Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
EnjoyIt
Posts: 1593
Joined: Sun Dec 29, 2013 8:06 pm

Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by EnjoyIt » Fri Sep 28, 2018 4:07 pm

Bogglers,
This is a really nice interactive website that uses the same data as cFIREsim, tracking historical periods from 1871 to 2016 to run through each scenario with inflation adjusted spending and then calculates the probability of success or failure, but adds, the probability of death from social security life expectancy tables depending on your sex. and then spits out a nice picture. The death part (in gray) is very humbling when looking to titrating a very high success rate of not going broke (red.)

I am sharing it with the permission of the creator, CCCA from MMM forums.
https://tinyurl.com/ybw8odwj

Image

RadAudit
Posts: 3062
Joined: Mon May 26, 2008 10:20 am
Location: Second star on the right and straight on 'til morning

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by RadAudit » Fri Sep 28, 2018 4:32 pm

Darn. Probably worked too long.

But, thanks. Helps allay my concerns that my soon-to-be (?) surviving DW might outlive the estate because of an overly conservative asset allocation.
Last edited by RadAudit on Sat Sep 29, 2018 9:28 am, edited 1 time in total.
FI is the best revenge. LBYM. Invest the rest. Stay the course. - PS: The Calvary isn't coming, kids. You are on your own.

User avatar
Blue
Posts: 1151
Joined: Sat Jul 12, 2008 10:18 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Blue » Fri Sep 28, 2018 5:14 pm

My new favorite tool. Very helpful.

edgeagg
Posts: 105
Joined: Tue Jan 23, 2018 1:27 pm
Location: WA-US

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by edgeagg » Fri Sep 28, 2018 5:35 pm

Kind of nice, but cannot displace Portfolio Visualizer as my favorite tool by any means. Would be nice to have:

(1) Median portfolio in the hover
(2) Age in the hover so that I can look at percentiles of the age distribution
ea

User avatar
Sandtrap
Posts: 5379
Joined: Sat Nov 26, 2016 6:32 pm
Location: Hawaii😀 Northern AZ.😳 Retired.

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Sandtrap » Fri Sep 28, 2018 5:48 pm

Very interesting.
Thanks for posting.
According to the chart and projection, the odds are 100% that I'm going to be dead before running out of money :oops:

delamer
Posts: 6284
Joined: Tue Feb 08, 2011 6:13 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by delamer » Fri Sep 28, 2018 6:40 pm

Scott Burns, a now-retired personal finance journalist who introduced me to index funds, used to make the point that the probability of your portfolio “failing” at a ripe old age needed to be balanced with the probability of you (or your spouse) not being around to care.

Here’s his analysis: https://assetbuilder.com/knowledge-cent ... end-freely

And this all ignores any Social Security or pensions, too.

User avatar
Kenkat
Posts: 4369
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Kenkat » Fri Sep 28, 2018 7:09 pm

This simulation has some really nice features - especially the ability to add future income streams such as social security or pensions at specified ages. Based on this, I may be able to retire fairly soon if I needed or wanted to. That’s a pretty nice feeling, even if I will probably work for awhile longer. Nice to have the Johnny Paycheck option - i.e., take this job and shove it, I ain’t workin’ here no more. :D

dknightd
Posts: 965
Joined: Wed Mar 07, 2018 11:57 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by dknightd » Fri Sep 28, 2018 7:17 pm

As near as I can figure, I'll die young, or filthy rich. Or maybe the other way around.

User avatar
Watty
Posts: 14399
Joined: Wed Oct 10, 2007 3:55 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Watty » Fri Sep 28, 2018 7:54 pm

I was playing with it and one thing that might not be obvious is that you can enter a negative number in "extra annual expenses" to simulate cutting your expenses if you need to.


Here is a simple calculator by Vanguard which might also be of interest. It shows the probability that either spouse in a couple will survive a number of years.

https://personal.vanguard.com/us/insigh ... ol?lang=en

Sailorgirl
Posts: 6
Joined: Fri Mar 23, 2018 4:28 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Sailorgirl » Fri Sep 28, 2018 9:47 pm

Longevity, Asset Allocation and Safe withdrawl rates are the cornerstone of fanancial planning. Real life causes each of us to agonize and evaluate probable and potential outcomes.

My parents 92 and 93 are living examples of distribution tails. My Mother had her first heart bypass surgery at 57 and her second at 60. She turned 92 this summer. My father who had bypass at 84 has lived 23 years past his father and 17 years past both of his brothers. He is still going strong! My husband father died at 93 and lived 26 years longer than his father. My parents ran out of money five years ago and are being Assited Living supported by loving son inlaws.

Simulation analysis based on outdated SS tables are helpful but more current tables are desperately needed. One wants to enjoy life while they can without the fear and hesitation that comes with longevity, sequence risk and bear markets.

MnD
Posts: 3793
Joined: Mon Jan 14, 2008 12:41 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by MnD » Fri Sep 28, 2018 10:21 pm

The "4% SWR is bleeding edge risky especially if you retire early" crowd isn't going to like this tool.

randomguy
Posts: 6504
Joined: Wed Sep 17, 2014 9:00 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by randomguy » Fri Sep 28, 2018 10:31 pm

MnD wrote:
Fri Sep 28, 2018 10:21 pm
The "4% SWR is bleeding edge risky especially if you retire early" crowd isn't going to like this tool.
This tool changes nothing for those discussions. If we knew the worst of the next 50 years was going to be as bad as the worst of the past 100, then nobody would be arguing with the 4% rule. Once you start assuming you can do 20-100% worse (i.e. the 3.5 to 2% SWR crowds), you end needing piles of money. And there are enough countries with 2% SWR that you just can't write it off as insanely paranoid. Granted pretty much everyone of those countries was bombed into rubble once or twice during the periods of those low SWR.:)

It should be pointed out that the extreme early retirement are in a different group than the early retirement crowd. Retire at 35 and if in 10 years things are bad, you can get a job and make enough money to make difference. Especially if flipping burgers covers half your expenses:). Retire at 55, and you run into the issue that 10 years you ability to work might be compromised.

User avatar
VictoriaF
Posts: 18604
Joined: Tue Feb 27, 2007 7:27 am
Location: Black Swan Lake

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by VictoriaF » Fri Sep 28, 2018 11:02 pm

An excellent tool, thank you.

But I can't help thinking that ten years ago during the Great Recession the outcome would have looked very differently.

Victoria
WINNER of the 2015 Boglehead Contest. | Every joke has a bit of a joke. ... The rest is the truth. (Marat F)

Ron Scott
Posts: 1090
Joined: Tue Apr 05, 2016 5:38 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Ron Scott » Fri Sep 28, 2018 11:04 pm

MnD wrote:
Fri Sep 28, 2018 10:21 pm
The "4% SWR is bleeding edge risky especially if you retire early" crowd isn't going to like this tool.
Well I’m a proud, card-carrying member of that crowd and I think the tool itself probably works well and is gorgeous. The problem is the dataset used by the tool to make important life decisions: garbage in, garbage out. I have yet to hear a logical argument for the validity of using historical data, much of it from 20th century America, to predict a specific 30-year retirement period in the future.

