Author Topic: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement  (Read 14387 times)

CCCA

  • Bristles
  • ***
  • Posts: 487
  • Location: Bay Area, California
  • born before the 80's
    • FI programming
Wow, you weren't kidding. At least one person over there arguing for assuming 0% real returns going forward others for sub 2% withdrawal rates (which if you don't retire until your 60s means you're assuming less than 0% real returns or that you'll live past 110).

That just Boggles the mind! They are cunning though all those extra years of work mean they'll die with fewer years of FIRE to fund.

0% real returns; How crazy is that?  Even TIPs pay better than that nothing.


I don't know if we here at the MMM forums are too optimistic, but it's almost as if the bogleheads (at least the vocal ones about the risks moving forward) are planning for some sort of black swan event.


Also I've upgraded the tool to be much more mobile friendly.  Anyone want to take a look on a mobile device and let me know if it becomes unusable or any issues they have?  I've tested on a few small devices but can't test everything obviously.  This is my first attempt at responsive web design so it's not perfect, but should make it better on a phone.  I think tablets were fine before.  Desktop/laptop is still the ideal way to view the tool, since it allows for hovering with the cursor (and still easiest to enter data into the calculator).


thanks!

dragoncar

  • Walrus Stache
  • *******
  • Posts: 8106
  • Registered member
Wow, you weren't kidding. At least one person over there arguing for assuming 0% real returns going forward others for sub 2% withdrawal rates (which if you don't retire until your 60s means you're assuming less than 0% real returns or that you'll live past 110).

That just Boggles the mind! They are cunning though all those extra years of work mean they'll die with fewer years of FIRE to fund.

0% real returns; How crazy is that?  Even TIPs pay better than that nothing.


I don't know if we here at the MMM forums are too optimistic, but it's almost as if the bogleheads (at least the vocal ones about the risks moving forward) are planning for some sort of black swan event.


Also I've upgraded the tool to be much more mobile friendly.  Anyone want to take a look on a mobile device and let me know if it becomes unusable or any issues they have?  I've tested on a few small devices but can't test everything obviously.  This is my first attempt at responsive web design so it's not perfect, but should make it better on a phone.  I think tablets were fine before.  Desktop/laptop is still the ideal way to view the tool, since it allows for hovering with the cursor (and still easiest to enter data into the calculator).


thanks!

Personally, it's not that I don't think black swan can't happen.  It's just that I don't think there's a good way to plan for it.  Saving for a 0% real return isn't really going to help you if capitalism collapses or an asteroid hits or something like that.  Even with preppers, I think they are probably prepping for the wrong event.  Unless you have a crystal ball, I lean towards cautious optimism

markbike528CBX

  • Pencil Stache
  • ****
  • Posts: 775
  • Location: the Everbrown part of the Evergreen State (WA)
Wow, you weren't kidding. At least one person over there arguing for assuming 0% real returns going forward others for sub 2% withdrawal rates (which if you don't retire until your 60s means you're assuming less than 0% real returns or that you'll live past 110).

That just Boggles the mind! They are cunning though all those extra years of work mean they'll die with fewer years of FIRE to fund.

0% real returns; How crazy is that?  Even TIPs pay better than that nothing.

I don't know if we here at the MMM forums are too optimistic, but it's almost as if the bogleheads (at least the vocal ones about the risks moving forward) are planning for some sort of black swan event.

Also I've upgraded the tool to be much more mobile friendly.  Anyone want to take a look on a mobile device and let me know if it becomes unusable or any issues they have?  I've tested on a few small devices but can't test everything obviously.  This is my first attempt at responsive web design so it's not perfect, but should make it better on a phone.  I think tablets were fine before.  Desktop/laptop is still the ideal way to view the tool, since it allows for hovering with the cursor (and still easiest to enter data into the calculator).


thanks!

Personally, it's not that I don't think black swan can't happen.  It's just that I don't think there's a good way to plan for it.  Saving for a 0% real return isn't really going to help you if capitalism collapses or an asteroid hits or something like that.  Even with preppers, I think they are probably prepping for the wrong event.  Unless you have a crystal ball, I lean towards cautious optimism
And how do " bogleheads and preppers" not acknoledge that is 1929 Not a black swan?   The 1970s*** were spread over a decade, but that's probably a black swan too. The 4% "rule"  does work with these to say that you would still have 95% (19 in20) success.

