Help with Retirement Projection Assumptions

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
accbh
Posts: 40
Joined: Tue Jul 09, 2013 7:32 pm

Help with Retirement Projection Assumptions

Post by accbh »

I'm 45 and hope to be FI in the next couple of years so I'm looking for opinions on some of the assumptions I'm making in my early retirement spreadsheets. Please review the items below and let me know if any of these things seem out of line, or if there is something I may be overlooking.

In my calculations, I am assuming:
1) My ACA premium (currently calculated at $8000) will not be subsidized in the future
2) The ACA premium will increase by 5 4% per year due to age + 5 4% per year due to premium increases
3) Medicare and Medigap costs will increase by 5% per year
4) Yearly expense projections include an additional out-of-pocket health insurance expense of $4000 (max out-of-pocket includes deductibles, co-pays, etc as listed on the benefits summary), adjusted with inflation
5) Stocks return 4% (real) per year (the portfolio will survive as low as 2% per year)
6) Bonds return 0.5% (real) per year
7) Zero social security income
8) Zero taxes during retirement. (I don't have a lot of unrealized gains at this point, and I'm assuming current tax law.)
9) My yearly expense projection includes:
a) $1000/yr for car repairs
b) $2000/yr for general home upgrades and home appliance maintenance/replacement
c) $6000/yr for house repairs including roof maintenance, slab leaks, plumbing and HVAC maintenance, new flooring when necessary, termite treatments, etc
10) Social security income will cover medicare and medigap premiums, and taxes during retirement (if any). Any SS left over will be an unexpected bonus.

Regarding the last item, I listed things that do not normally show up as yearly expenses (such as needing a new $40,000 roof after 25 years). I realize that looking for my omissions is difficult without seeing a detailed list of every expense I have, but I'm just looking for suggestions on large expenses that may come up infrequently (or maybe 20+ years from now) that I may not have considered. Thank you for any insight you can give me as I plan my future.

Update: After receiving many great responses in this thread, I have updated my spreadsheet assumptions as reflected above. I agree with AlohaJoe and flyingaway that medical premiums cannot indefinitely increase at current rates, so assumption #2 was changed to reflect that. Most of the skepticism in this thread centers around assumptions #7 and #8, so I replaced it with assumption #10 which may be slightly more realistic (although somewhat less conservative)
Last edited by accbh on Mon Jan 16, 2017 5:57 pm, edited 6 times in total.
User avatar
PhysicianOnFIRE
Posts: 470
Joined: Fri Jan 08, 2016 2:46 pm
Location: Up North

Re: Help with Retirement Projection Assumptions

Post by PhysicianOnFIRE »

Congrats on approaching your financial goals at 45. You're clearly doing well for yourself!

You're free to make assumptions, and a spreadsheet will demand it, but any good plan will have flexibility and contingency plans if the assumptions turn out to be wildly inaccurate. Some of them will -- perhaps in your favor, perhaps not.

The ACA is clearly on the chopping block, and we have no idea what, if anything, might replace it.

$4,000 for health insurance without a subsidy seems awfully low. I assume this covers you only (no family)?

Will you not receive SS, or do you plan to consider it a bonus?

Is your money invested in a way that it can be easily accessed without incurring taxes?

I had brake work done on an '06 Chevy this year. $1,200. Hit a deer with our minivan. $1,000. Might want to budget more there.

If you haven't yet, track annual spending. Consider yourself FI when you've got 25x to 33x your anticipated annual spending in retirement.

:beer
-PoF
delamer
Posts: 17348
Joined: Tue Feb 08, 2011 5:13 pm

Re: Help with Retirement Projection Assumptions

Post by delamer »

Consider budgeting for a new car every 5-10 years too (or whatever time frame fits your lifestyle).

The majority of current retirees depend on Social Security for all or a significant portion of their income. It is overly pessimistic to assume that you won't receive anything. Maybe less than projected, but not zero.

Otherwise I think your assumptions are reasonable. However, I would run additional scenarios with lower assumptions for real returns to look at worsr-case scenarios.
One thing that humbles me deeply is to see that human genius has its limits while human stupidity does not. - Alexandre Dumas, fils
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Help with Retirement Projection Assumptions

Post by Rodc »

PhysicianOnFIRE wrote:Congrats on approaching your financial goals at 45. You're clearly doing well for yourself!

You're free to make assumptions, and a spreadsheet will demand it, but any good plan will have flexibility and contingency plans if the assumptions turn out to be wildly inaccurate. Some of them will -- perhaps in your favor, perhaps not.

The ACA is clearly on the chopping block, and we have no idea what, if anything, might replace it.