People like me believe “the 4% rule” might actually work out well...or it might not. (IF the future works out like the past THEN we’ll be OK!) The problem is we simply cannot predict the future, regardless of truly pretty tools like this one, Firecalc et al. I wish we could.
Retirement is a game best played by those prepared for more volatility in the future than has been seen in the past. The solution is not to predict investment losses but to prepare for them.

james22
Posts: 1217
Joined: Tue Aug 21, 2007 2:22 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by james22 » Sat Sep 29, 2018 12:13 am

For later.
This whole episode is likely to end so badly that future children will learn about it in school and shake their heads in wonder at the rank stupidity of it all... Hussman

AlohaJoe
Posts: 3849
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by AlohaJoe » Sat Sep 29, 2018 12:53 am

Sailorgirl wrote:
Fri Sep 28, 2018 9:47 pm
Simulation analysis based on outdated SS tables are helpful but more current tables are desperately needed.
There are more current tables available. Nothing is needed. Most academic analysis use these tables, bot the ones from the Social Security Administration.

http://www.longevityillustrator.org

ignition
Posts: 201
Joined: Sun Dec 11, 2016 11:28 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by ignition » Sat Sep 29, 2018 3:16 am

Ron Scott wrote:
Fri Sep 28, 2018 11:04 pm
MnD wrote:
Fri Sep 28, 2018 10:21 pm
The "4% SWR is bleeding edge risky especially if you retire early" crowd isn't going to like this tool.
Well I’m a proud, card-carrying member of that crowd and I think the tool itself probably works well and is gorgeous. The problem is the dataset used by the tool to make important life decisions: garbage in, garbage out. I have yet to hear a logical argument for the validity of using historical data, much of it from 20th century America, to predict a specific 30-year retirement period in the future.

People like me believe “the 4% rule” might actually work out well...or it might not. (IF the future works out like the past THEN we’ll be OK!) The problem is we simply cannot predict the future, regardless of truly pretty tools like this one, Firecalc et al. I wish we could.
It might not but it will probably work out just fine. Nothing is guaranteed of course.

User avatar
Watty
Posts: 14399
Joined: Wed Oct 10, 2007 3:55 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Watty » Sat Sep 29, 2018 5:50 am

Ron Scott wrote:
Fri Sep 28, 2018 11:04 pm
I have yet to hear a logical argument for the validity of using historical data, much of it from 20th century America, to predict a specific 30-year retirement period in the future.

People like me believe “the 4% rule” might actually work out well...or it might not.
The biggest problem with the “the 4% rule” is that it really doesn't exist.

What there is are some academic studies that have shown what would have worked in the past which a very specific set of assumptions that are not meant to model how people in retirement actually spend their money.

No one has constant expenses and everyone will alter their spending based on how their investments are doing.

When looking at the various models the results are often stated in the odds of success, failure or even broke like this model. For most people "failure" would mean something like having to reduce their spending by 20% when they are in their 70 which might not be fun but would usually be a long way from being broke, homeless and hungry.

I would suspect that a lot of people that do end up broke in retirement were either clearly not financially prepared for it under almost any assumptions, or they had an unfortunate situation with extremely high expenses.

Ron Scott
Posts: 1090
Joined: Tue Apr 05, 2016 5:38 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Ron Scott » Sat Sep 29, 2018 8:05 am

Watty wrote:
Sat Sep 29, 2018 5:50 am
Ron Scott wrote:
Fri Sep 28, 2018 11:04 pm
I have yet to hear a logical argument for the validity of using historical data, much of it from 20th century America, to predict a specific 30-year retirement period in the future.

People like me believe “the 4% rule” might actually work out well...or it might not.
The biggest problem with the “the 4% rule” is that it really doesn't exist.

What there is are some academic studies that have shown what would have worked in the past which a very specific set of assumptions that are not meant to model how people in retirement actually spend their money.

No one has constant expenses and everyone will alter their spending based on how their investments are doing.

When looking at the various models the results are often stated in the odds of success, failure or even broke like this model. For most people "failure" would mean something like having to reduce their spending by 20% when they are in their 70 which might not be fun but would usually be a long way from being broke, homeless and hungry.

I would suspect that a lot of people that do end up broke in retirement were either clearly not financially prepared for it under almost any assumptions, or they had an unfortunate situation with extremely high expenses.
The financial experience of retirees can be dynamic and complex.

I think it’s safe to say the majority of Americans retiring today are not financially prepared for it because they did not or could not save enough. I feel for these people.

There are others who have the means and the desire to adequately prepare and are looking for guidance on how much to save before retiring. Many are willing to delay retirement and continue working to improve their chances of success. To this group I say, do not trust the simplistic output of online retirement planning tools like the one the OP points to, Firecalc, and the various Monte Carlo sims that all crunch the same old US historical datasets. These tools require an unwarranted leap of faith that the past can predict the future and logic does not back that assumption.

The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
Retirement is a game best played by those prepared for more volatility in the future than has been seen in the past. The solution is not to predict investment losses but to prepare for them.

RadAudit
Posts: 3062
Joined: Mon May 26, 2008 10:20 am
Location: Second star on the right and straight on 'til morning

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by RadAudit » Sat Sep 29, 2018 8:13 am

Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less
Now, if I can get the kids to buy into that idea I think that would be one less concern on the plate.
FI is the best revenge. LBYM. Invest the rest. Stay the course. - PS: The Calvary isn't coming, kids. You are on your own.

EddyB
Posts: 557
Joined: Fri May 24, 2013 3:43 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by EddyB » Sat Sep 29, 2018 9:31 am

RadAudit wrote:
Sat Sep 29, 2018 8:13 am
Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less
Now, if I can get the kids to buy into that idea I think that would be one less concern on the plate.
But it’s easy to turn that into trading one pair of blinders for another—as someone must have already pointed out, none of us is guaranteed even another day, and we’ve all heard the tales of people who “one more yeared” themselves out of ever getting to the rewards of retirement. (If you’re one of the people who says, “there’s nothing I would do in retirement that I can’t do now,” that eases the pressure, but it’s not true of everyone.)

User avatar
Sandtrap
Posts: 5379
Joined: Sat Nov 26, 2016 6:32 pm
Location: Hawaii😀 Northern AZ.😳 Retired.

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Sandtrap » Sat Sep 29, 2018 9:55 am

Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
Watty wrote:
Sat Sep 29, 2018 5:50 am
Ron Scott wrote:
Fri Sep 28, 2018 11:04 pm
I have yet to hear a logical argument for the validity of using historical data, much of it from 20th century America, to predict a specific 30-year retirement period in the future.

People like me believe “the 4% rule” might actually work out well...or it might not.
The biggest problem with the “the 4% rule” is that it really doesn't exist.

What there is are some academic studies that have shown what would have worked in the past which a very specific set of assumptions that are not meant to model how people in retirement actually spend their money.

No one has constant expenses and everyone will alter their spending based on how their investments are doing.

When looking at the various models the results are often stated in the odds of success, failure or even broke like this model. For most people "failure" would mean something like having to reduce their spending by 20% when they are in their 70 which might not be fun but would usually be a long way from being broke, homeless and hungry.

I would suspect that a lot of people that do end up broke in retirement were either clearly not financially prepared for it under almost any assumptions, or they had an unfortunate situation with extremely high expenses.
The financial experience of retirees can be dynamic and complex.

I think it’s safe to say the majority of Americans retiring today are not financially prepared for it because they did not or could not save enough. I feel for these people.

There are others who have the means and the desire to adequately prepare and are looking for guidance on how much to save before retiring. Many are willing to delay retirement and continue working to improve their chances of success. To this group I say, do not trust the simplistic output of online retirement planning tools like the one the OP points to, Firecalc, and the various Monte Carlo sims that all crunch the same old US historical datasets. These tools require an unwarranted leap of faith that the past can predict the future and logic does not back that assumption.

The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
Well said.
It would be far easier to project these things if there was an expiration date stamped somewhere on our body. :shock:
j

ignition
Posts: 201
Joined: Sun Dec 11, 2016 11:28 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by ignition » Sat Sep 29, 2018 11:09 am

Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
So how much is enough? 30x, 40x, 50x? Or just keep working until you drop dead?

It's all fine if you enjoy your work, but most people do not unfortunately and would rather do something else. I would much rather take the risk of the 4% rule compared to the risk of working a lot longer and dropping dead 5 years into retirement.