*** If it gets worse than Disco, no amount of ammo and firepower is enough.

DreamFIRE

  • Handlebar Stache
  • *****
  • Posts: 1291

It's reasonably possible for 0% returns over a decade or longer when the next bear market hits.   There were negative returns in the S&P500 over a 13+ year period not too many years ago.

https://forum.mrmoneymustache.com/investor-alley/why-not-do-100-allocation-draw-4-at-retirement-and-yolo-it/msg2023973/#msg2023973

And that's not even the longest period historically.

maizeman

  • Magnum Stache
  • ******
  • Posts: 2850

It's reasonably possible for 0% returns over a decade or longer when the next bear market hits.   There were negative returns in the S&P500 over a 13+ year period not too many years ago.

https://forum.mrmoneymustache.com/investor-alley/why-not-do-100-allocation-draw-4-at-retirement-and-yolo-it/msg2023973/#msg2023973

And that's not even the longest period historically.

There is a big difference between 0% real returns for a decade if you pick exactly the right starting point and negative real returns over five decades. And that's putting aside that if you're in your sixties and retiring with a sub 2% withdrawal rate, there is really no need to be invested in the stock market at all vs something like TIPS.*

*I'd argue there also wasn't a need to work to the point where you have 50 years worth of expenses saved in the first place, but that's a separate discussion.

DreamFIRE

  • Handlebar Stache
  • *****
  • Posts: 1291

It's reasonably possible for 0% returns over a decade or longer when the next bear market hits.   There were negative returns in the S&P500 over a 13+ year period not too many years ago.

https://forum.mrmoneymustache.com/investor-alley/why-not-do-100-allocation-draw-4-at-retirement-and-yolo-it/msg2023973/#msg2023973

And that's not even the longest period historically.

There is a big difference between 0% real returns for a decade if you pick exactly the right starting point and negative real returns over five decades.

Who said anything about 50 years?   20 years was mentioned in recent posts and so was a "decade".  Not everyone has 50 years of retirement ahead of them, especially before their pension/SS kicks in and pays most of the bills.  I believe negative real returns have run close to 20 years during one historical time period, but a negative return over 13 years as happened more recently is significant as well, even if your retirement horizon is longer.  Read up on sequence of returns risk.  And also consider SS and possible pension income that might kick in for many after a decade or so of drawing more heavily from the stash.

Quote
And that's putting aside that if you're in your sixties and retiring with a sub 2% withdrawal rate, there is really no need to be invested in the stock market at all vs something like TIPS.*

I never said anything about sub 2%.  But remember, when you sell your stock funds in your taxable brokerage account to invest in TIPS, you'll be paying federal and state capital gains taxes.  Also, general good investing advice isn't investing in 100% TIPS either before or during retirement.

Quote
*I'd argue there also wasn't a need to work to the point where you have 50 years worth of expenses saved in the first place, but that's a separate discussion.

I never commented regarding years of expenses or SWR in my post that you replied to.  But since you mentioned it, saving just enough to pay barebones expenses doesn't sound like a good way to FIRE to me.  I saved significantly more than my barebones expenses primarily so that I can enjoy far more travel and entertainment than I have during my working years.  After paying for necessary expenses, the additional savings allows more dollars for unbudgeted discretionary spending than it does paying required living expenses.
« Last Edit: October 04, 2018, 06:21:13 PM by DreamFIRE »

maizeman

  • Magnum Stache
  • ******
  • Posts: 2850
Ah, my mistake. I naturally assumed your post about 0% returns was related to our existing discussion in this thread about the sub zero percent returns being assumed by folks on bogleheads. Now that I know it was an unrelated non sequitur, carry on.

Telecaster

  • Handlebar Stache
  • *****
  • Posts: 1317
  • Location: Seattle, WA
Personally, it's not that I don't think black swan can't happen.  It's just that I don't think there's a good way to plan for it.  Saving for a 0% real return isn't really going to help you if capitalism collapses or an asteroid hits or something like that.  Even with preppers, I think they are probably prepping for the wrong event.  Unless you have a crystal ball, I lean towards cautious optimism

Exactly.  The Black Swan by definition is the thing you don't foresee happening.  I'm planning that over the next 40 years or so, capitalism will continue working basically the way it has, there will be a number of wars, but nothing too dramatic, a number of economic recessions and market crashes...just like we've seen for a long time in the past. 

Nuclear war, asteroid strikes, pandemic plague, zombie apocalypse, etc. could happen, but it is just too much trouble to try to plan for all those things.  I'll just call it good when I have 25X expenses and let the chips fall where they may. 

EnjoyIt

  • Handlebar Stache
  • *****
  • Posts: 1099
There was an interesting comment on bogleheads the other day that I will poorly paraphrase.