$4,000 for health insurance without a subsidy seems awfully low. I assume this covers you only (no family)?

Will you not receive SS, or do you plan to consider it a bonus?

Is your money invested in a way that it can be easily accessed without incurring taxes?

I had brake work done on an '06 Chevy this year. $1,200. Hit a deer with our minivan. $1,000. Might want to budget more there.

If you haven't yet, track annual spending. Consider yourself FI when you've got 25x to 33x your anticipated annual spending in retirement.

:beer
-PoF
All that.

You and I have no idea whatsoever what is going to happen with healthcare costs.

30, even 20 years ago no one knew or predicted (as far as I know) that healthcare costs were going to blow up. 20 years from now it will be something else - we just don't know what yet.

Retiring at 45 requires either an ability to live very very frugally if needed and lots buffer in your assumptions, or keeping up skills so you can jump back in to work if needed.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Topic Author
accbh
Posts: 40
Joined: Tue Jul 09, 2013 7:32 pm

Re: Help with Retirement Projection Assumptions

Post by accbh »

Thanks PoF. I just read a few of your posts (including the one linking to the SWR series) and was hoping you'd be one of the people responding to my thread :)
PhysicianOnFIRE wrote:The ACA is clearly on the chopping block, and we have no idea what, if anything, might replace it.

$4,000 for health insurance without a subsidy seems awfully low. I assume this covers you only (no family)?
[Sorry for the confusion, my original post has been edited for clarity] Regarding the ACA, I am calculating future premiums based on a current unsubsidized premium for a 45yo (about $8000 per year currently and increasing this number by 10% per year.) Per assumption #4, I am factoring in an ADDITIONAL $4000 per year for out-of-pocket costs for deductables, copays, etc. Per assumption #1, I am assuming that subsidies will not exist for me, so I am basing my premium on the full premium cost. If the ACA is replaced in the future, this is still my "best guess" as to what it will cost because the big variable is probably the subsidies.
PhysicianOnFIRE wrote:Will you not receive SS, or do you plan to consider it a bonus?
I will probably receive SS, but I plan to consider it a bonus for the following reasons: 1) I believe future benefits will be cut significantly, 2) I believe the age of receiving benefits will increase, 3) it may be means-tested in the future, and 4) I plan for any SS received to go to covering errors/unforeseen expenses of assumptions #3 #8, and #9.
PhysicianOnFIRE wrote:Is your money invested in a way that it can be easily accessed without incurring taxes?
My taxable funds have very little unrealized gains and assuming the current tax structure, anything that I do pull out should be minimally taxed.
PhysicianOnFIRE wrote:If you haven't yet, track annual spending. Consider yourself FI when you've got 25x to 33x your anticipated annual spending in retirement.
I have been tracking annual spending very closely. The problem I have with typical 25x-33x spending estimates is that I believe health insurance/costs will grow at a rate far greater than inflation. Here is the problem in a nutshell: Today my medical premiums are 15% of total expenses. In 20 years, they will be 47% of expenses, using assumption #2, because health costs are projected to dramatically outpace inflation. How can any SWR calculation account for this properly?
delamer wrote:Consider budgeting for a new car every 5-10 years too (or whatever time frame fits your lifestyle).
I've given a lot of thought to this, and I'm not sure that I view a new car as a necessary expense. If investments do significantly better than the conservative estimates that I projected, I may get a new car. If not, I will keep the one I have. After all, having a reliable car may not be that important if I don't need to drive to work. :)
delamer
Posts: 17348
Joined: Tue Feb 08, 2011 5:13 pm

Re: Help with Retirement Projection Assumptions

Post by delamer »

You can't keep any car forever. Eventually it will need to be replaced. Unless you really decide that you don't need one.
One thing that humbles me deeply is to see that human genius has its limits while human stupidity does not. - Alexandre Dumas, fils
AlohaJoe
Posts: 6609
Joined: Mon Nov 26, 2007 1:00 pm
Location: Saigon, Vietnam

Re: Help with Retirement Projection Assumptions

Post by AlohaJoe »

accbh wrote:The problem I have with typical 25x-33x spending estimates is that I believe health insurance/costs will grow at a rate far greater than inflation. Here is the problem in a nutshell: Today my medical premiums are 15% of total expenses. In 20 years, they will be 47% of expenses, using assumption #2, because health costs are projected to dramatically outpace inflation. How can any SWR calculation account for this properly?
Any forecast that assumes healthcare will consume 50% of 400 million American's spending seems like it is probably built on shaky foundations.
Topic Author
accbh
Posts: 40
Joined: Tue Jul 09, 2013 7:32 pm