User avatar
willthrill81
Posts: 6154
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by willthrill81 » Sat Sep 29, 2018 11:17 am

EnjoyIt wrote:
Fri Sep 28, 2018 4:07 pm
Bogglers,
This is a really nice interactive website that uses the same data as cFIREsim, tracking historical periods from 1871 to 2016 to run through each scenario with inflation adjusted spending and then calculates the probability of success or failure, but adds, the probability of death from social security life expectancy tables depending on your sex. and then spits out a nice picture. The death part (in gray) is very humbling when looking to titrating a very high success rate of not going broke (red.)

I am sharing it with the permission of the creator, CCCA from MMM forums.
https://tinyurl.com/ybw8odwj

Image
Thanks a lot! It's a very useful way of demonstrating that the risk of dying with a pile of money left behind is far greater than the risk of going broke for nearly everyone. Further, these analyses assume that one would blindly spend down their portfolio to zero, which is obviously a false assumption. Research has shown that most retirees do not significantly draw down their assets over time.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
vineviz
Posts: 2079
Joined: Tue May 15, 2018 1:55 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by vineviz » Sat Sep 29, 2018 11:23 am

willthrill81 wrote:
Sat Sep 29, 2018 11:17 am
Thanks a lot! It's a very useful way of demonstrating that the risk of dying with a pile of money left behind is far greater than the risk of going broke for nearly everyone.
This probably isn’t true for the majority of Americans who have saved exactly $0.00 for their retirement.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
willthrill81
Posts: 6154
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by willthrill81 » Sat Sep 29, 2018 11:25 am

ignition wrote:
Sat Sep 29, 2018 11:09 am
Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
So how much is enough? 30x, 40x, 50x? Or just keep working until you drop dead?
Actually, working until you die solves the entire saving, investing, and withdrawal 'problem' in one fail swoop. :wink:
ignition wrote:
Sat Sep 29, 2018 11:09 am
It's all fine if you enjoy your work, but most people do not unfortunately and would rather do something else. I would much rather take the risk of the 4% rule compared to the risk of working a lot longer and dropping dead 5 years into retirement.
:thumbsup

This is further compounded by the fact that you aren't going to spend your portfolio down to nothing, meaning that 'going broke' is merely an artifact of the analyses conducted and not necessarily akin to reality. Also, starting retirement with a relatively large starting balance, as most Bogleheads do, adds another layer of security beyond the withdrawal rate. For instance, a retiree with a $3M portfolio, for instance, spending $120k annually has a lot more room to reduce their spending if they really needed to than someone with a $1M portfolio and spending $40k annually. The bigger the portfolio that you start with, the higher the withdrawal rate that you can afford to start with.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
willthrill81
Posts: 6154
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by willthrill81 » Sat Sep 29, 2018 11:26 am

vineviz wrote:
Sat Sep 29, 2018 11:23 am
willthrill81 wrote:
Sat Sep 29, 2018 11:17 am
Thanks a lot! It's a very useful way of demonstrating that the risk of dying with a pile of money left behind is far greater than the risk of going broke for nearly everyone.
This probably isn’t true for the majority of Americans who have saved exactly $0.00 for their retirement.
But the majority of Americans aren't reading this thread either. :wink: The OP's analysis assumes that one is starting with a pile of money.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

JackoC
Posts: 275
Joined: Sun Aug 12, 2018 11:14 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by JackoC » Sat Sep 29, 2018 11:31 am

AlohaJoe wrote:
Sat Sep 29, 2018 12:53 am
Sailorgirl wrote:
Fri Sep 28, 2018 9:47 pm
Simulation analysis based on outdated SS tables are helpful but more current tables are desperately needed.
There are more current tables available. Nothing is needed. Most academic analysis use these tables, bot the ones from the Social Security Administration.

http://www.longevityillustrator.org
There are various longevity calculators online. The one you link gives my LE as around 87, the tool in the OP assumes ~81.5. This one gives 95 (just saying my health is 'very good', the one you linked limits it to 'fair' or 'excellent' where I said excellent). The latter asks more questions, such as race, educational/socio-econ background, exercise habits etc.
https://www.blueprintincome.com/tools/l ... ll-i-live/

Those are significant differences. Although, I would assume the probability of broke|alive is more or less a constant. IOW, if you just pull up the tool and leave the financial entries as is (my personal financial entries would be very different) but switch to 'male', the tool says there's a 95.4% chance I'll be dead 95 years after my birth. The 50% given in second longevity is a huge difference in absolute terms. However the OP tool says 0.7% chance I'll be broke, so around 15% chance of broke|alive. If my chance of being live is really ~10 times higher (50% v 4.6%), the probability of broke is probably 10 times higher. But if so my chance of never being broke would only decline from 99.7% to ~93% (and duly noting the comment that this is just under the assumption of no cutback in consumption as one's portfolio gradually goes south). Is that really that significant given all the uncertainties? Some people, I know, think so. Possibly including me: my withdrawal rate is far lower, but not necessarily just to make being broke a tiny %, also to leave a legacy, and also probably for reasons I don't fully understand, maybe it's just miserly tendencies in part. I think this has lots of potential for not strictly rational thought. As long as you don't fool yourself into thinking you're being 100% rational when you aren't, fine.

I think the bigger flaw in this tool is the implicit assumption that long term expected returns now are the long term historical ones. I don't think that's true. Now's are lower by reasonable analysis, IMO, for somebody near retirement anyway.

Ron Scott
Posts: 1090
Joined: Tue Apr 05, 2016 5:38 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Ron Scott » Sat Sep 29, 2018 12:01 pm

ignition wrote:
Sat Sep 29, 2018 11:09 am
Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
So how much is enough? 30x, 40x, 50x? Or just keep working until you drop dead?

It's all fine if you enjoy your work, but most people do not unfortunately and would rather do something else. I would much rather take the risk of the 4% rule compared to the risk of working a lot longer and dropping dead 5 years into retirement.
That question you need to answer for yourself. But the fact that it can be challenging does not make the historical data viable.

I personally ignore the underlying data in these models and use 0% real return for planning purposes.
Retirement is a game best played by those prepared for more volatility in the future than has been seen in the past. The solution is not to predict investment losses but to prepare for them.

ResearchMed
Posts: 7476
Joined: Fri Dec 26, 2008 11:25 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by ResearchMed » Sat Sep 29, 2018 12:25 pm

Sailorgirl wrote:
Fri Sep 28, 2018 9:47 pm

<snip>

Simulation analysis based on outdated SS tables are helpful but more current tables are desperately needed. One wants to enjoy life while they can without the fear and hesitation that comes with longevity, sequence risk and bear markets.
Updated life expectancy tables won't eliminate "the fear and hesitation that comes with longevity, sequence risk and bear markets".
That would just change the schedule.

Unless one does indeed have an expiration date stamped on one's body, there's no way to be precise during what could be years or decades in advance.

Life annuities come closest, but one shouldn't put all of one's pot into that for several reasons. So the same problem remains, but probably with less desperate risks, but possibly also adding more inflation risk.
And some people prefer not to use them at all, of course, which is a separate issue.

RM
This signature is a placebo. You are in the control group.

MnD
Posts: 3793
Joined: Mon Jan 14, 2008 12:41 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by MnD » Sat Sep 29, 2018 12:42 pm

Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
Watty wrote:
Sat Sep 29, 2018 5:50 am
Ron Scott wrote:
Fri Sep 28, 2018 11:04 pm
I have yet to hear a logical argument for the validity of using historical data, much of it from 20th century America, to predict a specific 30-year retirement period in the future.

People like me believe “the 4% rule” might actually work out well...or it might not.
The biggest problem with the “the 4% rule” is that it really doesn't exist.

What there is are some academic studies that have shown what would have worked in the past which a very specific set of assumptions that are not meant to model how people in retirement actually spend their money.

