FI Calc - with changes in spending, known and unknown

[retired?]

Let's suppose one is fine with using the 4% rule.  Aside from putting the safety nets in place that MMM discusses, how would you alter your calc if spending will change?

I have two changes I'd like to incorporate:  1) kids fly the coop in about 10 years and 2) possible high medical costs late in life.


1)  Not precise, but I can make some reasonable estimates.  Right now, spending is about 60k.  Once kids are gone they will go down a lot.  Main reason is that we'll downsize and move to area with lower prop taxes.  For the sake of the example, I'll guess 36k.

2)  Somewhat unknown, but, for example, may FiL had Parkinson's and the cost for care was about 90k per year for last three years of his life (once it got to where my MiL could not take care of him).

I'm 45, say I live to 80-85.  One simple method would be to assume (aside from inflation), 60k for ten years, 36k for 25 years, and near 100k for three years.  At worst (best, really) my wife nor I have a major illness and we leave a little extra to the kids.

Anyone incorporate this sort of known/unknown spending changes in their FI/SWR calc?  How did you approach it?  Other than simply choosing 3%, for example.

[eyePod]

Let me consult my tea leaves.

OK, they were green in a bag. That's all I got.

You're asking a lot of what ifs and maybes and then trying to multiply them all together to make a number. None of us know what will happen in 1 year let alone 45. Just try to minimize expenses now and save up as much as possible.

[EricP]

To me worrying about skyrocketing health care costs is unnecessary.  Unless there is a significant difference in age between your spouse and you then you can just plan on spending down your principal when healthcare costs rise in your late years.  There are also a lot of variables, 45 years from now healthcare in your country could be significantly different.  Technological advances, transition to NHS, etc.  Don't worry about that for now, because you always have the option to spend down principal.

As for the "transition years" between having kids and not having kids you just need to calculate your two end points and then figure out how much additional principal you need to get from the high years to the low years.  For the low years you'll need 900k and for the high years you would need 1.5M if you were planning to sustain that, but you aren't.

You yearly expense difference is 24k so if you multiply that by ten and add that amount to the 900k I would say if you get 1.14M then that would be enough for FI.  You'll spent down during your child years until they leave and then you'll have your 900k once they have flown the coop.  I ignored gains on the spentdown principal during the 10 years because that should give an extra level of safety and you may need that if a kid decides to stick around an extra year or something.

[retired?]

Let me consult my tea leaves.

OK, they were green in a bag. That's all I got.

You're asking a lot of what ifs and maybes and then trying to multiply them all together to make a number. None of us know what will happen in 1 year let alone 45. Just try to minimize expenses now and save up as much as possible.

I understand there is a huge amount of uncertainty in future cash flows and that SWR's are a safe rule of thumb.  But, it would be foolish to not include known changes, even if they are not known with complete precision.  Not covering kids or not having high housing costs are fairly known for me.

That said, I am simply saying, if you thought your spending was a step function.......medium for a certain time, lower for an extended time, then higher for a short time, how would you change your FI calc.

Still lots of assumptions, but one less than in the usual calc (which only assumes current spending habits). 

Minimizing expenses and saving as much as possible....good advice, but that alone does not tell one when they are FI.

Thanks for your (non)input, but I expect others have incorporated this simple adjustment, and I'd like to hear about the various approaches.

[nereo]

eyePod is right - you are asking a lot fo the tea-leaves.
BUT, you can "force" fireCalc to make some interesting adjustments if you explore the other tabs.  For instance, you can enter in a postive OR negative income for pension income.  In your case, if your spending goes from $60k to $30k you could tell fireCalc that you will be getting a 'pension' of $30k/year once the kids leave the nest (resulting in you only 'needing' $30k per year from your portfolio).