Re: Help with Retirement Projection Assumptions

Post by accbh »

AlohaJoe wrote:Any forecast that assumes healthcare will consume 50% of 400 million American's spending seems like it is probably built on shaky foundations.
That's probably true. What would be a more realistic forecast in your opinion?
User avatar
Portfolio7
Posts: 1128
Joined: Tue Aug 02, 2016 3:53 am

Re: Help with Retirement Projection Assumptions

Post by Portfolio7 »

FWIW SS is expected to pay out 75% of current promised benefits worst case :annoyed . I am almost 50, you and I are very near the 'worst case' profile, retiring when SS Program Income/Outgo gap is at it's largest. I assume 70% of SS benefits (from SSA.Gov) in my outlooks for a small built-in margin of safety (the amount you are taking 70% of, or whatever percent you choose to assume, will vary depending on what age you are when you start to draw. 8% Returns on your invested assets is the equilibrium point, roughly, so with the returns you are assuming, I would suspect you'd choose to retire at age 70 for SS purposes, as it seems likely your other funds will last long enough to enable that. Maybe there are better assumptions, but that's what I've used so far for my rough outlooks, in case it helps any.

The health care is a huge wildcard. I am trying to save an extra $1M over my 'number' just to be able to cover what may come. My wife was in an auto accident in her college years, and there will be medical issues to deal with as we age, so I can't be chintzy on my preparation.

I have a special needs child, so I am also trying to build a legacy for him, separate from the above. We are working with a lawyer on the Trust right now.

Hope this helps in a small way, it looks like you have a lot of great input.
"An investment in knowledge pays the best interest" - Benjamin Franklin
Topic Author
accbh
Posts: 40
Joined: Tue Jul 09, 2013 7:32 pm

Re: Help with Retirement Projection Assumptions

Post by accbh »

Portfolio7 wrote:FWIW SS is expected to pay out 75% of current promised benefits worst case :annoyed . I am almost 50, you and I are very near the 'worst case' profile, retiring when SS Program Income/Outgo gap is at it's largest. I assume 70% of SS benefits (from SSA.Gov) in my outlooks for a small built-in margin of safety (the amount you are taking 70% of, or whatever percent you choose to assume, will vary depending on what age you are when you start to draw. 8% Returns on your invested assets is the equilibrium point, roughly, so with the returns you are assuming, I would suspect you'd choose to retire at age 70 for SS purposes, as it seems likely your other funds will last long enough to enable that. Maybe there are better assumptions, but that's what I've used so far for my rough outlooks, in case it helps any.
What are your thoughts on means testing though?
Sandman62
Posts: 103
Joined: Wed Feb 06, 2013 2:22 pm
Location: Rhode Island

Re: Help with Retirement Projection Assumptions

Post by Sandman62 »

accbh wrote:
Portfolio7 wrote:FWIW SS is expected to pay out 75% of current promised benefits worst case :annoyed . I am almost 50, you and I are very near the 'worst case' profile, retiring when SS Program Income/Outgo gap is at it's largest. I assume 70% of SS benefits (from SSA.Gov) in my outlooks for a small built-in margin of safety (the amount you are taking 70% of, or whatever percent you choose to assume, will vary depending on what age you are when you start to draw. 8% Returns on your invested assets is the equilibrium point, roughly, so with the returns you are assuming, I would suspect you'd choose to retire at age 70 for SS purposes, as it seems likely your other funds will last long enough to enable that. Maybe there are better assumptions, but that's what I've used so far for my rough outlooks, in case it helps any.
What are your thoughts on means testing though?
Exactly. At some point, the government may look at those of us sitting on large retirement accounts (especially those who also have pensions!) as not needing SS as much as those who blew every penny they earned on every gadget, clothing item and restaurant or cafe latte that came along. So some of us could take maybe as big as a 50% SS haircut, while others might still get 100%. :x
User avatar
PhysicianOnFIRE
Posts: 470
Joined: Fri Jan 08, 2016 2:46 pm
Location: Up North

Re: Help with Retirement Projection Assumptions

Post by PhysicianOnFIRE »

Based on your replies, I would say you've clearly got a plan and have considered contingency plans, as well.

I think the best approach for any early retiree is to figure out your "target number," overshoot it by a bit unless you can't stand your job, and have a Plan B to earn money again if the need arises. With a solid and adequately funded plan, it's unlikely you'll need to execute Plan B, but it's wise to at least think about one.

