 90s early 2000s and those folks appear to be doing just fine today. Valuations help, but they don't predict the future. If you had 100+ data points, you still could not predict future returns. Even if the odds of failure at 95 is 20% then my risk of running out of money and living longer than 95 is 2.4%. Guess what, I still like those odds.
If we add in that retirees spend less and less as they get older, my odds improve. If you add in that I am not a robot and will make some adjusts in my spending if the markets acts atrociously my odds of failure drop well below a fraction of a fraction of a percent. Guess what, I still like those odds.
"Even if the odds of failure at 95 is 20% then my risk of running out of money and living longer than 95 is 2.4%. Guess what, I still like those odds."

This is probably closer to our current circumstance.

"Valuations help, but they don't predict the future."

They do predict the future, just not with precision:

Larry Swedroe alluded to a paper by Cliff Asness (Swedroe here: https://www.etf.com/sections/index-inve ... nopaging=1 and Asness here: file:///C:/Users/Curt/Downloads/An%20Old%20Friend%20The%20Stock%20Markets%20Shiller%20PE%20(1).pdf ):

“In a November 2012 paper, “An Old Friend: The Stock Market’s Shiller P/E,” Asness, of AQR Capital Management, found that the Shiller CAPE 10 provides valuable information. Specifically, he found 10-year-forward average real returns drop nearly monotonically as starting Shiller P/Es increase.

He also found that, as the starting Shiller CAPE 10 ratio increased, worst cases became worse, and best cases became weaker. Additionally, he found that, while the metric provided valuable insights, there were still very wide dispersions of returns. For instance:

When the CAPE 10 was below 9.6, 10-year-forward real returns averaged 10.3%. In relative terms, that is more than 50% above the historical average of 6.8% (9.8% nominal return less 3.0% inflation). The best 10-year-forward real return was 17.5%. The worst 10-year-forward real return was still a pretty good 4.8%, just 2.0 percentage points below the average and 29% below it in relative terms. The range between the best and worst outcomes was a 12.7 percentage point difference in real returns.

When the CAPE 10 was between 15.7 and 17.3 (about its long-term average of 16.5), the 10-year-forward real return averaged 5.6%. The best and worst 10-year-forward real returns were 15.1% and 2.3%, respectively. The range between the best and worst outcomes was a 12.8 percentage point difference in real returns.

When the CAPE 10 was between 21.1 and 25.1, the 10-year-forward real return averaged just 0.9%. The best 10-year-forward real return was still 8.3%, above the historical average of 6.8%. However, the worst 10-year-forward real return was now -4.4%. The range between the best and worst outcomes was a difference of 12.7 percentage points in real terms.

When the CAPE 10 was above 25.1, the real return over the following 10 years averaged just 0.5%—virtually the same as the long-term real return on the risk-free benchmark, one-month Treasury bills. The best 10-year-forward real return was 6.3%, just 0.5 percentage points below the historical average. But the worst 10-year-forward real return was now -6.1%. The range between the best and worst outcomes was a difference of 12.4 percentage points in real terms.”
Lack of precision is the problem. To add to all that CAPE 10 analysis above, there really was only 1 time in history where 4% failed and that was during 1966 where cape 10 was lower than in 1999/2000. And those 1999/2000 retirees are doing just fine. I don’t think CAPE 10 has any predictive value in one’s ability to sustain 4% withdrawals for 30 years because 4% already includes some really low returns. In fact, if valuations were lower we probably would be able to pull 4.5 or 5% every year and be fine.

So sure, valuations are high and returns maybe lower, but 4% withdrawals accounts for that.

Post Reply