You can also add extra, one-time expenses under the "Portfolio Changes" tab.  There you could add 3 years of $30k additional expenses for projected, end-of-life medical care.  The catch is you have to figure out in what year those are likely to happen.  But you can just keep changing the years to fit different scenerios (e.g. "let's suppose I need $30k/year medical for three years starting when I am 75, which is 20XX.  Now let's supposed I need that earlier, at age 70 - repeat simulation"

Just play around with the different inputs.    However, one ultimate take-home is that the spread (or 'uncertainty/variance') increases the further you get from the present.  30 years from now using the 4% WR you might have 4x your original portfolio, or in some cases you might have $0.  Ultimately, flexibility and/or additional assets are key.

[retired?]

To me worrying about skyrocketing health care costs is unnecessary.  Unless there is a significant difference in age between your spouse and you then you can just plan on spending down your principal when healthcare costs rise in your late years.  There are also a lot of variables, 45 years from now healthcare in your country could be significantly different.  Technological advances, transition to NHS, etc.  Don't worry about that for now, because you always have the option to spend down principal.

As for the "transition years" between having kids and not having kids you just need to calculate your two end points and then figure out how much additional principal you need to get from the high years to the low years.  For the low years you'll need 900k and for the high years you would need 1.5M if you were planning to sustain that, but you aren't.

You yearly expense difference is 24k so if you multiply that by ten and add that amount to the 900k I would say if you get 1.14M then that would be enough for FI.  You'll spent down during your child years until they leave and then you'll have your 900k once they have flown the coop.  I ignored gains on the spentdown principal during the 10 years because that should give an extra level of safety and you may need that if a kid decides to stick around an extra year or something.

Thanks for the suggestion.  I like the approach of funding each period separately.  It is additive.  Yeah, my concern isn't about the general level of healthcare costs, b/c as you noted there is no way to predict it.  I am thinking of a high cost event for me or my wife and using recent costs.

[retired?]

I'll maintain that I am not asking any more of the tea leaves than the ordinary calculation. 

Instead of a stream of 36k with uncertainty around it, or a stream of 60k with uncertainty around it, I am assuming a stream of:

60, 60, 60,........60, 36, 36, 36..........36, 90, 90, 90.

with uncertainty around it.  Think of it as moving the mean.

In response, I'll ask - do people actually not include kids moving out and the associated drop in spending?

For the sake of future responses - let's assume the changes are known with the same level of certainty that one would place on a fixed sequence of spending.  No tea leaves, it is an assumption.

[eyePod]

I think it's important. I'd start back calculating how much you need per chunk of life (pre-pension, post pension, pre kid, post kid, pre 401k, post 401k, etc.) and then figure out your current assets and how long that will get you. Once everything meets up, then you're done.

And then life will happen and everything will change and the calcs will have to be re-done.

[EricP]

I don't have kids so I can only say what I imagine I would do, and I'm also still fairly early in my FI journey.  I'm only 25 and only decided that this is what I want 6 months ago.  I still haven't gotten into overdrive with Savings Rate (just a little over 30%) because circumstances prevent me from getting into bike range of my job without losing a substantial portion of my income.

So, enough with the background, I'm guessing most people fall into one of two camps:  Kids expenses are not that substantial (<5k extra) or they are not going to hit FI until after the kids leave.  MMM had that article where he advocated only having 1 kid and that would obviously help push people into the first group.

As for me, I expect to not have very high expenses when having kids.  I'm getting married in April and hopefully kids are still 5 years out, but when they do come I would like to only have 2 and then not get a McMansion to house them.  Definitely something larger than her current house, but nothing that I would have to downsize out of after they leave.

[retired?]

I think it's important. I'd start back calculating how much you need per chunk of life (pre-pension, post pension, pre kid, post kid, pre 401k, post 401k, etc.) and then figure out your current assets and how long that will get you. Once everything meets up, then you're done.

And then life will happen and everything will change and the calcs will have to be re-done.

Very true.  Dynamic Control.  That's life, and we use the information we have available.  All predictions/models are wrong, but some are better than others.