No one has constant expenses and everyone will alter their spending based on how their investments are doing.

When looking at the various models the results are often stated in the odds of success, failure or even broke like this model. For most people "failure" would mean something like having to reduce their spending by 20% when they are in their 70 which might not be fun but would usually be a long way from being broke, homeless and hungry.

I would suspect that a lot of people that do end up broke in retirement were either clearly not financially prepared for it under almost any assumptions, or they had an unfortunate situation with extremely high expenses.
The financial experience of retirees can be dynamic and complex.

I think it’s safe to say the majority of Americans retiring today are not financially prepared for it because they did not or could not save enough. I feel for these people.

There are others who have the means and the desire to adequately prepare and are looking for guidance on how much to save before retiring. Many are willing to delay retirement and continue working to improve their chances of success. To this group I say, do not trust the simplistic output of online retirement planning tools like the one the OP points to, Firecalc, and the various Monte Carlo sims that all crunch the same old US historical datasets. These tools require an unwarranted leap of faith that the past can predict the future and logic does not back that assumption.

The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
The past includes things like the Great Depression, world wars, polio, urban riots, global flu pandemics, 40-year runs of interest rate increases, stagflation, multiple ~50% market declines, presidential assassinations and resignations - you get the picture. This idea that the past just too rosy so we need some ultra-low SWR beyond what history suggests that might add a decade or more of work to achieve, or mean no retirement given that people die in their 50's, 60's and 70's all the time, just doesn't hold water. While the large majority of US households are woefully underfunded for retirement, many Bogleheads will derive only a small portion of the potential utility of their retirement savings. For many the game seems to be money collecting or hoarding versus a plan to save and invest to provide a comfortable retirement income early enough in life that one can still enjoy it.

Ron Scott
Posts: 1090
Joined: Tue Apr 05, 2016 5:38 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Ron Scott » Sat Sep 29, 2018 1:23 pm

MnD wrote:
Sat Sep 29, 2018 12:42 pm
Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
Watty wrote:
Sat Sep 29, 2018 5:50 am
Ron Scott wrote:
Fri Sep 28, 2018 11:04 pm
I have yet to hear a logical argument for the validity of using historical data, much of it from 20th century America, to predict a specific 30-year retirement period in the future.

People like me believe “the 4% rule” might actually work out well...or it might not.
The biggest problem with the “the 4% rule” is that it really doesn't exist.

What there is are some academic studies that have shown what would have worked in the past which a very specific set of assumptions that are not meant to model how people in retirement actually spend their money.

No one has constant expenses and everyone will alter their spending based on how their investments are doing.

When looking at the various models the results are often stated in the odds of success, failure or even broke like this model. For most people "failure" would mean something like having to reduce their spending by 20% when they are in their 70 which might not be fun but would usually be a long way from being broke, homeless and hungry.

I would suspect that a lot of people that do end up broke in retirement were either clearly not financially prepared for it under almost any assumptions, or they had an unfortunate situation with extremely high expenses.
The financial experience of retirees can be dynamic and complex.

I think it’s safe to say the majority of Americans retiring today are not financially prepared for it because they did not or could not save enough. I feel for these people.

There are others who have the means and the desire to adequately prepare and are looking for guidance on how much to save before retiring. Many are willing to delay retirement and continue working to improve their chances of success. To this group I say, do not trust the simplistic output of online retirement planning tools like the one the OP points to, Firecalc, and the various Monte Carlo sims that all crunch the same old US historical datasets. These tools require an unwarranted leap of faith that the past can predict the future and logic does not back that assumption.

The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
The past includes things like the Great Depression, world wars, polio, urban riots, global flu pandemics, 40-year runs of interest rate increases, stagflation, multiple ~50% market declines, presidential assassinations and resignations - you get the picture. This idea that the past just too rosy so we need some ultra-low SWR beyond what history suggests that might add a decade or more of work to achieve, or mean no retirement given that people die in their 50's, 60's and 70's all the time, just doesn't hold water.
The datasets used in these tools typically represent America during the years in which it became the most dominant superpower in the history of the planet. (Talk about bias.) These years also saw a fairly dormant Asia (not anymore) and the advent of prolonged, expensive adolescence...and "retirement" as we know them. But aside from all that, economists are perpetually lousy at predicting the future and they torture all sorts of historical data for a living.

I don't think you need to work for another decade or more as you suggest. And most people planning for retirement, except those with known medical conditions, are not setting goals assuming they die in their 70s. Contrasting the use of bad data with extremes isn't helpful. Working a few extra years grows the kitty and reduces it's needed life at the same time, and spending a bit less probably only means you're spending a bit more than the next guy posting a response to this thread; your world won't end.

No need to go overboard and throw the baby out with the bathwater. A little caution, and a lot of skepticism about "expert predictions" can go a long way.
Retirement is a game best played by those prepared for more volatility in the future than has been seen in the past. The solution is not to predict investment losses but to prepare for them.

User avatar
Kenkat
Posts: 4369
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Kenkat » Sat Sep 29, 2018 1:35 pm

MnD wrote:
Sat Sep 29, 2018 12:42 pm
Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
Watty wrote:
Sat Sep 29, 2018 5:50 am
Ron Scott wrote:
Fri Sep 28, 2018 11:04 pm
I have yet to hear a logical argument for the validity of using historical data, much of it from 20th century America, to predict a specific 30-year retirement period in the future.

People like me believe “the 4% rule” might actually work out well...or it might not.
The biggest problem with the “the 4% rule” is that it really doesn't exist.

What there is are some academic studies that have shown what would have worked in the past which a very specific set of assumptions that are not meant to model how people in retirement actually spend their money.

No one has constant expenses and everyone will alter their spending based on how their investments are doing.

When looking at the various models the results are often stated in the odds of success, failure or even broke like this model. For most people "failure" would mean something like having to reduce their spending by 20% when they are in their 70 which might not be fun but would usually be a long way from being broke, homeless and hungry.

I would suspect that a lot of people that do end up broke in retirement were either clearly not financially prepared for it under almost any assumptions, or they had an unfortunate situation with extremely high expenses.
The financial experience of retirees can be dynamic and complex.

I think it’s safe to say the majority of Americans retiring today are not financially prepared for it because they did not or could not save enough. I feel for these people.

There are others who have the means and the desire to adequately prepare and are looking for guidance on how much to save before retiring. Many are willing to delay retirement and continue working to improve their chances of success. To this group I say, do not trust the simplistic output of online retirement planning tools like the one the OP points to, Firecalc, and the various Monte Carlo sims that all crunch the same old US historical datasets. These tools require an unwarranted leap of faith that the past can predict the future and logic does not back that assumption.

The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
The past includes things like the Great Depression, world wars, polio, urban riots, global flu pandemics, 40-year runs of interest rate increases, stagflation, multiple ~50% market declines, presidential assassinations and resignations - you get the picture. This idea that the past just too rosy so we need some ultra-low SWR beyond what history suggests that might add a decade or more of work to achieve, or mean no retirement given that people die in their 50's, 60's and 70's all the time, just doesn't hold water. While the large majority of US households are woefully underfunded for retirement, many Bogleheads will derive only a small portion of the potential utility of their retirement savings. For many the game seems to be money collecting or hoarding versus a plan to save and invest to provide a comfortable retirement income early enough in life that one can still enjoy it.
The above is why I find this simulator interesting. Here’s my situation: 54 years old, wife is 52. We have about 10x income saved, probably around 14x expenses, with a legit shot at the 25x goal by 65. That said, I also have a pension, and with social security, will probably be able to cover most if not all of our expenses at 65 from that alone.