When the market returns are strong people find themselves in a good position and retire early.  This usually will occur in a nice bull market as we have today.  This increases the chance of a recession coming in early retirement since after a long bull market historically a bear market will follow.

Does that mean that many of us who just retired or are looking to retire very soon are more likely to experience sequence of returns risk and therefor stress on their portfolio? I don't want to use portfolio failure since most sane people will cut spending significantly and/or find income to sustain their portfolio during those bad times.

maizeman

  • Magnum Stache
  • ******
  • Posts: 2850
The effect appears to be real, although as long as you're saving at least 50% of your income each year, it shouldn't create a dramatically large bias in failure rates.

@gerardc did some nice calculations that showed that the failure rate of a 90%/10% portfolio with a 50% savings rate until you hit 25x expenses and 4% withdrawal rate thereafter might decline by a couple of percentage points if you account for the fact people are more likely to hit their magic number during market run ups than at other times.

https://forum.mrmoneymustache.com/welcome-to-the-forum/cfiresim-severely-overestimates-success-rates-for-mustachians/msg1625045/#msg1625045

EnjoyIt

  • Handlebar Stache
  • *****
  • Posts: 1099
The effect appears to be real, although as long as you're saving at least 50% of your income each year, it shouldn't create a dramatically large bias in failure rates.

@gerardc did some nice calculations that showed that the failure rate of a 90%/10% portfolio with a 50% savings rate until you hit 25x expenses and 4% withdrawal rate thereafter might decline by a couple of percentage points if you account for the fact people are more likely to hit their magic number during market run ups than at other times.

https://forum.mrmoneymustache.com/welcome-to-the-forum/cfiresim-severely-overestimates-success-rates-for-mustachians/msg1625045/#msg1625045

Thanks for the link.  The few percentage points are legit and not insignificant.  This is just one more reason for us to go part time instead of FIRE completely.

TomTX

  • Magnum Stache
  • ******
  • Posts: 3006
  • Location: Texas
Made a couple of small updates to the calculator:
  • added a >5x initial starting balance wedge as well.  This is the largest wedge in most scenarios after about 45 years at 4% withdrawal rate (not counting the death wedge of course). 
  • added a Generate URL button to save the parameters into the URL so you can share your scenarios with other folks.
In other news, I shared these calculators with friends and family and a friend starting a new business said he needed some interactive visualizations done for his website.  So I've now got a small side gig of web programmer (after just learning this web stuff earlier this year).
This tool was already great when I last checked it, and it just keeps getting better!  I'm starting to think I may have built a teensy bit too much of a safety margin into my plans given the embarrassingly large size of my 5x wedge even on the inflation adjusted basis.  I'm too scared to look how bad it will be if I stop budgeting for a 10% spending increase over my pre retirement budget (which most years I've been beating by 10%).  I've worked too long haven't I?  Doh!

This tool is the reason I moved from the 2027 FIRE group to 2025, with a stretch goal of 2023.

2027 is when I can start drawing a pension (with medical) if I keep working there until retirement date.

CCCA

  • Bristles
  • ***
  • Posts: 487
  • Location: Bay Area, California
  • born before the 80's
    • FI programming
Made a couple of small updates to the calculator:
  • added a >5x initial starting balance wedge as well.  This is the largest wedge in most scenarios after about 45 years at 4% withdrawal rate (not counting the death wedge of course). 
  • added a Generate URL button to save the parameters into the URL so you can share your scenarios with other folks.
In other news, I shared these calculators with friends and family and a friend starting a new business said he needed some interactive visualizations done for his website.  So I've now got a small side gig of web programmer (after just learning this web stuff earlier this year).
This tool was already great when I last checked it, and it just keeps getting better!  I'm starting to think I may have built a teensy bit too much of a safety margin into my plans given the embarrassingly large size of my 5x wedge even on the inflation adjusted basis.  I'm too scared to look how bad it will be if I stop budgeting for a 10% spending increase over my pre retirement budget (which most years I've been beating by 10%).  I've worked too long haven't I?  Doh!

This tool is the reason I moved from the 2027 FIRE group to 2025, with a stretch goal of 2023.

2027 is when I can start drawing a pension (with medical) if I keep working there until retirement date.


Awesome!  Glad it helped in framing your thinking. 


Regarding black swans, I guess some critical questions are:
1. Does a large stache still have much of it's value afterwards?
2. Would having a job (i.e. not being retired) help during this black swan?