Of course, you have to redo the calcs whether you use a simple model or one that is more accurate.

Yes, the chunk of life approach seems to make good sense and straightforward.  Thanks.

[nereo]

I'll maintain that I am not asking any more of the tea leaves than the ordinary calculation. 

what I meant by the "asking a lot of the tea leaves" is simply that anytime you add in another variable, in inherently increases the uncertainty.  If you are certain how much and when you want to change your spending then it's a simple calcualtion.  But if you think that you 'might' spend 'about' $30k/year extra towards the end of life (which could be anywhere between 2050 and 2065), then each calculation has is a floating calculation and hence... more uncertainty.

But as an example of what I was talking about, let's say you are planning on retiring in 2016, and will need $60k until the kids leave the nest in 2025, and then you will need only $30k until later in life (say 2050, when you will want to have a minimum $50k for added health costs.  You can 'force' fireCalc to simulate that by setting your initial spending to $60k, then entering a  pension of $30,000 in 2025 (requiring only $30k/year from your portfolio), and then adding a negative pension of $20k in 2050 (to allow for an increase in spending).
Just for fun I ran that scenario a few times.  With a portfolio of $1.5M you get no failures.  With $1M you get 72.8% success. 

[Goldielocks]

I am posting as I am exploring the same question myself. And want to see the responses from this forum.

Where I am landing is to consider this as three pots of money for me:

1. Secure long term retirement so I don't have to think about income or work after 60. FireSim is good for that, and I add in lump sums or downsizing if I need a lot of money for unexpected reasons. Such as your LTC. This is no touch money until 60 or older.

2. Second investment pool to cover and spend down basic expenses to allow FIRE. Almost a second retirement stream calc.   Here I assume SO salary will cover the majority. Extra money  at the end, if any,is for spendy pants buys at age 60. Maybe add enough to cover base living if he unable to work after all.

3. My immediate needs for 5-10 more years with kids. Cash savings draw down to offset SO modest income near term.   No investment gains assumed for this short term savings.

This puts me at Fire in 3 yrs with lots of annual spend, FIRE now with base spend or part time input or take on a boarder. Not mentally there yet to FIRE.

I am undecided about paying down mortgage faster, but reducing monthly expenses seems like a good idea to reduce risk of ARM increases especially if I would not choose to start working to cover the difference.   This impacts the above a lot in our high COL area.

That said I will likely change my mind in the next few months as I read others' thoughts and discuss with SO.

[arebelspy]

Go to www.cfiresim.com

It allows you to put in year by year spending.

Simulate your own situation and see what the success rate is.

:)

[MsSindy]

I prefer to go old-school and use Excel.  There's one column for each year.  I list my known income streams (interest earned on mortgage notes I hold, stocks/MF) across my accounts and then deduct what I plan to spend - repeat for the number of years we plan to live (90).  This allows me to play around with different ROIs (amazing what just 1% does over 30 years).  This way I can adjust my spending for different phases of my life - ie. when I'm past 70, I'll move into a small home, probably travel less, but healthcare may be more - so I adjust accordingly.  I have a fair bit of fat in my spending (hobby & travel), so I know I can always pull in the reigns there if needed.  I also don't include pulling any money from my home (reverse mortgage or sale), so I have contingency there...we can also downsize earlier in life.  I think flexibility is key.

What this doesn't take into consideration is the ebbs/flows of the market.  So I use a small ROI % and I only have a small percentage of my assets in the stock market anyway - most of it is holding mortgage notes where I have cash flow.

At the end of it all, it really depends how much risk you're willing to take on as to how close you want to cut the numbers....my DH is way more conservative than I am, so I feel we're a bit on overkill, but ER is still close enough that it works for us.

[deborah]

There is a lot of literature that during the last eight years of life people end to be in poor health. We also have a number of calculators of how your personal life expectancy, so both these factors can be simulated without just looking into tea leaves.