:beer
-PoF
sherwink
Posts: 332
Joined: Mon May 28, 2007 9:48 am

Re: Help with Retirement Projection Assumptions

Post by sherwink »

I'd include recurring expenses in your budget: Rent or mortgage, food, clothes and apparel, entertainment, subscriptions, cable and internet, (you've got a bad assumption in that your future taxes will be zero), gifts, charity, travel, etc... You need to do a thorough and complete current annual budget and document your current annual expenses, then tailor and adjust from there.
flyingaway
Posts: 3908
Joined: Fri Jan 17, 2014 9:19 am

Re: Help with Retirement Projection Assumptions

Post by flyingaway »

5) Stocks return 4% (real) per year
6) Bonds return 0.5% (real) per year


For a 50/50 retirement portfolio forward, I would use between 1% and 2% (real) return per year.
Topic Author
accbh
Posts: 40
Joined: Tue Jul 09, 2013 7:32 pm

Re: Help with Retirement Projection Assumptions

Post by accbh »

delamer wrote: I would run additional scenarios with lower assumptions for real returns to look at worsr-case scenarios.
flyingaway wrote:5) Stocks return 4% (real) per year
6) Bonds return 0.5% (real) per year


For a 50/50 retirement portfolio forward, I would use between 1% and 2% (real) return per year.
It's true that I used an average expected stock growth in my calculations (4% real) instead of lower, more conservative estimates. I just re-ran my numbers and they do work (barely) with a real 2% return on stocks if I forego most luxuries. How do most people feel... is this conservative enough, or should I calculate and plan for even lower investment growth?
flyingaway
Posts: 3908
Joined: Fri Jan 17, 2014 9:19 am

Re: Help with Retirement Projection Assumptions

Post by flyingaway »

AlohaJoe wrote:
accbh wrote:The problem I have with typical 25x-33x spending estimates is that I believe health insurance/costs will grow at a rate far greater than inflation. Here is the problem in a nutshell: Today my medical premiums are 15% of total expenses. In 20 years, they will be 47% of expenses, using assumption #2, because health costs are projected to dramatically outpace inflation. How can any SWR calculation account for this properly?
Any forecast that assumes healthcare will consume 50% of 400 million American's spending seems like it is probably built on shaky foundations.
I cannot believe that healthcare cost can go up freely without any challenges in the future. If 50% of people could not afford healthcare, it would be time to bring something other than credit cards to the doctor's office (and demand healthcare).
rgs92
Posts: 3436
Joined: Mon Mar 02, 2009 7:00 pm

Re: Help with Retirement Projection Assumptions

Post by rgs92 »

I think you should not consider future means testing for SS. It would create a tremendous distortion in economic behavior.
It would be a disincentive to save for retirement because (as the means-testing scenarios proposed have been envisioned),
any income from personal taxable retirement savings would dollar-for-dollar reduce your SS payment, starting at a gross income of $85,000/year.

So why save if much of it could be taken away by a SS reduction?
(The income the savings generate would be effectively be taken away, to be precise.)
This is a non-starter IMHO.

So I believe this would never happen. If anything, the retirement age could go up by 5 years or so, or SS could be taxed at 100%, not 85% like now,
or the cost-of-living adjustment could be reduced somewhat.

So while SS could be a little less generous, it's not going to be reduced enough to make a big difference to the degree that you should not include it in forecasting your finances. This is especially so for anyone over 55. (Even at 45, it should be OK).
Last edited by rgs92 on Mon Jan 16, 2017 1:12 pm, edited 1 time in total.
JimmyJammy
Posts: 322
Joined: Sat Oct 26, 2013 1:08 pm

Re: Help with Retirement Projection Assumptions

Post by JimmyJammy »

I've been using http://www.OnTrajectory.com for these kinds of calculations. You can get pretty detailed with the data you input. There are some problems/confusions I have with the tool and so I've been sending in some questions. They're very quick to respond!
MoonOrb
Posts: 1505
Joined: Thu Jan 24, 2013 5:58 pm

Re: Help with Retirement Projection Assumptions

Post by MoonOrb »

I'm assuming you've got this covered, but are none of your assets in 401k/457/IRA plans that will be taxed upon withdrawal?
Topic Author
accbh
Posts: 40
Joined: Tue Jul 09, 2013 7:32 pm

Re: Help with Retirement Projection Assumptions

Post by accbh »

I changed my spreadsheet to reflect the thoughts and opinions of this thread, and updated the OP. Please chime in... Do you agree or disagree with these changes? Thanks!
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Help with Retirement Projection Assumptions

Post by Rodc »

Strongly suggest you not use ONE set of assumptions.

Suggest you vary the assumptions to learn which changes make a significant difference in the result and which do not.

Then spend some time looking at what happens when things that matter are more pessimistic and more optimistic than assumed in the baseline case.