I think both of these questions cannot adequately be answered, so it's not clear that prepping with more money or not being retired will help.  In the 99% probability that the black swan doesn't happen, then you'd probably be better off by pulling the plug at 25-30x your annual spending.

Bird In Hand

  • Bristles
  • ***
  • Posts: 423
In the 99% probability that the black swan doesn't happen, then you'd probably be better off by pulling the plug at 25-30x your annual spending.

I know that your guess is as good as mine (and you probably didn't literally mean 99% anyway), but the US alone has had 2 financial black swans in the last 18 years -- really two in an ~8-9 year period followed by the very long ~9-10 bull market we're currently in.  Decades from now this might look like an odd statistical aberration, or it might be the new normal where financial black swans happen every decade or so.

What to do about it?  For those lucky enough to have a "safe" job that they enjoy, part-time work is not a bad strategy IMO.  Of course a lot of people really don't like their jobs, or don't have the flexibility to work part time.  Those folks can choose a lower SWR or just rely on their ingenuity, work ethic, and way-larger-than-average nest eggs to make it through financial black swans.

As @Telecaster says, other types of black swans are not really things we can reasonably mitigate with personal financial planning anyway.

maizeman

  • Magnum Stache
  • ******
  • Posts: 2850
In the 99% probability that the black swan doesn't happen, then you'd probably be better off by pulling the plug at 25-30x your annual spending.

I know that your guess is as good as mine (and you probably didn't literally mean 99% anyway), but the US alone has had 2 financial black swans in the last 18 years -- really two in an ~8-9 year period followed by the very long ~9-10 bull market we're currently in.  Decades from now this might look like an odd statistical aberration, or it might be the new normal where financial black swans happen every decade or so.

But are these black swans? It's too early to know for sure but so far looks like someone who retired in 2000 using the 4% rule may well turn out to be fine. People do these analyses every couple of years, but here's the most recent analysis I could find: https://www.bogleheads.org/forum/viewtopic.php?t=237334

Here is a somewhat older analysis (2015) which looked at both the 2000 and 2008 retiree and how they are doing today: https://www.kitces.com/blog/how-has-the-4-rule-held-up-since-the-tech-bubble-and-the-2008-financial-crisis/

CCCA

  • Bristles
  • ***
  • Posts: 487
  • Location: Bay Area, California
  • born before the 80's
    • FI programming
In the 99% probability that the black swan doesn't happen, then you'd probably be better off by pulling the plug at 25-30x your annual spending.

I know that your guess is as good as mine (and you probably didn't literally mean 99% anyway), but the US alone has had 2 financial black swans in the last 18 years -- really two in an ~8-9 year period followed by the very long ~9-10 bull market we're currently in.  Decades from now this might look like an odd statistical aberration, or it might be the new normal where financial black swans happen every decade or so.

But are these black swans? It's too early to know for sure but so far looks like someone who retired in 2000 using the 4% rule may well turn out to be fine. People do these analyses every couple of years, but here's the most recent analysis I could find: https://www.bogleheads.org/forum/viewtopic.php?t=237334

Here is a somewhat older analysis (2015) which looked at both the 2000 and 2008 retiree and how they are doing today: https://www.kitces.com/blog/how-has-the-4-rule-held-up-since-the-tech-bubble-and-the-2008-financial-crisis/


I agree with Maizeman.  I don't really consider these to be black swans.  Just bad instances of normal economic variability.  Like a monte carlo where your random draw is from the bad tail.  The black swans I'm talking about where there's some event that causes a significant economic collapse and it's not going to recover anytime soon. 


But yes, if the new normal is low to negative average economic growth with significant drops every once in awhile, then yes, it's going to be hard to retire in that context as well.   

Bird In Hand

  • Bristles
  • ***
  • Posts: 423
I agree with Maizeman.  I don't really consider these to be black swans.  Just bad instances of normal economic variability.  Like a monte carlo where your random draw is from the bad tail.  The black swans I'm talking about where there's some event that causes a significant economic collapse and it's not going to recover anytime soon. 

I see your point -- I guess it depends on what you consider a black swan.  Personally I view the DotCom bust and the 2008 events as quite a bit outside normal economic variability.  But it is true that the market recovery after each of those occurred much faster than it did after the Great Depression, or even the late 60's to early 80's doldrums.

I'll have to go and review Maizeman's links to see what they say about SORR for a Y2K retiree.  As he mentioned, the picture is incomplete (with respect to SWR) because we haven't reached 30 years yet.

EnjoyIt

  • Handlebar Stache
  • *****
  • Posts: 1099
Considering what the definition of a black swan is, dot com and the Great Recession do not qualify.