So, I can’t help to think, as I pile into the car day after day and drive to work, and deal with all of the normal stresses of the work day, and then pile into the car and drive back home, only to repeat day after day, week after week...when do I stop all of this? I’ve been doing this routine for close to 35 years. It’s not terrible, but it would be nice to do something different with life. Do I do this in January when I am 55? Next year? 59? 60? I keep having this nagging feeling that I might not need to work as long as I thought. This simulation makes me think maybe I am right.

Day9
Posts: 626
Joined: Mon Jun 11, 2012 6:22 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Day9 » Sat Sep 29, 2018 1:48 pm

Could we estimate lower expected future returns by simply increasing the "Investment Fees" field by the amount that expected returns will be lower going forward?

Plugged in OP's numbers:
Image

Again but with 2% investment fees, adjusting for 2 percentage point decrease in expected return going forward:
Image
I'm just a fan of the person I got my user name from

User avatar
David Jay
Posts: 5684
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by David Jay » Sat Sep 29, 2018 2:00 pm

Kenkat wrote:
Sat Sep 29, 2018 1:35 pm
The above is why I find this simulator interesting. Here’s my situation: 54 years old, wife is 52. We have about 10x income saved, probably around 14x expenses, with a legit shot at the 25x goal by 65. That said, I also have a pension, and with social security, will probably be able to cover most if not all of our expenses at 65 from that alone.

So, I can’t help to think, as I pile into the car day after day and drive to work, and deal with all of the normal stresses of the work day, and then pile into the car and drive back home, only to repeat day after day, week after week...when do I stop all of this? I’ve been doing this routine for close to 35 years. It’s not terrible, but it would be nice to do something different with life. Do I do this in January when I am 55? Next year? 59? 60? I keep having this nagging feeling that I might not need to work as long as I thought. This simulation makes me think maybe I am right.
That is why I look at it this way:
3. Expense needs of surviving spouse after SS drops to 1 benefit (I use 4% because there will hopefully be a lot of compounding between now and then)
2. Expense needs when both spouses are taking SS benefits. Again 4% because I don’t even plan to start SS until 68 at the earliest.
1. Expense needs to get to SS. I use actual balance for this number. If I need 80,000 a year to get from 62 to 68, I need to have $400,000 at retirement

Add up all 3 and that is what I need.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius

MnD
Posts: 3793
Joined: Mon Jan 14, 2008 12:41 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by MnD » Sat Sep 29, 2018 6:34 pm

Ron Scott wrote:
Sat Sep 29, 2018 1:23 pm
MnD wrote:
Sat Sep 29, 2018 12:42 pm
Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
Watty wrote:
Sat Sep 29, 2018 5:50 am
Ron Scott wrote:
Fri Sep 28, 2018 11:04 pm
I have yet to hear a logical argument for the validity of using historical data, much of it from 20th century America, to predict a specific 30-year retirement period in the future.

People like me believe “the 4% rule” might actually work out well...or it might not.
The biggest problem with the “the 4% rule” is that it really doesn't exist.

What there is are some academic studies that have shown what would have worked in the past which a very specific set of assumptions that are not meant to model how people in retirement actually spend their money.

No one has constant expenses and everyone will alter their spending based on how their investments are doing.

When looking at the various models the results are often stated in the odds of success, failure or even broke like this model. For most people "failure" would mean something like having to reduce their spending by 20% when they are in their 70 which might not be fun but would usually be a long way from being broke, homeless and hungry.

I would suspect that a lot of people that do end up broke in retirement were either clearly not financially prepared for it under almost any assumptions, or they had an unfortunate situation with extremely high expenses.
The financial experience of retirees can be dynamic and complex.

I think it’s safe to say the majority of Americans retiring today are not financially prepared for it because they did not or could not save enough. I feel for these people.

There are others who have the means and the desire to adequately prepare and are looking for guidance on how much to save before retiring. Many are willing to delay retirement and continue working to improve their chances of success. To this group I say, do not trust the simplistic output of online retirement planning tools like the one the OP points to, Firecalc, and the various Monte Carlo sims that all crunch the same old US historical datasets. These tools require an unwarranted leap of faith that the past can predict the future and logic does not back that assumption.

The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
The past includes things like the Great Depression, world wars, polio, urban riots, global flu pandemics, 40-year runs of interest rate increases, stagflation, multiple ~50% market declines, presidential assassinations and resignations - you get the picture. This idea that the past just too rosy so we need some ultra-low SWR beyond what history suggests that might add a decade or more of work to achieve, or mean no retirement given that people die in their 50's, 60's and 70's all the time, just doesn't hold water.
The datasets used in these tools typically represent America during the years in which it became the most dominant superpower in the history of the planet. (Talk about bias.) These years also saw a fairly dormant Asia (not anymore) and the advent of prolonged, expensive adolescence...and "retirement" as we know them. But aside from all that, economists are perpetually lousy at predicting the future and they torture all sorts of historical data for a living.

I don't think you need to work for another decade or more as you suggest. And most people planning for retirement, except those with known medical conditions, are not setting goals assuming they die in their 70s. Contrasting the use of bad data with extremes isn't helpful. Working a few extra years grows the kitty and reduces it's needed life at the same time, and spending a bit less probably only means you're spending a bit more than the next guy posting a response to this thread; your world won't end.

No need to go overboard and throw the baby out with the bathwater. A little caution, and a lot of skepticism about "expert predictions" can go a long way.
The odds of a 56 year old couple (M/F) _both_ being alive at age 77 is only 50:50.
http://eyebonds.info/downloads/pages/Od ... Alive.html

The odds of them both being alive, active and mentally and physically healthy at 77 are significantly lower than that. No-one knows how many years of active healthy life they have left and it's safe to say the earlier years of retirement are going to be a much more pleasant ride than the later ones. At age 56 I'm in a position of having vastly more than typical households this age and certainly "enough" by reasonable standards. But probably not enough to satisfy the "work a few more years" and/or SWR in the 2.X to low 3.X range crowd. Working more years, saving more, spending less to overcompensate beyond the worst series in history is not anything resembling a win-win in my book. More like locking in a horribly bad sequence of returns retirement plan from Day 1 of retirement before you even experience one.

User avatar
willthrill81
Posts: 6154
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by willthrill81 » Sat Sep 29, 2018 7:45 pm

MnD wrote:
Sat Sep 29, 2018 6:34 pm
Ron Scott wrote:
Sat Sep 29, 2018 1:23 pm
MnD wrote:
Sat Sep 29, 2018 12:42 pm
Ron Scott wrote:
Sat Sep 29, 2018 8:05 am
Watty wrote:
Sat Sep 29, 2018 5:50 am


The biggest problem with the “the 4% rule” is that it really doesn't exist.

What there is are some academic studies that have shown what would have worked in the past which a very specific set of assumptions that are not meant to model how people in retirement actually spend their money.

No one has constant expenses and everyone will alter their spending based on how their investments are doing.

When looking at the various models the results are often stated in the odds of success, failure or even broke like this model. For most people "failure" would mean something like having to reduce their spending by 20% when they are in their 70 which might not be fun but would usually be a long way from being broke, homeless and hungry.

I would suspect that a lot of people that do end up broke in retirement were either clearly not financially prepared for it under almost any assumptions, or they had an unfortunate situation with extremely high expenses.
The financial experience of retirees can be dynamic and complex.

I think it’s safe to say the majority of Americans retiring today are not financially prepared for it because they did not or could not save enough. I feel for these people.

There are others who have the means and the desire to adequately prepare and are looking for guidance on how much to save before retiring. Many are willing to delay retirement and continue working to improve their chances of success. To this group I say, do not trust the simplistic output of online retirement planning tools like the one the OP points to, Firecalc, and the various Monte Carlo sims that all crunch the same old US historical datasets. These tools require an unwarranted leap of faith that the past can predict the future and logic does not back that assumption.