You will likely find some parameters that are not important in that you could be fairly wrong and still get pretty much the same answer. You may find some parameters where even small changes have large results on your retirement situation. You would then want to focus attention on those - either investing in a way that they are under control, or tracking them to make sure they follow the assumed values or adjusting if they don't.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Topic Author
accbh
Posts: 40
Joined: Tue Jul 09, 2013 7:32 pm

Re: Help with Retirement Projection Assumptions

Post by accbh »

Rodc wrote:Strongly suggest you not use ONE set of assumptions.

Suggest you vary the assumptions to learn which changes make a significant difference in the result and which do not.

Then spend some time looking at what happens when things that matter are more pessimistic and more optimistic than assumed in the baseline case.

You will likely find some parameters that are not important in that you could be fairly wrong and still get pretty much the same answer. You may find some parameters where even small changes have large results on your retirement situation. You would then want to focus attention on those - either investing in a way that they are under control, or tracking them to make sure they follow the assumed values or adjusting if they don't.
Yes, that's exactly what I'm doing. I'm using the set of assumptions in the OP as the median, and adjusting parameters around that.
User avatar
David Jay
Posts: 14569
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Help with Retirement Projection Assumptions

Post by David Jay »

Assumptions 1 and 2 will not survive the month...
It's not an engineering problem - Hersh Shefrin | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
Portfolio7
Posts: 1128
Joined: Tue Aug 02, 2016 3:53 am

Re: Help with Retirement Projection Assumptions

Post by Portfolio7 »

accbh wrote:
Portfolio7 wrote:FWIW SS is expected to pay out 75% of current promised benefits worst case :annoyed . I am almost 50, you and I are very near the 'worst case' profile, retiring when SS Program Income/Outgo gap is at it's largest. I assume 70% of SS benefits (from SSA.Gov) in my outlooks for a small built-in margin of safety (the amount you are taking 70% of, or whatever percent you choose to assume, will vary depending on what age you are when you start to draw. 8% Returns on your invested assets is the equilibrium point, roughly, so with the returns you are assuming, I would suspect you'd choose to retire at age 70 for SS purposes, as it seems likely your other funds will last long enough to enable that. Maybe there are better assumptions, but that's what I've used so far for my rough outlooks, in case it helps any.
What are your thoughts on means testing though?
Sorry I missed your response. Two things:
Right, now, I'm assuming my taxes go from 25% Marginal to 28% Marginal upon retirement. I'm working on a process to move all my 401K/IRA to Roth in my late 50's and early 60's, so by 70 when I take SS, I don't have to worry about the tax on a portion of my SS income, which should bring my marginal back down to 25%, or hopefully lower. I think tax impacts are roughly similar before and after retirement without the Social Security wealth tax is factored in. Those who have been there/done that can add their wisdom and/or corrections.

Further Means Testing beyond what is already built into the tax code... I have no answer. Work a few more years, I guess. My investment results have been favorable, so maybe I'll get lucky that way. I'm still unclear on post-retirement expenses, and my project this year is to build a flexible estimate, with segments for necessities, general lifestyle, and Travel.
"An investment in knowledge pays the best interest" - Benjamin Franklin
Topic Author
accbh
Posts: 40
Joined: Tue Jul 09, 2013 7:32 pm

Re: Help with Retirement Projection Assumptions

Post by accbh »

David Jay wrote:Assumptions 1 and 2 will not survive the month...
I should have labeled 1&2 "Health care premiums." I am currently basing that on un-subsidized ACA as a best-guess estimate.
JimmyJammy
Posts: 322
Joined: Sat Oct 26, 2013 1:08 pm

Re: Help with Retirement Projection Assumptions

Post by JimmyJammy »

Sorry I missed your response. Two things:
Right, now, I'm assuming my taxes go from 25% Marginal to 28% Marginal upon retirement. I'm working on a process to move all my 401K/IRA to Roth in my late 50's and early 60's, so by 70 when I take SS, I don't have to worry about the tax on a portion of my SS income, which should bring my marginal back down to 25%, or hopefully low
er.

How can you do that? Is that even possible, to move pre-tax 401K/IRA assets into a post-tax Roth vehicle? I'm curious.
Sailor36
Posts: 21
Joined: Mon May 04, 2015 7:15 pm

Re: Help with Retirement Projection Assumptions

Post by Sailor36 »

JimmyJammy wrote:
Sorry I missed your response. Two things:
Right, now, I'm assuming my taxes go from 25% Marginal to 28% Marginal upon retirement. I'm working on a process to move all my 401K/IRA to Roth in my late 50's and early 60's, so by 70 when I take SS, I don't have to worry about the tax on a portion of my SS income, which should bring my marginal back down to 25%, or hopefully low
er.