Maybe the Great Depression does, the long collapse of the Japanese market does. The collapse of the Soviet Union is a great example. Maybe what is occuring in Venezuela will qualify. Aliens landing on Earth and enslaving is will be a Black Swan.

The last two recessions are far cries from Black Swans.

Quote from: ‘Taleb’
The theory was developed by Nassim Nicholas Taleb to explain:

1) The disproportionate role of high-profile, hard-to-predict, and rare events that are beyond the realm of normal expectations in history, science, finance, and technology.

2) The non-computability of the probability of the consequential rare events using scientific methods (owing to the very nature of small probabilities).

3) The psychological biases that blind people, both individually and collectively, to uncertainty and to a rare event's massive role in historical affairs.
« Last Edit: October 06, 2018, 12:40:08 PM by EnjoyIt »

maizeman

  • Magnum Stache
  • ******
  • Posts: 2850
EnjoyIt I agree, something like Venezuela today, or France in 1940, or the Incan empire when Pizarro showed up is not something that can be successfully mitigated by saving enough money and are true black swans.

I'd tend to say the great depression of japan's lost decade(s?) shouldn't count as black swans, but just the extreme edge of the normal range of volatility in modern economic systems at a national level.

Telecaster

  • Handlebar Stache
  • *****
  • Posts: 1317
  • Location: Seattle, WA
Considering what the definition of a black swan is, dot com and the Great Recession do not qualify.

Maybe the Great Depression does, the long collapse of the Japanese market does. The collapse of the Soviet Union is a great example. Maybe what is occuring in Venezuela will qualify. Aliens landing on Earth and enslaving is will be a Black Swan.

The last two recessions are far cries from Black Swans.


I'd say the last recession was a Black Swan.   There was a general belief that we were experiencing a housing bubble, but almost no one realized that complex financial instruments like credit default swaps had become highly leveraged and were poised to topple--and would have toppled--the entire financial system except for swift, massive, and unprecidented government intervention in world financial markets.   Even so, major Wall Street banks like Bear Stearns and Lehman Brothers went bankrupt (the largest bankruptcy filing in history) and even big non-Wall Street banks like Washington Mutual and IndyMac closed their door.  The collapse of the banking industry claimed, or almost claimed other victims like General Motors and Chrysler, who only survived thanks to government intervention.   

One hallmark of a Black Swan is that everyone saw it coming in hindsight.  In hindsight, yes of course this was going to happen.  But very few people saw it ahead of time. 

 

CCCA

  • Bristles
  • ***
  • Posts: 487
  • Location: Bay Area, California
  • born before the 80's
    • FI programming
Considering what the definition of a black swan is, dot com and the Great Recession do not qualify.

Maybe the Great Depression does, the long collapse of the Japanese market does. The collapse of the Soviet Union is a great example. Maybe what is occuring in Venezuela will qualify. Aliens landing on Earth and enslaving is will be a Black Swan.

The last two recessions are far cries from Black Swans.


I'd say the last recession was a Black Swan.   There was a general belief that we were experiencing a housing bubble, but almost no one realized that complex financial instruments like credit default swaps had become highly leveraged and were poised to topple--and would have toppled--the entire financial system except for swift, massive, and unprecidented government intervention in world financial markets.   Even so, major Wall Street banks like Bear Stearns and Lehman Brothers went bankrupt (the largest bankruptcy filing in history) and even big non-Wall Street banks like Washington Mutual and IndyMac closed their door.  The collapse of the banking industry claimed, or almost claimed other victims like General Motors and Chrysler, who only survived thanks to government intervention.   

One hallmark of a Black Swan is that everyone saw it coming in hindsight.  In hindsight, yes of course this was going to happen.  But very few people saw it ahead of time.


Regardless of whether 2008 is a black swan or not, it seems like "obvious in hindsight" should not be a defining characteristic of black swans, at least as the term is currently used and understood.  An alien invasion, or crazy sunspot, or other external event that is not predictable seems like the obvious example where "obvious" isn't applicable. 


However, I think the overall point is whether people generally think the future will be like the past or not.  Given the pace of technological change in the world, it's definitely possible that financial markets, capitalism and our political/social structures could change rapidly too as a result.  Or they may not. 