The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
The past includes things like the Great Depression, world wars, polio, urban riots, global flu pandemics, 40-year runs of interest rate increases, stagflation, multiple ~50% market declines, presidential assassinations and resignations - you get the picture. This idea that the past just too rosy so we need some ultra-low SWR beyond what history suggests that might add a decade or more of work to achieve, or mean no retirement given that people die in their 50's, 60's and 70's all the time, just doesn't hold water.
The datasets used in these tools typically represent America during the years in which it became the most dominant superpower in the history of the planet. (Talk about bias.) These years also saw a fairly dormant Asia (not anymore) and the advent of prolonged, expensive adolescence...and "retirement" as we know them. But aside from all that, economists are perpetually lousy at predicting the future and they torture all sorts of historical data for a living.

I don't think you need to work for another decade or more as you suggest. And most people planning for retirement, except those with known medical conditions, are not setting goals assuming they die in their 70s. Contrasting the use of bad data with extremes isn't helpful. Working a few extra years grows the kitty and reduces it's needed life at the same time, and spending a bit less probably only means you're spending a bit more than the next guy posting a response to this thread; your world won't end.

No need to go overboard and throw the baby out with the bathwater. A little caution, and a lot of skepticism about "expert predictions" can go a long way.
The odds of a 56 year old couple (M/F) _both_ being alive at age 77 is only 50:50.
http://eyebonds.info/downloads/pages/Od ... Alive.html

The odds of them both being alive, active and mentally and physically healthy at 77 are significantly lower than that. No-one knows how many years of active healthy life they have left and it's safe to say the earlier years of retirement are going to be a much more pleasant ride than the later ones. At age 56 I'm in a position of having vastly more than typical households this age and certainly "enough" by reasonable standards. But probably not enough to satisfy the "work a few more years" and/or SWR in the 2.X to low 3.X range crowd. Working more years, saving more, spending less to overcompensate beyond the worst series in history is not anything resembling a win-win in my book. More like locking in a horribly bad sequence of returns retirement plan from Day 1 of retirement before you even experience one.
:thumbsup

I think that if one has enough flexibility, it may actually be worthwhile to consider front-loading withdrawals. We certainly plan on doing that. I want to enjoy money when I'm both alive and well enough to enjoy it.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

EnjoyIt
Posts: 1593
Joined: Sun Dec 29, 2013 8:06 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by EnjoyIt » Sat Sep 29, 2018 8:24 pm

Kenkat wrote:
Sat Sep 29, 2018 1:35 pm


The above is why I find this simulator interesting. Here’s my situation: 54 years old, wife is 52. We have about 10x income saved, probably around 14x expenses, with a legit shot at the 25x goal by 65. That said, I also have a pension, and with social security, will probably be able to cover most if not all of our expenses at 65 from that alone.

So, I can’t help to think, as I pile into the car day after day and drive to work, and deal with all of the normal stresses of the work day, and then pile into the car and drive back home, only to repeat day after day, week after week...when do I stop all of this? I’ve been doing this routine for close to 35 years. It’s not terrible, but it would be nice to do something different with life. Do I do this in January when I am 55? Next year? 59? 60? I keep having this nagging feeling that I might not need to work as long as I thought. This simulation makes me think maybe I am right.
I get it, and if your pension and SS cover all your needs it may be reasonable and possible to spend down some of your savings right up to the day you start collecting SS and your pension.

It is graphs like the ones above that show our mortality and just how stupid it may be to save a few extra dollars one probably won’t need. But, fear of not having enough is a very strong feeling very difficult to break.

User avatar
LadyGeek
Site Admin
Posts: 48672
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by LadyGeek » Sat Sep 29, 2018 8:47 pm

The bottom of the chart is cutoff in Firefox.

For those with a software background - The chart is embedded in an <iframe>. Hover over the chart, right-click --> This Frame --> (Select how you want to view the frame).
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.

MnD
Posts: 3793
Joined: Mon Jan 14, 2008 12:41 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by MnD » Sun Sep 30, 2018 9:16 am

willthrill81 wrote:
Sat Sep 29, 2018 7:45 pm
MnD wrote:
Sat Sep 29, 2018 6:34 pm
Ron Scott wrote:
Sat Sep 29, 2018 1:23 pm
MnD wrote:
Sat Sep 29, 2018 12:42 pm
Ron Scott wrote:
Sat Sep 29, 2018 8:05 am


The financial experience of retirees can be dynamic and complex.

I think it’s safe to say the majority of Americans retiring today are not financially prepared for it because they did not or could not save enough. I feel for these people.

There are others who have the means and the desire to adequately prepare and are looking for guidance on how much to save before retiring. Many are willing to delay retirement and continue working to improve their chances of success. To this group I say, do not trust the simplistic output of online retirement planning tools like the one the OP points to, Firecalc, and the various Monte Carlo sims that all crunch the same old US historical datasets. These tools require an unwarranted leap of faith that the past can predict the future and logic does not back that assumption.

The safe thing for these people to do is to admit uncertainty and overcompensate before retirement by planning to save more and spend less. If the future turns out to be like the past they will have worked a bit more and have more money than they need. If the future is not as rosy as the past they will have a cushion. It’s as close to a win-win you will get.

We do not get a lot of do-overs in retirement, so better safe than sorry...
The past includes things like the Great Depression, world wars, polio, urban riots, global flu pandemics, 40-year runs of interest rate increases, stagflation, multiple ~50% market declines, presidential assassinations and resignations - you get the picture. This idea that the past just too rosy so we need some ultra-low SWR beyond what history suggests that might add a decade or more of work to achieve, or mean no retirement given that people die in their 50's, 60's and 70's all the time, just doesn't hold water.
The datasets used in these tools typically represent America during the years in which it became the most dominant superpower in the history of the planet. (Talk about bias.) These years also saw a fairly dormant Asia (not anymore) and the advent of prolonged, expensive adolescence...and "retirement" as we know them. But aside from all that, economists are perpetually lousy at predicting the future and they torture all sorts of historical data for a living.

I don't think you need to work for another decade or more as you suggest. And most people planning for retirement, except those with known medical conditions, are not setting goals assuming they die in their 70s. Contrasting the use of bad data with extremes isn't helpful. Working a few extra years grows the kitty and reduces it's needed life at the same time, and spending a bit less probably only means you're spending a bit more than the next guy posting a response to this thread; your world won't end.

No need to go overboard and throw the baby out with the bathwater. A little caution, and a lot of skepticism about "expert predictions" can go a long way.
The odds of a 56 year old couple (M/F) _both_ being alive at age 77 is only 50:50.
http://eyebonds.info/downloads/pages/Od ... Alive.html

The odds of them both being alive, active and mentally and physically healthy at 77 are significantly lower than that. No-one knows how many years of active healthy life they have left and it's safe to say the earlier years of retirement are going to be a much more pleasant ride than the later ones. At age 56 I'm in a position of having vastly more than typical households this age and certainly "enough" by reasonable standards. But probably not enough to satisfy the "work a few more years" and/or SWR in the 2.X to low 3.X range crowd. Working more years, saving more, spending less to overcompensate beyond the worst series in history is not anything resembling a win-win in my book. More like locking in a horribly bad sequence of returns retirement plan from Day 1 of retirement before you even experience one.
:thumbsup

I think that if one has enough flexibility, it may actually be worthwhile to consider front-loading withdrawals. We certainly plan on doing that. I want to enjoy money when I'm both alive and well enough to enjoy it.
A percentage of portfolio SWR effectively does that. 5% of portfolio balance with a 3% inflation-adjusted floor provides both additional security from portfolio failure and much more utility on average and at the median from a portfolio of a given size. The cost of course is the flexibility. If 2019 ends up being 1966 you'll be on the 3% floor in no-time and probably stay there for your entire retirement. But 3% inflation-adjusted won't look or feel bad compared to your peers in this scenario.

