FIREcalc Question

[tozz]

Are there any known bugs in the "FIRECalc's Research Suggests..." section of FIREcalc?  I've always gotten screwy withdrawal numbers there, and just ignored them.  But now I am wondering if I'm the one with the problem, and I've been entering incorrect data all along.

 

[Telly]

Not that I know of!  When I run FIRECalc, I'm always using that feature. Playing with it, it has always made sense.

My only common mistake in data entry is with the sign on increasing/decreasing withdrawals in x number of years. It's easy to forget to put the negative sign in when required.

 

[tozz]

OK, I think I've figured it out.  The following sample (not my own, but similar) reflects a pension-heavy scenario with the sale of a home in year 2, move to low cost of living area through year 7, and purchase of a smaller home in year 7.  $45,000 will be required for fifteen years, dropping down to the $25,000 pension only after year 15.  The bold areas confused me--apparently they are not the same thing? 

FIRECalc Results
You have proposed a withdrawal of 45.00% of your starting portfolio.

We looked at the 116 possible 15 year periods from 1871 until 2002, and the 15 partial periods from 1987 until 2002, starting with a portfolio of $100,000 and taking out $45,000 the first year, and the same amount after adjustments for inflation (PPI) each year except as follows:

Starting in year 1, the withdrawal was decreased by $25,000 (adjusted for inflation). Pension

Starting in year 2, the withdrawal was decreased by $25,000 (adjusted for inflation). Move to low cost of living area

Starting in year 7, the withdrawal was increased by $25,000 (adjusted for inflation). Return from low cost of living area

In year 2, the portfolio was increased by a single (inflation-adjusted) $250,000 addition (as from the sale of a house). Home sale

In year 7, the portfolio was decreased by a single (inflation-adjusted) $150,000 reduction (as for a major purchase). Purchase cheaper home

Your Success Rate is 92.4%
In 7.6% of those 131 periods, the portfolio would have been fully depleted at or before the withdrawal in year 15. In 92.4% of the years, the portfolio would have maintained a positive balance through the withdrawal in year 15.
The average (mean) portfolio balance following the withdrawal in year 15 was $455,037.

(This assumes your portfolio consists of 25% in Commercial Paper and 75% in equities that behave like the market as a whole, with an overall expense ratio of 0.18%.)

FIRECalc's Research Suggests...
Checking for a lower withdrawal to get a 95% safe rate. . . . . . . . . . . . . . . .

A withdrawal of about $16,000.00 is the highest withdrawal that would have survived 95% of the periods tested, using the equity split and other criteria you entered. This would be a withdrawal starting at about 16.00% of your starting portfolio, with adjustments for inflation.

 

[tozz]

That is, for clarity and consistency, shouldn't the bottom paragraph read:

A withdrawal of about $41,000.00 is the highest withdrawal that would have survived 95% of the periods tested, using the equity split and other criteria you entered. This would be a withdrawal starting at about 41.00% of your starting portfolio, with adjustments for inflation.

 

[SteveR]

$41,000
-
$25,000
__________

$16,000

The FIRE calc. says in year 1 you are reducing the withdrawls by $25,000. This give a net of only 16% ($16,000 divided into $100,000 the starting amount).

This is why it say 16% versus 41%.

 

[SteveR]

Speaking of FIRE calc. numbers.

I have run my numbers in several different calculators and have come to the conclusion that FIRE gives the best overall answer to the question "Will my $$$ last me until I drop dead?" The data from it should only be used as a tool and should be used with other calculators to determine your cashflow and how you are going to generate your cashflow year to year as each of our cases vary.

In my case, I am going to do a 72(t) to lower my tax hit later in life. My IRA is pretty heavy right now and will be major tax drain in my senior years. I have my fingers crossed on some aging stock options that are still underwater but I am not counting on them. Damn shame! Now I wish they had paid me in real money instead of options.

The plan is to sell off shares in my after tax account first as that will give me a lower tax (capital gain instead of income) and will give my other stuff a couple more years to grow before I hit them. My biggest gap in now until 59.5, but I can cover with after tax and savings. My life style will need to scale back some but we live pretty cheap except for some major stuff that was planned over the past couple of years. If I can only convince the spousal unit that new clothes are not required every time a new catalogue shows up in the mail.