Mgmny

  • Pencil Stache
  • ****
  • Posts: 505
  • Age: 28
  • Location: Northwest 'Burbs of MSP
JL Collins sums it up very well in his stock series: The market has never failed us yet. It has ALWAYS gone up with any type of X(20 year? 40 year? I can't remember...) span in the history of the market it has gone up. If you are concerned that there will be a black swan event and the market crashes, we have bigger issues to worry about - and there won't be a way to prepare. If Aliens enslave us, your 'stache won't matter. If Iran bombs us to smithereens, your 'stache won't matter. If US breaks into a crazy civil/nuclear war, the size of your 'stache won't matter.

As long as you believe that the US economy won't collapse (e.g. Black Swan catastrophe), the market will continue to grow.

maizeman

  • Magnum Stache
  • ******
  • Posts: 2850
I prefer to switch the order around but the conclusion is the same: if the market fails to provide a positive return for multiple decades, then our society (and by extension each of us individually) is going to have much bigger problems than worrying about haven’t spent down all of our principle.

Telecaster

  • Handlebar Stache
  • *****
  • Posts: 1317
  • Location: Seattle, WA

Regardless of whether 2008 is a black swan or not, it seems like "obvious in hindsight" should not be a defining characteristic of black swans, at least as the term is currently used and understood.  An alien invasion, or crazy sunspot, or other external event that is not predictable seems like the obvious example where "obvious" isn't applicable. 


Well, it kind of is.  When Nassim Taleb introduced the concept of black swan events, a central point was that we don't often see them ahead of time because our cognitive biases.   Perhaps the event is actually far more likely than we believed.   We use those same cognitive biases to convince ourselves that the black swan was obvious the whole time, even when that may not have been the case.   In case when an event was completely random, for example.   We're fooling ourselves, in other words.  Hence the name of the book "Fooled by Randomness." 

maizeman

  • Magnum Stache
  • ******
  • Posts: 2850

Regardless of whether 2008 is a black swan or not, it seems like "obvious in hindsight" should not be a defining characteristic of black swans, at least as the term is currently used and understood.  An alien invasion, or crazy sunspot, or other external event that is not predictable seems like the obvious example where "obvious" isn't applicable. 


Well, it kind of is.  When Nassim Taleb introduced the concept of black swan events, a central point was that we don't often see them ahead of time because our cognitive biases.   Perhaps the event is actually far more likely than we believed.   We use those same cognitive biases to convince ourselves that the black swan was obvious the whole time, even when that may not have been the case.   In case when an event was completely random, for example.   We're fooling ourselves, in other words.  Hence the name of the book "Fooled by Randomness."

Is that correct? Here's the definition I was able to find by Taleb for black swans:

"First, it is an outlier, as it lies outside the realm of regular expectations, because nothing in the past can convincingly point to its possibility. Second, it carries an extreme 'impact'. Third, in spite of its outlier status, human nature makes us concoct explanations for its occurrence after the fact, making it explainable and predictable."

That makes it sound not like the we fail to predict individual black swans because of cognitive biases, but that after the fact our cognitive biases lead us to we convince ourselves they could have been predicted. However, I haven't read Taleb's books in any depth. I attempted Antifragile once but simply disliked his writing style too much to make it through to the end, even though I think there were some useful ideas contained within. Anyway, please do tell me if I'm misunderstanding the core concept by taking the one quote from him out of context.

CCCA

  • Bristles
  • ***
  • Posts: 487
  • Location: Bay Area, California
  • born before the 80's
    • FI programming

Regardless of whether 2008 is a black swan or not, it seems like "obvious in hindsight" should not be a defining characteristic of black swans, at least as the term is currently used and understood.  An alien invasion, or crazy sunspot, or other external event that is not predictable seems like the obvious example where "obvious" isn't applicable. 


Well, it kind of is.  When Nassim Taleb introduced the concept of black swan events, a central point was that we don't often see them ahead of time because our cognitive biases.   Perhaps the event is actually far more likely than we believed.   We use those same cognitive biases to convince ourselves that the black swan was obvious the whole time, even when that may not have been the case.   In case when an event was completely random, for example.   We're fooling ourselves, in other words.  Hence the name of the book "Fooled by Randomness."

Is that correct? Here's the definition I was able to find by Taleb for black swans:

"First, it is an outlier, as it lies outside the realm of regular expectations, because nothing in the past can convincingly point to its possibility. Second, it carries an extreme 'impact'. Third, in spite of its outlier status, human nature makes us concoct explanations for its occurrence after the fact, making it explainable and predictable."