A low fixed SWR approach implements a poor retirement income outcome for every sequence of returns. You can reverse course and start spending more later but by then you may have gone past the "go-go" years of retirement and maybe are into the "go-slow" or "no-go" years. "Working longer" guarantees eliminating XX% of your retirement and and a higher XX% of the best years (healthy, active, mentally fit) of your retirement. Working a few more years and that loss might be "10% and 15%" or "60% and 80%", but in any event it won't be an insignificant percentage loss.

User avatar
Kenkat
Posts: 4369
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Kenkat » Sun Sep 30, 2018 10:11 am

EnjoyIt wrote:
Sat Sep 29, 2018 8:24 pm
But, fear of not having enough is a very strong feeling very difficult to break.
Yes this is exactly it. How much are you comfortable spending down because once you shut down the money making engine of your career, it is not always easy to restart it.

EnjoyIt
Posts: 1593
Joined: Sun Dec 29, 2013 8:06 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by EnjoyIt » Sun Sep 30, 2018 2:17 pm

Kenkat wrote:
Sun Sep 30, 2018 10:11 am
EnjoyIt wrote:
Sat Sep 29, 2018 8:24 pm
But, fear of not having enough is a very strong feeling very difficult to break.
Yes this is exactly it. How much are you comfortable spending down because once you shut down the money making engine of your career, it is not always easy to restart it.
I that really true? Is the ability to make money really shut down?

One can always do some credit card and bank churning, drive for uber, dog sit, baby sit if they really wanted to, sell some junk on eBay, etc.
How many years of one's life are we willing to give up to decrease the sequence of return risk by 1% or 2%. Is it worth giving up 5% of your free time for 1% of risk reduction? What about 15%? At 60 years old on may have about 25 years left. Is it worth it to work another 5 years to 65 which is 20% of one's remaining life to decrease the risk of running out of money from 96% to 99%? Personally I don't think so, but many on this forum don't think like I do.

Life is short and you can't account for every risk out there. Otherwise we might as well stay at home, grow out our finger nails and start collecting our urine in jars.

EnjoyIt
Posts: 1593
Joined: Sun Dec 29, 2013 8:06 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by EnjoyIt » Sun Sep 30, 2018 2:20 pm

Day9 wrote:
Sat Sep 29, 2018 1:48 pm
Could we estimate lower expected future returns by simply increasing the "Investment Fees" field by the amount that expected returns will be lower going forward?

Plugged in OP's numbers:
Again but with 2% investment fees, adjusting for 2 percentage point decrease in expected return going forward:
Image
If returns used to be about 8.7% historically at a 60/40 portfolio after inflation, returns of 2% less would be 23% worse than what we have ever seen. Is that very likely over the next 40 years? Although I can not predict the future, I suspect the odds are very low.
Try changing it around and making it 1% with maybe 15k social security at 70 years old. Then instead of 4% make it 3.7% withdrawal rate which is 1,081K and you end up with this which looks amazing to me. We do not need these crazy sub 3% withdrawal rates I keep hearing about on these forums. That is unless I am looking to just keep working and die in my office.

https://tinyurl.com/y8gsbmnv

User avatar
willthrill81
Posts: 6154
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by willthrill81 » Sun Sep 30, 2018 3:08 pm

EnjoyIt wrote:
Sun Sep 30, 2018 2:17 pm
How many years of one's life are we willing to give up to decrease the sequence of return risk by 1% or 2%. Is it worth giving up 5% of your free time for 1% of risk reduction? What about 15%? At 60 years old on may have about 25 years left. Is it worth it to work another 5 years to 65 which is 20% of one's remaining life to decrease the risk of running out of money from 96% to 99%? Personally I don't think so, but many on this forum don't think like I do.
I agree that this is the logical way to consider the situation. If one's life expectancy is 85, for instance, then working from age 65 to 67 consumes 10% of your remaining life expectancy for what may be a tiny improvement in the expected 'success' rate of your withdrawal strategy. On top of that, the years spent doing this are likely to be the healthiest, perhaps the most potentially enjoyable, years you have remaining. Further, the problem with discussions of 'success' in this context are a misleading artifact of Monte Carlo analyses. No sane person blindly spends down their portfolio to zero. For most Bogleheads retiring with over $1 million plus Social Security, 'failure' means only spending half as much on their grandchildren, traveling, nice cars, etc. as they had originally planned. It most definitely does not equate to going broke, financial ruin, and eating cat food.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

JackoC
Posts: 275
Joined: Sun Aug 12, 2018 11:14 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by JackoC » Sun Sep 30, 2018 4:22 pm

EnjoyIt wrote:
Sun Sep 30, 2018 2:20 pm
Day9 wrote:
Sat Sep 29, 2018 1:48 pm
Could we estimate lower expected future returns by simply increasing the "Investment Fees" field by the amount that expected returns will be lower going forward?
If returns used to be about 8.7% historically at a 60/40 portfolio after inflation, returns of 2% less would be 23% worse than what we have ever seen. Is that very likely over the next 40 years? Although I can not predict the future, I suspect the odds are very low.


https://tinyurl.com/y8gsbmnv
The return on 60/40 *after* inflation since 1871-2010 was 5.4%. The tool implies it uses data from 1871 (ie Schiller's data, which a lot of things do), maybe with a few more years added which would bump it up a little.
https://www.researchaffiliates.com/docu ... Return.pdf

Past safe bond returns are irrelevant to current expected returns out to the horizon for which safe bonds already exist, assuming they remain safe. The 30 yr TIPS yields ~1% real, that's the expected real safe bond return, basically (give or take coupon reinvestment rate). An investor can speculate (use shorter maturity bonds and roll them hoping yields rise, credit risky bonds hoping downgrade/defaults are less than expected, nominal bonds hoping inflation is lower than expected) to get a higher (or lower) return from bonds, but that's the benchmark: real long term yields now. They are significantly lower than historical.

Which should suggest how questionable it is to assume expected stock returns are equal to historical regardless of current valuation. I would estimate stock expected return now around 3-4% real. 1/CAPE would be 3.3% real expected return. Add something for foreign stocks not being as expensive in general, some general handwaving :D etc and say 4%

Then 60/40 expected return would be around 3% real (with rebalancing the expected return would probably be a little higher than .6*4%+.4*1%). IOW 2% less than historical for today's expected returns on 60/40 is not particularly conservative IMO, it's roughly equivalent to a 5% real expected return on stocks (Jeremy Siegel says 5.5% and he's a big time perma-bull), though I didn't think it worth complaining that it was too aggressive either.

visualguy
Posts: 859
Joined: Thu Jan 30, 2014 1:32 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by visualguy » Sun Sep 30, 2018 4:56 pm

As often mentioned, expenses related to health and long-term care are the big issue, and they can be huge and hard to predict. I don't feel comfortable ignoring this whole thing, so I add significant extra money to the 30X (or whatever) expenses thing. I also add money for other unpredictable large expenses - fire or earthquake where the insurance doesn't cover the whole thing, something really bad happening to someone in the family and help is needed, etc. Basically, it's a sum of extra money to cover whatever life throws at us. I don't know what those things will be exactly and how much they will cost, but I know there will be some such things, most likely related to health and long-term care at the very least.

I've seen much of this with my family... My grandmother spent years in a nursing home with Alzheimer's. My parents are currently in a situation where they can't take care of themselves anymore. A variety of debilitating health issues with my dad, and early Alzheimer's with my mom. My dad just turned 80 and my mom is in her mid-70s. They'll have good care because there's family to manage it, and my parents own properties that we will liquidate to pay for their care which is expected to be very expensive.

If you don't have family to provide some help, it's even more complicated and expensive. Maybe a CCRC is the way to go, but that's big bucks, so you need to factor that into the math.