How can you do that? Is that even possible, to move pre-tax 401K/IRA assets into a post-tax Roth vehicle? I'm curious.
That would be a Roth Conversion. You pay full taxes on the amount converted in the year you do the conversion. With a 401k, you can typically only do it with accounts from former employers. Often the years between when you stop working and when you start taking Social Security and RMDs from tIRAs and 401Ks are a good time to do these conversions, because you have very little other fully taxable income at that time.
User avatar
Lieutenant.Columbo
Posts: 1189
Joined: Sat Sep 05, 2015 9:20 pm
Location: Cicely AK

Re: Help with Retirement Projection Assumptions

Post by Lieutenant.Columbo »

accbh wrote:My taxable funds have very little unrealized gains and assuming the current tax structure, anything that I do pull out should be minimally taxed
accbh,
1. how can this be? how can you be ready to retire (I'm jealous) but only have little taxable gains? are you taking into consideration that, even if little gains now, over the years, you might accrue taxable gains?
2. would you mind sharing your FI number and what age you have in mind for retiring?
thanks
Rodc wrote:...vary the assumptions to learn which changes make a significant difference in the result and which do not.

Then spend some time looking at what happens when things that matter are more pessimistic and more optimistic than assumed in the baseline case...
Rodc,
The question I have is how does one know by how much to change any of the variables in a way that the change is meaningfully different from the original assumption, but not so unreasonably/unrealistically different that there's no way that scenario will play out.
Thanks.
Lt. Columbo: Well, what do you know. Here I am talking with some of the smartest people in the world, and I didn't even notice!
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Help with Retirement Projection Assumptions

Post by Rodc »

Lieutenant.Columbo wrote:
Rodc wrote:...vary the assumptions to learn which changes make a significant difference in the result and which do not.

Then spend some time looking at what happens when things that matter are more pessimistic and more optimistic than assumed in the baseline case...
Rodc,
The question I have is how does one know by how much to change any of the variables in a way that the change is meaningfully different from the original assumption, but not so unreasonably/unrealistically different that there's no way that scenario will play out.
Thanks.
That is an excellent question.

I would suggest for the sensitivity analysis you choose values that are modestly worse and modestly better than baseline. For stocks maybe +/- 2%. For things like car repairs that would depend on your situation. But as far as "badness" you would want to be consistent - you would not use Zombie Apocalypse bad for one variable and slightly bad for another since the goal is to see which variables you are sensitive to. Hard to do actual statistics on all these categories but think in terms of +/- one standard deviation (so for stocks while you might get +/- 15% year to year, over 20-40 years you might only get 2% or so), other variables use your best guess).

That said, you could certainly go more "worst case", just be consistent.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Quark
Posts: 1045
Joined: Sun Nov 01, 2015 4:32 pm

Re: Help with Retirement Projection Assumptions

Post by Quark »

Sandman62 wrote:
accbh wrote:
Portfolio7 wrote:FWIW SS is expected to pay out 75% of current promised benefits worst case :annoyed . I am almost 50, you and I are very near the 'worst case' profile, retiring when SS Program Income/Outgo gap is at it's largest. I assume 70% of SS benefits (from SSA.Gov) in my outlooks for a small built-in margin of safety (the amount you are taking 70% of, or whatever percent you choose to assume, will vary depending on what age you are when you start to draw. 8% Returns on your invested assets is the equilibrium point, roughly, so with the returns you are assuming, I would suspect you'd choose to retire at age 70 for SS purposes, as it seems likely your other funds will last long enough to enable that. Maybe there are better assumptions, but that's what I've used so far for my rough outlooks, in case it helps any.
What are your thoughts on means testing though?
Exactly. At some point, the government may look at those of us sitting on large retirement accounts (especially those who also have pensions!) as not needing SS as much as those who blew every penny they earned on every gadget, clothing item and restaurant or cafe latte that came along. So some of us could take maybe as big as a 50% SS haircut, while others might still get 100%. :x
There aren't enough high net worth or high income people to make a substantial difference for SS. The only way for means testing to make a substantial difference is to bring the trigger levels down well into the middle class. There already is a limited type of means testing, in that SS benefits start to be taxable above a certain income level. None of this will necessarily prevent means testing.
User avatar
Portfolio7
Posts: 1128
Joined: Tue Aug 02, 2016 3:53 am

Re: Help with Retirement Projection Assumptions

Post by Portfolio7 »

