Variance in FIRE calculators

[dodojojo]

I check the calculators a few times each year just to see where I am.  I don't get too wrapped up with them because I thought I was about 10-14 years away.  This is based on using Cfiresim most of the time.

Yesterday, I checked Cfiresim and it gave me the same answer I have been seeing the last few years.  If I want to fire on 70K annually (my Lean FIRE is 40K), I will have wait 13 years.  I'm 46, so as usual, it's kind of depressing to think of 59, especially with so many examples of much younger FIREs.  Usually, I keep a stiff upper lip and carry on.  But today, I tried other calculators and used 70K for all of them.

Firecalc, 95% success rate FIRE at 56.  Hmmm...3 years faster than Cfiresim.

Nerdwallet, retire at 55, 6% investment annual return rate

Vanguard, retire at 55, selecting 6% investment annual return rate (which is pretty conservative, right?)

Now, I'm wondering if I have been using Cfiresim incorrectly?  For someone doing an older FIRE, 55 versus 59 is a huge difference!

EDIT: Clearly I struggle with math and calculators--had to redo the numbers and the calculators are now giving 55-56 in age.  They seem more in line.  Thought Vanguard was odd at 52...I was so excited!

[dodojojo]

I checked Cfiresim and there is one section I have always doubted if I understood it correctly.  The "Other Income" section where I have always input the amount I would invest for the years from now until FIRE age rather than my annual income.  I changed it to my actual annual income and my FIRE age dropped from 59 to 55.  So, did I misunderstand this section and that accounts for Cfiresim being the outlier?

Someone say yes because this just about makes my week!

[nereo]

It's important to understand that cFiresim and FireCalc are drawing upon historical data while things like Nerdwallet and Vanguard are simply blindly assuming a return of x% every year.  Averages are easy to program/calculate, but are less informative.  When cFireSim says that you have a 95% success rate what it's doing is saying that in 95% of historical scenarios the portfolio did not go to $0 with the timeframe specified (e.g. a 30 year retirement).


Regarding your questions about "additional income" in cFireSim - the calculator is assuming that this is additional income, which it uses to reduce the amount you need to live off of.  For example, if you plan on having a part-time job as a swim coach which will bring in $5k/year in retirement, and your spending plan is for $40k/year, cFireSim will calculate that you need only $35,000/year because of that additioanl income.  It's a particularly useful column if your pays you $X/year as a buy out or if you know you will continue working part-time with predictable income.


Also:  Don't sell yourself short for retiring before age 60; even if you retire at age 59 you will be ahead of roughly 90% of Americans. It may not be as early as some on this forum but you sound like you are already in a very good financial situation.  It sounds like you could easily be retired by 55 with some good fortune and a little hard work.

[dodojojo]

Regarding your questions about "additional income" in cFireSim - the calculator is assuming that this is additional income, which it uses to reduce the amount you need to live off of.  For example, if you plan on having a part-time job as a swim coach which will bring in $5k/year in retirement, and your spending plan is for $40k/year, cFireSim will calculate that you need only $35,000/year because of that additioanl income.  It's a particularly useful column if your pays you $X/year as a buy out or if you know you will continue working part-time with predictable income.

Hmmm...so how do I account for my investment additions from now until retirement in Cfiresim?

In Firecalc there is this:

What year will you retire?: (Or how many years before you retire?) We'll assume NO withdrawals until then.
   
How much will you add to your portfolio until then, per year? (Enter values in 2019 dollars; FIRECalc will assume future savings will keep up with inflation.)

For Cfiresim, I was using the Other Income section to mirror the above.  Doing that my FIRE age comes out to 59.  That's 3 years longer than FIRECalc.  It's only when I entered my gross income, did Cfiresim drop to 56 and closer inline with FIRECalc's estimate. 

Also:  Don't sell yourself short for retiring before age 60; even if you retire at age 59 you will be ahead of roughly 90% of Americans. It may not be as early as some on this forum but you sound like you are already in a very good financial situation.  It sounds like you could easily be retired by 55 with some good fortune and a little hard work.

Trust me that's what I tell myself, just to keep sane on the days when I'm ready to jack it in. It's especially hard when I go on a FIRE blog reading binge and it's all 30/40 YOs traveling the world, etc.

But, you're right of course.  Gotta keep things in perspective. 

[Rdy2Fire]

FIRECALC gives me the numbers I want to see however I realize it's based on historical data and that's somewhat concerning so I look at that but also play but the 3-4% rule. So using 3.5% you'd need 2M to get your 70K per year. If you only cared about 40K then obviously you could have significantly less with about 1.2m. I had a couple of cocktails so maybe my math is off :)

[nereo]

Hmmm...so how do I account for my investment additions from now until retirement in Cfiresim?

