Longevity

[jon-nyc]

Some quick thoughts:

(1) Surely some researcher has come up with a quantile regression model that gives better results than period/cohort table conditioned only on sex.

There are great calculators out there that take into account loads of demographic and medical information. But they're not free.

 

[Big_Hitter]

There are great calculators out there that take into account loads of demographic and medical information. But they're not free.

you mean like life underwriting software?

 

[jon-nyc]

Yep.

 

[photoguy]

you mean like life underwriting software?

Is it possible to get access to something like this as a non-employee?

I see a lot of free calculators on the web, but I think they are probably garbage, not actually based on real data / statistical model, and likely ask way more questions than needed to collect your data.

 

[samclem]

There are great calculators out there that take into account loads of demographic and medical information. But they're not free.

I gotta think that these tools are very useful in accurately predicting mortality over a large number of cases. But isn't the variance in any one particular case (the one I care about--ME!) fairly high? How precise is this likely to me? How precise would we want it to be? ("Ask not for whom the bell tolls . . . .")

 

[haha]

This is likely just part of the general obsessiveness found here.

As to its practical value, maybe it is worth 4 hours early in ones retirement planning to understand that life expectancy at birth is not what you are interested in, unless you are literally going to be born again. Also what is the relevant group you belong in-ie life insurance buyers or annuity buyers.

Maybe a few very general principles, like you should be much more interested in annuity tables than life tables for insurance buyers. Glance in the mirror to get a fix on your sex. If you are a member of a couple use the relevant factor for the last to die.

Beyond this, one's own health very broadly assessed, and maybe a few lifestyle factors, like do you smoke a pack a day and drink until you pass out every night? Finer than this needn't concern us, because we don't want to run out of money, but we really do not care if we don't zero out the bank account on the date of our demise.

Overall we will do more for our security by working an extra year than anything we can do by becoming amateur actuaries.


Ha

 

[daylatedollarshort]

I don't think it's a black swan event for one of two people in their 60s to make it to 100.

+1. It has happened in my family. I'd be dumb not to plan to age 100+. If getting a programming job at age 55 is hard for many middle aged people, I think 90 would be super tough. I am okay with leaving money to charity. My goal is not to spend the portfolio to zero. I don't want to have to worry about running out of money in my old age.

 

[photoguy]

A somewhat related article from 538:

Why The Oldest Person In The World Keeps Dying | FiveThirtyEight

I found that more people than ever are clustering at the outer edge of human aging and that the tenure of the world’s oldest living person isn’t as long as it used to be. Better record-keeping and longer lifespans have helped lead to quite a crowd.

Supercentenarians — people who have lived past their 110th birthday — generally come from a heartier stock than most people. They tend to have few age-related health issues and are much physically and mentally sharper than their peers during their 80s and 90s. Weaver, for example, didn’t move into the rehabilitation home until she was 109. As we enter an age with less war and infection and fewer accidents, more and more people with these superior aging genes have been able to make it to a point in time when they can show them off. It’s getting crowded at the top.

Wow staying out of a nursing home until 109!

 

[imoldernu]

I believe that this article from The Atlantic, deserves a read. Not just from a standpoint of statistics, but some expectations of how an aging society may change the entire face of society, such as the increase in the aging population creating a different type of political landscape. A power bloc of aging voters who may influence Federal budget items such as Taxes, Social Security, Pensions, Healthcare and the many benefits that may accrue to the elderly.

What Happens When We All Live to 100? - The Atlantic

Longer life has obvious appeal, but it entails societal risks. Politics may come to be dominated by the old, who might vote themselves ever more generous benefits for which the young must pay. Social Security and private pensions could be burdened well beyond what current actuarial tables suggest. If longer life expectancy simply leads to more years in which pensioners are disabled and demand expensive services, health-care costs may balloon as never before, while other social needs go unmet.

 

[2017ish]

I believe that this article from The Atlantic, deserves a read. Not just from a standpoint of statistics, but some expectations of how an aging society may change the entire face of society, such as the increase in the aging population creating a different type of political landscape. A power bloc of aging voters who may influence Federal budget items such as Taxes, Social Security, Pensions, Healthcare and the many benefits that may accrue to the elderly.

What Happens When We All Live to 100? - The Atlantic

Yes; thanks for the link. Worth the time to read and consider. Hadn't seen it before. This was not the crux of the article, but an observation that most will sigh and agree with, regardless of political leanings:

A fair guess is that the government will do nothing about Social Security reform until a crisis strikes—and then make panicked, ill-considered moves that foresight might have avoided.

 

[Rick_Head]

I've also wondered about family history of longevity and how it figures into calculations.

My stats are:
Paternal:
- GGF died at 105
- GF died at 87
- F died at 55
- GM died at 97, her brother at 98 (he "RE'd" at 97). F's siblings at 14 and 80ish.

Maternal
- GF at ~62
- GM (and her sisters) made upper 90's
- M died at 78, her siblings at 90ish.

So, when my dad was my age he'd been dead for 10 years. When my mother was my age she was a 10 year cancer survivor and had another 10 yr or so..

I think, on paternal lines, I just don't know I'm dead (although DW might). On maternal lines, who knows.

How should I tweak the tables to guess a lifespan?

 

[Tandemlovers]

3-5% of the population is hardly going to bankrupt SS.

 

[daylatedollarshort]

Our biggest longevity insurance is frugal living relative to our retirement income and assets. That far outweighs most other factors for us.

 

[Totoro]

Maybe interesting: I obviously don't know what the US government will do, but this is typically what most governments in Europe have done for the SS equivalent of benefits because of shifting demographics:

• Keep the current qualifying age and benefit levels for everyone within 10 years of achieving it (sometimes up to 15 years).
• Increase gradually the age when entitlements start, e.g. every 5 years increase one year.
• Increase taxation for groups with a very high entitlement (typically retired governement officials)
• Lower taxes for individuals who choose to defer receiving SS (and typically keep on working)
• Limit early retirement options

What they haven't done:

• Block inflation adjustments
• Lower SS-type benefits for individuals having already hit the qualifying age
• Lower SS-type benefits for future retirees in amount received per year.

In addition, sometimes defined benefit pensions have gotten haircuts (usually through postponing inflation adjustments). And many pensions (on top of any SS-type system) are being converted to defined contribution.