Quark wrote:
Sandman62 wrote:
accbh wrote:
Portfolio7 wrote:FWIW SS is expected to pay out 75% of current promised benefits worst case :annoyed . I am almost 50, you and I are very near the 'worst case' profile, retiring when SS Program Income/Outgo gap is at it's largest. I assume 70% of SS benefits (from SSA.Gov) in my outlooks for a small built-in margin of safety (the amount you are taking 70% of, or whatever percent you choose to assume, will vary depending on what age you are when you start to draw. 8% Returns on your invested assets is the equilibrium point, roughly, so with the returns you are assuming, I would suspect you'd choose to retire at age 70 for SS purposes, as it seems likely your other funds will last long enough to enable that. Maybe there are better assumptions, but that's what I've used so far for my rough outlooks, in case it helps any.
What are your thoughts on means testing though?
Exactly. At some point, the government may look at those of us sitting on large retirement accounts (especially those who also have pensions!) as not needing SS as much as those who blew every penny they earned on every gadget, clothing item and restaurant or cafe latte that came along. So some of us could take maybe as big as a 50% SS haircut, while others might still get 100%. :x
There aren't enough high net worth or high income people to make a substantial difference for SS. The only way for means testing to make a substantial difference is to bring the trigger levels down well into the middle class. There already is a limited type of means testing, in that SS benefits start to be taxable above a certain income level. None of this will necessarily prevent means testing.
Agreed, though at present you can get around the SS Tax if you can manage to move most of your retirement accounts into Roth vehicles before you start to take SS payments. The latest proposals from the GOP include some sort of means testing requirement, but there seem to be few details on how they envision that - it may make the Roth strategy more important, or may eliminate that work-around entirely, but I don't want to go too far into speculation. I only mention this because it seems prudent when planning for the future to take these possibilities into account. I've already lost 95% of my future pension :( ; I'm not sure I can also handle a big SS change :x without a decline in lifestyle. It's a good lesson to those starting out. There are many things you take for granted that you may not be able to count on as you get older. These are risks that can only be handled via increasing your savings rate. I'm just glad the 401(k)s are in our names.
"An investment in knowledge pays the best interest" - Benjamin Franklin
User avatar
Lieutenant.Columbo
Posts: 1189
Joined: Sat Sep 05, 2015 9:20 pm
Location: Cicely AK

Re: Help with Retirement Projection Assumptions

Post by Lieutenant.Columbo »

Rodc wrote:
Lieutenant.Columbo wrote:Rodc,
The question I have is how does one know by how much to change any of the variables in a way that the change is meaningfully different from the original assumption, but not so unreasonably/unrealistically different that there's no way that scenario will play out.
Thanks.
That is an excellent question.

I would suggest for the sensitivity analysis you choose values that are modestly worse and modestly better than baseline. For stocks maybe +/- 2%. For things like car repairs that would depend on your situation. But as far as "badness" you would want to be consistent - you would not use Zombie Apocalypse bad for one variable and slightly bad for another since the goal is to see which variables you are sensitive to. Hard to do actual statistics on all these categories but think in terms of +/- one standard deviation (so for stocks while you might get +/- 15% year to year, over 20-40 years you might only get 2% or so), other variables use your best guess).

That said, you could certainly go more "worst case", just be consistent.
Thank you, Rodc.
I ask so many questions that only by chance sooner or later one of them would be found excellent by someone :wink:.
If one wanted to simplify the "calculation of The Number if retiring at age 55 years", what "yearly expenses multiplier" would you suggest using for a 90% chance of dying before funds run out?
Thanks.
Lt. Columbo: Well, what do you know. Here I am talking with some of the smartest people in the world, and I didn't even notice!
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Help with Retirement Projection Assumptions

Post by Rodc »

Lieutenant.Columbo wrote:
Rodc wrote:
Lieutenant.Columbo wrote:Rodc,
The question I have is how does one know by how much to change any of the variables in a way that the change is meaningfully different from the original assumption, but not so unreasonably/unrealistically different that there's no way that scenario will play out.
Thanks.
That is an excellent question.

I would suggest for the sensitivity analysis you choose values that are modestly worse and modestly better than baseline. For stocks maybe +/- 2%. For things like car repairs that would depend on your situation. But as far as "badness" you would want to be consistent - you would not use Zombie Apocalypse bad for one variable and slightly bad for another since the goal is to see which variables you are sensitive to. Hard to do actual statistics on all these categories but think in terms of +/- one standard deviation (so for stocks while you might get +/- 15% year to year, over 20-40 years you might only get 2% or so), other variables use your best guess).