Sadly, you can't.. and I think it's kindof a chicken-and-egg problem.  cFireSim runs off your "FI number" and allows for adjustments in retirement to determine whether or not your portfolio will survive however many years you plan on needing it for.  What it doesn't do is predict how much savings you will need to GET to particular FI number. 

Frankly I find the approached used by cFireSim to be the more useful method, as I wouldn't pull the trigger on retirement until I hit my pre-determined FI number, regardless of whether it was longer than anticipated to get there.

You can use a variety of other calculators to estimate how long it will take you to reach $X, but there's a ton of variance if its in the market.  Once your portfolio is much larger than your annual contributions its the market that makes most of the gains (or losses) in a given year.

[habanero]

I've not done the math on it, but the power of additional income can be viewed as a lower withdrawal rate which massively increases the probability of success - way beyond the seemingly meagre sum earned every year. It effectively means that during bad years you would need to sell a lower part of your nest egg and consequently be better positioned for the eventual rebound and you have a larger 'stash on which the magic off compounded interest works. If you take the obvious case with "additional income" equal to your spending you would never need to sell anything and the probability of success will be 100% and you don't even need a nest egg to live off.

[dodojojo]

Thank you all for your insights.  I'm going to use both FireCalc and Cfiresim going forward rather than just Cfiresim.

[jeroly]

Consider looking at the VPW calculator on bogleheads as well.

[bacchi]

Consider looking at the VPW calculator on bogleheads as well.

Yeah, it'll make your time-to-FIRE (TTF) dramatically shorter and is "guaranteed" not to fail. The kicker is that the withdrawal amount is very volatile. With your ideal FIRE at $70k and your lean FIRE at $40k, it's probably perfect for your situation.

cfiresim has a VPW mode with a floor and ceiling.

[ysette9]

You absolutely can add additional savings in cFIREsim. It is in the additional income section and you have to remember to give it an end date the year before you FIRE.

I suspect your situation is better than what cFIREsim is telling you because the interface takes some time to get used to. It took me a while of using it to get comfortable with inputting my info correctly.

[dodojojo]

You absolutely can add additional savings in cFIREsim. It is in the additional income section and you have to remember to give it an end date the year before you FIRE.

That's what I have been doing.

I guess cFiresim really does estimate my FIRE age is 59 and Firecalc estimates 56. I thought they were using the same/very similar models and expected the estimates to be closer.

I downloaded the VPW worksheet.  Haven't had time to read up on it yet. It's a little bewildering at first glance.

[ysette9]

An internal check I use is on the results page, I check the average stash at retirement number. If this number seems in the ballpark of what I expect my stash to be after saving $x/yr for y more years, then I got that input syntax correct.

[dodojojo]

I was making a mistake in Firecalc.  For "years" in "Start Here",  I was inputting the number of years after retirement.  But just read this:

Because you indicated a future retirement date (2029), the withdrawals won't start until that year. Your contributions will continue until then. The tested period is 10 years of preretirement plus 40 years of retirement, or 50 years.

So after adjusting, Firecalc is putting me at 57 years old rather 56.  Putting me a little closer to the cFiresim estimate.

What blew my mind though is if I adjusted my additional savings for my remaining work years, it really made very little difference.  So instead of saving 38K annually, saving 25K pushed my retirement age out by one entire year.  I guess the power of my number now and years remaining makes more of difference than adding in a tiny percentage of my overall nest egg for the next 10 years or so.

It's mind blowing because I have to find a new job but have held back because I was unwilling to take a pay cut.  I don't want to get into it but it's not likely I can find comparable pay to what I currently earn in my next job for the same type of work.  But knowing that I can basically max my 401K and IRA each year and only have my retirement age increase by one year...wow, maybe I can drop this mental block of "I have to stay at this job if I want to retire before 60." 

Maxing out 401K and IRA is realistic because that's what I was doing when I was earning 65-75K.

Also calculated if I made no additional savings ever to my current total--FIRE would happen at 63.  Not ideal but it's still before 65.

[nereo]

What blew my mind though is if I adjusted my additional savings for my remaining work years, it really made very little difference.  So instead of saving 38K annually, saving 25K pushed my retirement age out by one entire year.  I guess the power of my number now and years remaining makes more of difference than adding in a tiny percentage of my overall nest egg for the next 10 years or so.