That makes it sound not like the we fail to predict individual black swans because of cognitive biases, but that after the fact our cognitive biases lead us to we convince ourselves they could have been predicted. However, I haven't read Taleb's books in any depth. I attempted Antifragile once but simply disliked his writing style too much to make it through to the end, even though I think there were some useful ideas contained within. Anyway, please do tell me if I'm misunderstanding the core concept by taking the one quote from him out of context.
I haven’t read Taleb either but since j was the one who originally brought up black swans I guess I meant it more as generic unpredictable and improbable event rather than a specific subset of these events.

letsdoit

  • Stubble
  • **
  • Posts: 230
i tried to listen to the author speak once on youtube, he was intolerable

matchewed

  • Magnum Stache
  • ******
  • Posts: 4256
  • Location: CT
I've been having fun using this tool to visualize certain scenarios. It makes me realize how resilient the plan can be in later years. Factoring in SS as soon as possible and assuming some terrible health event that is going to make me spend $50k more/year from 70 onto death at 90 or some number I probably won't hit. It barely budges the thin sliver of broke.

It also reinforces that even a minor amount of income early in FIRE can mitigate and increase success chances, especially if you have a relatively small expenses number compared to the population/forum at large.

CCCA just wanted to say thanks again. This is a cool tool.

Telecaster

  • Handlebar Stache
  • *****
  • Posts: 1317
  • Location: Seattle, WA

Is that correct? Here's the definition I was able to find by Taleb for black swans:

"First, it is an outlier, as it lies outside the realm of regular expectations, because nothing in the past can convincingly point to its possibility. Second, it carries an extreme 'impact'. Third, in spite of its outlier status, human nature makes us concoct explanations for its occurrence after the fact, making it explainable and predictable."

That makes it sound not like the we fail to predict individual black swans because of cognitive biases, but that after the fact our cognitive biases lead us to we convince ourselves they could have been predicted. However, I haven't read Taleb's books in any depth. I attempted Antifragile once but simply disliked his writing style too much to make it through to the end, even though I think there were some useful ideas contained within. Anyway, please do tell me if I'm misunderstanding the core concept by taking the one quote from him out of context.

It is nuanced, he's written a couple books on the topic, and it is hard to explain or unpack in a few sentences.  You're pretty much right.  The reason why I said "don't often see them ahead of time" because cognitive bias is part of the definition.*   For example, many people assume that stock returns are normal, because that's how they learned how to calculate standard deviations.  But they aren't normal.  That's an example of a cognitive bias blinding you to a potential problem. 

There is a parallel discussion to this in another thread.  A poster is trying to reduce risk by reducing volatility--which lots of people do.   Even Michael Kitces who I think is great does this.  However, that's wrong.   Taleb points out that real risk comes from things like the banking crisis, which is totally outside the "volatility is risk model."  So if you are say, selecting a portfolio of low-volatility banking stocks, thinking lower volatility is safer,  your cognitive biases will cause you to over look the possibility of real risks.   Some people saw the banking crisis coming so it wasn't impossible to predict, but most people didn't. 

Edit: Of course, this is all hair splitting anyway.  I'm sure we're all in agreement there are risks out there that we don't know about.   If we don't know about them, it is hard to plan for them.   I'm personally just putting the possibility of a Black Swan event in the "too hard" pile and calling it good.   

And I agree with you.  I don't really like his writing style either. 

"not often" probably wasn't the right phrase.  Maybe "very rarely" would have been better.
« Last Edit: October 10, 2018, 12:49:16 AM by Telecaster »

CCCA

  • Bristles
  • ***
  • Posts: 487
  • Location: Bay Area, California
  • born before the 80's
    • FI programming
Back to the issue of longevity data for differnet populations, someone on reddit suggested looking at the Society of Actuaries.  It looks like they have data on different populations including Blue Collar vs White Collar and Bottom Quartile and Top Quartile and split by sex.  It's in an excel doc on this website: 

https://www.soa.org/experience-studies/2014/research-2014-rp/

Unfortunately, it looks almost perfect but not quite.  For these populations there's an employee number that goes to 80 and a healthy annuitant number that goes from 50-120.  From the report: The term “Healthy Annuitant Tables” refers to tables based on the combined populations of Healthy Retirees and Beneficiaries.

First of all the rates are very different for the years that overlap.  Second, it seems like having the beneficiaries in there then conflates male mortality with female surviving mortality and vice versa as well as age differences between beneficiaries and primaries.

