Firecalc vs. Fido RIP

[big-papa]

Not sure if this one has been discussed. I'm not FIREd yet (but wanna be in the next 3 years or so). I recently ran some "what if's" through both Firecalc and Fido RIP. What I was looking at is the difference if my wife and I take SS at age 62 vs 67 vs 70.

Ignoring the absolute withdrawals, I noticed something I wasn't expecting.
Fido indicated that the longer I waited, the higher my withdrawal rate could be, which is what I expected.
Firecalc showed just the opposite. Max withdrawal for 100% safe dropped the longer we waited to take SS.

I also have my own spreadsheets and did the same analysis and the results were similar to Firecalc.

I suspect that the main difference might be taxes. Firecalc and my simplistic spreadsheet ignores them and Fido asks what you what your expected marginal tax rate would be.

FYI - in FIDO I'm using today's dollars and an underperforming market.

Big-Papa

[braumeister]

FIRECalc doesn't ignore taxes -- they are part of your total spending input.

[mathjak107]

a few differences.

fido automatically increases healthcare and long term care costs by 5.50% a year , fido uses monte carlo simulations to try to find even worse case wscenario's that we could have had but didn't . but , fido uses a 90% success rate for worst case . i much prefer 95 to 100%

so the 90% success rate with the more stressful testing and assumptions comes close to firecalc at 95-100%

[big-papa]

Quote:

Originally Posted by braumeister FIRECalc doesn't ignore taxes -- they are part of your total spending input.

What I meant was that it isn't treated separately from the rest of your spending.

[big-papa]

Quote:

Originally Posted by mathjak107 a few differences. fido automatically increases healthcare and long term care costs by 5.50% a year , fido uses monte carlo simulations to try to find even worse case wscenario's that we could have had but didn't . but , fido uses a 90% success rate for worst case . i much prefer 95 to 100% so the 90% success rate with the more stressful testing and assumptions comes close to firecalc at 95-100%

That would make sense since we're most likely dealing with a long tail in the distributions - longer even than the worst historical case which is what I use in my own spreadsheet which includes assets not included in Firecalc (like intermediate term treasury bonds).

A question - how do you translate the score Fido gives into a 90% success rate? Or, rather, what score = 90% success rate? What I did was to alter total expenses each time I ran simulations with different SS start dates until I got the same high score.

[mathjak107]

being retired i don't get a score . i just get a pass or fail based on worst case scenario stress testing which according to the data is at a 90% success rate

[nwsteve]

BTW, Fido's tax rate input is AVERAGE/Effective expected Fed Rate will be Not marginal.
per Fido "Your effective tax rate is the total rate you pay on all of your taxable income after deductions. You can determine your annual effective rate by dividing the taxes you paid in the year by your taxable income for that year. Your effective rate will always be lower than your marginal tax rate, which is the rate you pay on the income that falls into the highest tax bracket you reach."
Just sayin ;-)
Nwsteve

[big-papa]

Quote:

Originally Posted by mathjak107 being retired i don't get a score . i just get a pass or fail based on worst case scenario stress testing which according to the data is at a 90% success rate

Ah that's a good piece of info. I might be able to fool it into thinking I'm already retired, but will need to fudge the next 3 years of savings into it. Thanks!