As you edge closer to retirement and your investment account grows much, much larger than your annual contributions the amount you contribute matters less and less.  This is entirely because market forces (i.e. gains and losses in your investment funds) dwarf any monthly savings. 

[Metalcat]

What blew my mind though is if I adjusted my additional savings for my remaining work years, it really made very little difference.  So instead of saving 38K annually, saving 25K pushed my retirement age out by one entire year.  I guess the power of my number now and years remaining makes more of difference than adding in a tiny percentage of my overall nest egg for the next 10 years or so.

As you edge closer to retirement and your investment account grows much, much larger than your annual contributions the amount you contribute matters less and less.  This is entirely because market forces (i.e. gains and losses in your investment funds) dwarf any monthly savings.

And is a VERY good thing to consider when feeling trapped in a job you don't love. When staring down the barrel of a long timeline to FI, it's worth doing some calculations to see just how soon one can pull the plug on their job if they were to coast for awhile and let compounding do it's thing.

DH plans to stay at his job until he can get his full golden handcuffs pension (14 years), but it really helps him enjoy his job when on difficult days, he reminds himself that he could leave in 2.5 years and find something less stressful or drop to very part time if he wanted to.

[dodojojo]

I'm playing around with the variable spending option in Cfiresim--illuminating.  I'm really glad I started this thread--I was a bit fearful I was asking a dumb question, but I have learnt a lot. Most important, I now see more FIRE options.

[ysette9]

Good! Please keep asking questions because we all learn this way. I have spent hours and hours with cFIREsim and still learned something totally new the other weekend from my husband who had only just started playing with it.

[dodojojo]

Instead of using my time to look for a new job, I'm still doodling away on Cfiresim.  Discovering tiny new nuances, mainly along the lines of:

What blew my mind though is if I adjusted my additional savings for my remaining work years, it really made very little difference.  So instead of saving 38K annually, saving 25K pushed my retirement age out by one entire year.  I guess the power of my number now and years remaining makes more of difference than adding in a tiny percentage of my overall nest egg for the next 10 years or so.

I'm tinkering with CoastFI scenarios.  For example, if I continue my max annual investment of 36K until I'm 54 AND I use variance withdrawal of a low of 35K to a ceiling of 100K with annual spending at 70K.  I have a 95% chance of retiring at 55.

Compared to maxing out my investments at 26K annually to age 51 and then not investing as I work/CoastFI enough to cover expenses only.  I have a 95% chance of retiring fully at...56. 

The wide variance of low and high withdrawal is because 1) my leanfire number is around 40-45K.  And I have a cash reserve of about 1 year.  I do not count this in my withdrawal stache.  Scenario includes growing the cash to about 3 years so in theory I could go a year or two with $0 withdrawal.  But I use 35K as a floor just to be safe.

For context, I'm at a career's crossroads and I'm afraid to take a job at a lower pay.  But again, it looks like the bulk of the stache and time is the key driver at this point and killing myself and being unhappy at a job so I can save an extra 10-20K annually for a handful of more years doesn't really make that much of a difference.  Retiring early in my 30's and 40's--well that boat has sailed so now realistically I'm looking at years in my 50's and at this point, the math seems to point to just a year or two difference.

Now if I can break through they psychological barrier and put myself out there on the job market!

[SwordGuy]

CFireSim is pretty flexible.   I just wish the owner would read his email and fix the sql database issue that's keeping people from saving or loading scenarios (because you can't log in anymore). :(

[ysette9]

CFireSim is pretty flexible.   I just wish the owner would read his email and fix the sql database issue that's keeping people from saving or loading scenarios (because you can't log in anymore). :(

THIS!!

[Classical_Liberal]

I'm tinkering with CoastFI scenarios.  For example, if I continue my max annual investment of 36K until I'm 54 AND I use variance withdrawal of a low of 35K to a ceiling of 100K with annual spending at 70K.  I have a 95% chance of retiring at 55.

The more you tinker, the less useful these calculators are.  Really, they are just a guide.  Can you really predict your exact spending 25 years from now?  Will the past look exactly like the future?  Will you keep exactly to the same allocation even if markets change, new investment opportunities arise? 

They should be used as a useful tool to see if you are in the ballpark (or for fun).  If you are, plan for adaptability and you'll be fine.

[lauren_knows]

@ysette9 - @SwordGuy managed to get in contact with me.  You should be able to log in now, and I'm working on a re-write of the site, too. :D

[ysette9]

@ysette9 - @SwordGuy managed to get in contact with me.  You should be able to log in now, and I'm working on a re-write of the site, too. :D

Fantastic! Thank you so much