Mgmny

  • Pencil Stache
  • ****
  • Posts: 505
  • Age: 28
  • Location: Northwest 'Burbs of MSP
Back to the issue of longevity data for differnet populations, someone on reddit suggested looking at the Society of Actuaries.  It looks like they have data on different populations including Blue Collar vs White Collar and Bottom Quartile and Top Quartile and split by sex.  It's in an excel doc on this website: 

https://www.soa.org/experience-studies/2014/research-2014-rp/

Unfortunately, it looks almost perfect but not quite.  For these populations there's an employee number that goes to 80 and a healthy annuitant number that goes from 50-120.  From the report: The term “Healthy Annuitant Tables” refers to tables based on the combined populations of Healthy Retirees and Beneficiaries.

First of all the rates are very different for the years that overlap.  Second, it seems like having the beneficiaries in there then conflates male mortality with female surviving mortality and vice versa as well as age differences between beneficiaries and primaries.

It looks like by working as long as possible I have about a 50% less chance of dying in any given year. Is that true?

Hvillian

  • Stubble
  • **
  • Posts: 126
[...]
First of all the rates are very different for the years that overlap.  Second, it seems like having the beneficiaries in there then conflates male mortality with female surviving mortality and vice versa as well as age differences between beneficiaries and primaries.

I haven't looked at the SOA study recently, but two quick points IIRC:
1.  Each of the tables should have a separate Male and Female version.
2.  Yes, mortality does show an increase for people that quick working.

Personally, I don't get too caught up in projecting mortality very exactly.  I just can't figure out what it adds to my projections that is helpful for actual planning purposes.  Longevity risk is real, but the steps to mitigate it wouldn't change that much for me based on these probabilities.  Ask me if I feel the same way when I am a bit older, and get into the relatively higher mortality rates.

maizeman

  • Magnum Stache
  • ******
  • Posts: 2850
Personally, I don't get too caught up in projecting mortality very exactly.  I just can't figure out what it adds to my projections that is helpful for actual planning purposes.  Longevity risk is real, but the steps to mitigate it wouldn't change that much for me based on these probabilities.  Ask me if I feel the same way when I am a bit older, and get into the relatively higher mortality rates.

I think it's less about getting people worried about the risk of extensive longevity and more about making people feel "better" (in quotes intentionally) because, particularly if you FIRE in middle age or later, you're much less likely to go broke than you think if you run all your numbers assuming you're going to live to 90 or 100.

zolotiyeruki

  • Magnum Stache
  • ******
  • Posts: 2747
  • Location: State: Denial
[...]
First of all the rates are very different for the years that overlap.  Second, it seems like having the beneficiaries in there then conflates male mortality with female surviving mortality and vice versa as well as age differences between beneficiaries and primaries.

I haven't looked at the SOA study recently, but two quick points IIRC:
1.  Each of the tables should have a separate Male and Female version.
2.  Yes, mortality does show an increase for people that quick working.

Personally, I don't get too caught up in projecting mortality very exactly.  I just can't figure out what it adds to my projections that is helpful for actual planning purposes.  Longevity risk is real, but the steps to mitigate it wouldn't change that much for me based on these probabilities.  Ask me if I feel the same way when I am a bit older, and get into the relatively higher mortality rates.
Are they concluding that's there's a causal relationship between leaving work and mortality rates, or just a correlation between them?

Western_sean

  • 5 O'Clock Shadow
  • *
  • Posts: 2
] Are they concluding that's there's a causal relationship between leaving work and mortality rates, or just a correlation between them?

The thing to bear in mind here is that retiree mortality is driven by averages. Historically most early retirees do so for health reasons or some type of incapacity so they are arguably substandard as a population. While a general causal relationship may be evident historically - extrapolating that to a specific case today is likely to be inappropriate.

Actuarial data generally is all about large population averages - it is really up to the underwriters to manage specific case deviation from the mean in practice.

CCCA

  • Bristles
  • ***
  • Posts: 487
  • Location: Bay Area, California
  • born before the 80's
    • FI programming
] Are they concluding that's there's a causal relationship between leaving work and mortality rates, or just a correlation between them?

The thing to bear in mind here is that retiree mortality is driven by averages. Historically most early retirees do so for health reasons or some type of incapacity so they are arguably substandard as a population. While a general causal relationship may be evident historically - extrapolating that to a specific case today is likely to be inappropriate.

Actuarial data generally is all about large population averages - it is really up to the underwriters to manage specific case deviation from the mean in practice.


Unfortunately, I'm not sure how to use the SOA data in a way that makes sense.  It would have been nice to prove some different survival curves based upon different demographic attributes or even just the users expectations.  If someone has any thoughts or suggestions to get this data in a useable format, let me know.




LateStarter

  • 5 O'Clock Shadow
  • *
  • Posts: 10
  • Location: UK
Great job CCCA - very useful, and very easy to use.