I read that having a mother with Alzheimer's increases significantly your chances of getting it too. My grandmother had it (started in her 70s), my mother has it (also started in her 70s), so it's somewhat frightening for me... The money needs to be there to deal with it if and when it happens, and also to continue supporting my wife who is likely to outlive me since she's younger, and she may have her own expensive health and long-term care issues, so the money needs to be there for that as well.

EnjoyIt
Posts: 1593
Joined: Sun Dec 29, 2013 8:06 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by EnjoyIt » Sun Sep 30, 2018 5:03 pm

JackoC wrote:
Sun Sep 30, 2018 4:22 pm
EnjoyIt wrote:
Sun Sep 30, 2018 2:20 pm
Day9 wrote:
Sat Sep 29, 2018 1:48 pm
Could we estimate lower expected future returns by simply increasing the "Investment Fees" field by the amount that expected returns will be lower going forward?
If returns used to be about 8.7% historically at a 60/40 portfolio after inflation, returns of 2% less would be 23% worse than what we have ever seen. Is that very likely over the next 40 years? Although I can not predict the future, I suspect the odds are very low.


https://tinyurl.com/y8gsbmnv
The return on 60/40 *after* inflation since 1871-2010 was 5.4%. The tool implies it uses data from 1871 (ie Schiller's data, which a lot of things do), maybe with a few more years added which would bump it up a little.
https://www.researchaffiliates.com/docu ... Return.pdf

Past safe bond returns are irrelevant to current expected returns out to the horizon for which safe bonds already exist, assuming they remain safe. The 30 yr TIPS yields ~1% real, that's the expected real safe bond return, basically (give or take coupon reinvestment rate). An investor can speculate (use shorter maturity bonds and roll them hoping yields rise, credit risky bonds hoping downgrade/defaults are less than expected, nominal bonds hoping inflation is lower than expected) to get a higher (or lower) return from bonds, but that's the benchmark: real long term yields now. They are significantly lower than historical.

Which should suggest how questionable it is to assume expected stock returns are equal to historical regardless of current valuation. I would estimate stock expected return now around 3-4% real. 1/CAPE would be 3.3% real expected return. Add something for foreign stocks not being as expensive in general, some general handwaving :D etc and say 4%

Then 60/40 expected return would be around 3% real (with rebalancing the expected return would probably be a little higher than .6*4%+.4*1%). IOW 2% less than historical for today's expected returns on 60/40 is not particularly conservative IMO, it's roughly equivalent to a 5% real expected return on stocks (Jeremy Siegel says 5.5% and he's a big time perma-bull), though I didn't think it worth complaining that it was too aggressive either.
My mistake on the 8.7%..It is 8.4% nominal or 5.4% real. Either way, we are not talking about returns over only 5-10 years. We are talking about returns over 20-30 years. I don't know about you, but to me it appears there is a lot of crystal ball utilization in what is being predicted. 3% real returns would mean that we should expect worse than 56% of our average over the last century for the next 20-30 years which seams like a pretty strong prediction.

And hey, maybe you are right. If that is the case I would rather fly to Europe every other year as opposed to once a year if it means not wasting another 10-20% of my healthy life saving for that extra vacation.

User avatar
Kenkat
Posts: 4369
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by Kenkat » Sun Sep 30, 2018 6:33 pm

EnjoyIt wrote:
Sun Sep 30, 2018 2:17 pm
Kenkat wrote:
Sun Sep 30, 2018 10:11 am
EnjoyIt wrote:
Sat Sep 29, 2018 8:24 pm
But, fear of not having enough is a very strong feeling very difficult to break.
Yes this is exactly it. How much are you comfortable spending down because once you shut down the money making engine of your career, it is not always easy to restart it.
I that really true? Is the ability to make money really shut down?

One can always do some credit card and bank churning, drive for uber, dog sit, baby sit if they really wanted to, sell some junk on eBay, etc.
How many years of one's life are we willing to give up to decrease the sequence of return risk by 1% or 2%. Is it worth giving up 5% of your free time for 1% of risk reduction? What about 15%? At 60 years old on may have about 25 years left. Is it worth it to work another 5 years to 65 which is 20% of one's remaining life to decrease the risk of running out of money from 96% to 99%? Personally I don't think so, but many on this forum don't think like I do.

Life is short and you can't account for every risk out there. Otherwise we might as well stay at home, grow out our finger nails and start collecting our urine in jars.
I didn’t say it was shut down, I said it was hard to restart. I should add restart at your current level of salary, benefits, etc. Credit card churning, eBay, Uber or dog sitting do not come with medical and dental benefits, pension, 401k matching and what is for me a pretty nice salary. Sure, I’d dog sit some day, but it’s probably going to pay $10 or $15 dollars an hour. I’d be fine with that, actually, as long as I felt like I had enough money saved and was just doing it for a little extra pocket cash and because I enjoy those doggone doggies.

JackoC
Posts: 275
Joined: Sun Aug 12, 2018 11:14 am

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by JackoC » Sun Sep 30, 2018 8:27 pm

EnjoyIt wrote:
Sun Sep 30, 2018 5:03 pm

My mistake on the 8.7%..It is 8.4% nominal or 5.4% real. Either way, we are not talking about returns over only 5-10 years. We are talking about returns over 20-30 years. I don't know about you, but to me it appears there is a lot of crystal ball utilization in what is being predicted. 3% real returns would mean that we should expect worse than 56% of our average over the last century for the next 20-30 years which seams like a pretty strong prediction.

And hey, maybe you are right. If that is the case I would rather fly to Europe every other year as opposed to once a year if it means not wasting another 10-20% of my healthy life saving for that extra vacation.
Again for bonds now the expected return is definitely lower than historical. Not for the next 5 or 10 yrs, but all the way out to the longest bonds, 30 yr TIPS yields 1%. It's hard to see a reasonable argument to assume more than the yield of bonds (more or less anyway) for expected return out to their maturity. And yields now *are* about 2% lower than the historical realized return of bonds. IOW the real yield on bonds now is more like 2/3's lower than the historical average realized return...yet that's clearly what it is. So that 'it's X% lower' argument is not very compelling IMO.

For stocks of course it's fuzzier, inherently. But I think the bond case illustrates the general idea that there's nothing that special about the historical return in estimating the expected return. Stock like bond valuations are higher now than they usually have been, implying a lower expected return than historical. How much lower is open to reasonable debate, but there just isn't a lot of logic in assuming the expected return of a given financial asset will be the historical return, as often as that assumption is employed. That would be true if assets didn't ever become steadily cheaper or more expensive relative to their payoffs, but both stocks and bonds have, become steadily more expensive relative to their payoffs in last few decades. Assuming 'the market knows what it's doing'* that means lower expected return now than historical return, and 2% lower is somewhere in the ballpark, though not as conservative as I assume for my own planning.

*assuming the market is mispricing and valuations will fall, the expected return should be lower still, a lot more than 2% below historical, anyway I don't assume that.

EnjoyIt
Posts: 1593
Joined: Sun Dec 29, 2013 8:06 pm

Re: Rich, Broke or Dead: Visualizing probabilities of outcomes in early retirement

Post by EnjoyIt » Sun Sep 30, 2018 11:30 pm

JackoC,
I get it. I really do. I get the fear that this time is different. The math makes it so compelling. Maybe the answer is to buy an annuity and risk a little more in equities. Maybe it is being ready and able to spend just a little bit less if the markets are not favorable. I would be okay with planning only 1 overseas trip every other year instead of every year. I would also be okay dog sitting for an extra $1-$2K/yr. What I am not okay with is wasting 15-20% of my healthy and active life working so that I have a better garuantee that I won’t have to do the above things. It kinda seams foolish even writing it down right now.

Post Reply