Health insurance will always be a year to year risk as it could be dropped. DW is holding out two more years to get her pension and health insurance. That will give us two chances to have more or less full coverage until one or the other drops the coverage. We will both get a small pension that will be about what SS will pay when we start it at 62. We plan to live a bit large at first since we feel the need to kick up our heels after being enslaved for the past 30+ years. Travel will be frequent and we want to fly/drive cross country several times as well as visit family in Australia. I have 400,000 frequent flier miles to burn up so going first class Down Under will be great. Oh, well, we all need to have some sort of plan once we break away from MegaCorp and we do.

 

[Cb]

FIREcalc is one fine piece of work, but it sure would be nice to have '03 & '04 data added for those of us who look at 50+ year windows...

 

[tozz]

I agree.  I hope I'm not coming across as critical.  FIREcalc is da bomb. 

I don't think I am making myself clear, though.  No big deal, but I will try one more time.  Wouldn't the changes below increase the clarity and consistence of the FIREcalc results and recommendations?
_________________________________________________
FIRECalc Results
You have proposed a withdrawal of 45.00% of your starting portfolio.

We looked at the 116 possible 15 year periods from 1871 until 2002, and the 15 partial periods from 1987 until 2002, starting with a portfolio of $100,000 and taking out $45,000 the first year, and the same amount after adjustments for inflation (PPI) each year except as follows:

Starting in year 1, the withdrawal was decreased by $25,000 (adjusted for inflation). Pension

Starting in year 2, the withdrawal was decreased by $25,000 (adjusted for inflation). Move to low cost of living area

Starting in year 7, the withdrawal was increased by $25,000 (adjusted for inflation). Return from low cost of living area

In year 2, the portfolio was increased by a single (inflation-adjusted) $250,000 addition (as from the sale of a house). Home sale

In year 7, the portfolio was decreased by a single (inflation-adjusted) $150,000 reduction (as for a major purchase). Purchase cheaper home

Your Success Rate is 92.4%
In 7.6% of those 131 periods, the portfolio would have been fully depleted at or before the withdrawal in year 15. In 92.4% of the years, the portfolio would have maintained a positive balance through the withdrawal in year 15.
The average (mean) portfolio balance following the withdrawal in year 15 was $455,037.

(This assumes your portfolio consists of 25% in Commercial Paper and 75% in equities that behave like the market as a whole, with an overall expense ratio of 0.18%.)

FIRECalc's Research Suggests...
Checking for a lower withdrawal to get a 95% safe rate. . . . . . . . . . . . . . . .

A withdrawal of about $16,000.00 $41,000 is the highest withdrawal that would have survived 95% of the periods tested, using the equity split and other criteria you entered. This would be a withdrawal starting at about 16.00% 41.00% of your starting portfolio, with adjustments for inflation.

 

[Nords]

FIREcalc is one fine piece of work, but it sure would be nice to have '03 & '04 data added for those of us who look at 50+ year windows...

Here's the last I think that Dory had to say on the 2003 data:

I've had a few questions on this...

The source of the numbers is Yale Prof Robert Schiller, who has (so far) updated them at the end of each year for the previous year (i.e., 2003 results become available near the end of 2004).  As of today, he has published final stock and bond numbers, but only estimated inflation numbers. 

The detailed numbers on Firecalc output show the retirement year, and then the results each subsequent year. Since we don't have 2003 numbers there yet, there is no "next year" for someone retiring in 2002. That will show up as soon as I get and incorporate the new data. (And from a practical viewpoint, little will change when a new year is added, unless that new year is pretty abnormal compared to the existing years.

More when I see the updated data.  Dory36 

 

[Cb]

It looks like Schiller's data is updated through June of 2004 now.

http://www.econ.yale.edu/~shiller/data/ie_data.htm

Cb

 

[dory36]

Thanks for the link. I'll look at those numbers. When I last looked, a couple of months ago, there were missing pieces that prevented any useful update.