That said, you could certainly go more "worst case", just be consistent.
Thank you, Rodc.
I ask so many questions that only by chance sooner or later one of them would be found excellent by someone :wink:.
If one wanted to simplify the "calculation of The Number if retiring at age 55 years", what "yearly expenses multiplier" would you suggest using for a 90% chance of dying before funds run out?
Thanks.
Quickly reading the opening post I did not see a "yearly expenses multiplier". Is this an inflation factor?
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
User avatar
Lieutenant.Columbo
Posts: 1189
Joined: Sat Sep 05, 2015 9:20 pm
Location: Cicely AK

Re: Help with Retirement Projection Assumptions

Post by Lieutenant.Columbo »

Rodc wrote:
Lieutenant.Columbo wrote:
Rodc wrote:
Lieutenant.Columbo wrote:Rodc,
The question I have is how does one know by how much to change any of the variables in a way that the change is meaningfully different from the original assumption, but not so unreasonably/unrealistically different that there's no way that scenario will play out.
Thanks.
That is an excellent question.

I would suggest for the sensitivity analysis you choose values that are modestly worse and modestly better than baseline. For stocks maybe +/- 2%. For things like car repairs that would depend on your situation. But as far as "badness" you would want to be consistent - you would not use Zombie Apocalypse bad for one variable and slightly bad for another since the goal is to see which variables you are sensitive to. Hard to do actual statistics on all these categories but think in terms of +/- one standard deviation (so for stocks while you might get +/- 15% year to year, over 20-40 years you might only get 2% or so), other variables use your best guess).

That said, you could certainly go more "worst case", just be consistent.
Thank you, Rodc.
I ask so many questions that only by chance sooner or later one of them would be found excellent by someone :wink:.
If one wanted to simplify the "calculation of The Number if retiring at age 55 years", what "yearly expenses multiplier" would you suggest using for a 90% chance of dying before funds run out?
Thanks.
Quickly reading the opening post I did not see a "yearly expenses multiplier". Is this an inflation factor?
I meant something like x33 or x50 one's yearly expenses
Lt. Columbo: Well, what do you know. Here I am talking with some of the smartest people in the world, and I didn't even notice!
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Help with Retirement Projection Assumptions

Post by Rodc »

Lieutenant.Columbo wrote:
Rodc wrote:
Lieutenant.Columbo wrote:
Rodc wrote:
Lieutenant.Columbo wrote:Rodc,
The question I have is how does one know by how much to change any of the variables in a way that the change is meaningfully different from the original assumption, but not so unreasonably/unrealistically different that there's no way that scenario will play out.
Thanks.
That is an excellent question.

I would suggest for the sensitivity analysis you choose values that are modestly worse and modestly better than baseline. For stocks maybe +/- 2%. For things like car repairs that would depend on your situation. But as far as "badness" you would want to be consistent - you would not use Zombie Apocalypse bad for one variable and slightly bad for another since the goal is to see which variables you are sensitive to. Hard to do actual statistics on all these categories but think in terms of +/- one standard deviation (so for stocks while you might get +/- 15% year to year, over 20-40 years you might only get 2% or so), other variables use your best guess).

That said, you could certainly go more "worst case", just be consistent.
Thank you, Rodc.
I ask so many questions that only by chance sooner or later one of them would be found excellent by someone :wink:.
If one wanted to simplify the "calculation of The Number if retiring at age 55 years", what "yearly expenses multiplier" would you suggest using for a 90% chance of dying before funds run out?
Thanks.
Quickly reading the opening post I did not see a "yearly expenses multiplier". Is this an inflation factor?
I meant something like x33 or x50 one's yearly expenses
Well, a net present value using a discount/growth rate of 1.31% means at 30 years gives a 4% rule, or multiplier of 25. If we extend to 40 years we get 3.19% or a multiplier of 31.38557211. (extra digits just for fun, so call it 30).

I don't have any of the historical SWR studies handy, but by memory they show something similar. The key is being flexible and cutting expenses when markets are down. If you can do that you would likely be able to make the 4% WD work most years.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Topic Author
accbh
Posts: 40
Joined: Tue Jul 09, 2013 7:32 pm

Re: Help with Retirement Projection Assumptions

Post by accbh »

Lieutenant.Columbo wrote:how can this be? how can you be ready to retire (I'm jealous) but only have little taxable gains? are you taking into consideration that, even if little gains now, over the years, you might accrue taxable gains?
My gains are low because I changed my portfolio fairly recently, and I realized gains in the process. As for new gains, I am assuming current tax law, which allows for up to $75,000 worth of gains (as income) at the 0% LTCG rate. If this changes, it will be covered by social security (see assumption #10.)
Post Reply