Firecalc compared with Vanguard Retirement Nest Egg Calculator

[ranchoparque]

I notice that when I plug in my numbers into Firecalc I get much more optimistic results than with the Vanguard calculator. Firecalc gives me a 3.65% SWR and Vanguard just under 3%. This is for 40% equities and a 32 year time frame. 100% success rate.
Has anyone noticed this difference between the two calculators? Try plugging in your numbers and see your results. How do you choose your SWR? What rate would you feel comfortable with if you had my time frame? I am 63 and I am planning to age 95.

https://retirementplans.vanguard.com/VGApp/pe/pubeducation/calculators/RetirementNestEggCalc.jsf

 

[REWahoo]

I believe the Vanguard calculator uses Monte Carlo simulation. FIRECalc looks at actual history, and that could account for the difference.

 

[MRG]

I notice that when I plug in my numbers into Firecalc I get much more optimistic results than with the Vanguard calculator. Firecalc gives me a 3.65% SWR and Vanguard just under 3%. This is for 40% equities and a 32 year time frame. 100% success rate.
Has anyone noticed this difference between the two calculators? Try plugging in your numbers and see your results. How do you choose your SWR? What rate would you feel comfortable with if you had my time frame? I am 63 and I am planning to age 95.

https://retirementplans.vanguard.com/VGApp/pe/pubeducation/calculators/RetirementNestEggCalc.jsf

I've run most every calculator on the net before retiring. Firecalc, Fidelity RIP, T. Rowe., Vanguard... and they all produce wildly different results.

What's the question?

Seriously they all have different approaches to solve the same problem. Different people tried to solve it with their ideas of the best software and DATA! The answer is different between them(like REWahoo said). They better give different results or they're wrong! Of course that doesn't make any of them right.

I took the most conservative(Fidelity RIP for my situation at that time) and asked if that was an OK worst case scenario for planning.

Good luck and best wishes.

 

[43210]

100% success rate.

You get silly answers if you want 100% success rate, for a predetermined spending rate. It's just an abstract calculation.

Instead, you simply have some flexibility for spending level. You can spend more on average if you are flexible.

 

[ranchoparque]

I've run most every calculator on the net before retiring. Firecalc, Fidelity RIP, T. Rowe., Vanguard... and they all produce wildly different results.

What's the question?

.

I think I am looking at opinions on what would be a conservative safe withdrawal rate for my time frame (and asset allocation). Obviously if I am conservative I would pick the lowest rate.

 

[braumeister]

I think I am looking at opinions on what would be a conservative safe withdrawal rate for my time frame (and asset allocation). Obviously if I am conservative I would pick the lowest rate.

Every calculator will give you an opinion, but they will never be quite the same and often will be very different. The reason is as MRG stated above.

IOW, there is no certainty -- only opinions and you have to make your own decisions. They do give you a good general idea though.

 

[CardsFan]

Trying to get a "realistic" 100% withdrawal rate with a Monte carlo calculator is a fool's errand.
The Vangard simulator runs 100,000 random cases. Some (many?) will not be remotely realistic relative to how the market actually moves. So, to get 100% you need to drop the WR quite a bit.

As 43210 said, you should only use these calculators as an indicator. They are just a calculation based on possibilities.

 

[GravitySucks]

You get silly answers if you want 100% success rate, for a predetermined spending rate. It's just an abstract calculation.

Instead, you simply have some flexibility for spending level. You can spend more on average if you are flexible.

Yup. It's foolish to think there is a 100% success rate to any of this.
The end of the world won't care what your SWR is.

 

[43210]

OP, consider this. Suppose you get your first job ever, and you decide that you are going to spend exactly $X per year (inflation adjusted) for the rest of your life and you will stick to that spending regardless of what transpires financially.

Do you see how ridiculous that is?

Also if you make that initial once-only choice of $X so that you have 100% chance of never going broke, then you'll have to choose $X very low.

While working, you can adjust spending depending on circumstances, and it's the same with retirement.

 

[OldShooter]

Really, retirement calculators are pretty much doomed to be garbage in and gospel out devices.

Question: How do we know that the designers of retirement calculators have a sense of humor? Answer: They use decimal points.

Why garbage? Because no one knows what will happen in the future. The past really does not predict the future to a degree where we can rely on it in any major way.

First and most importantly we do not know when we are going to die.

But if you want to obsess over methodology, take standard deviation for example. The economists have a lot of fun with standard deviation by ignoring the fact that the mathematics only applies to nice, neat, gaussian distributions. We know that stock prices distributions have long tails, are not centered on zero, and are not symmetric. This already blows statistical mathematics and classic Monte Carlo methods out of the water, but the big torpedo is that stock price samples are not independent. If they were, there would be no talk of "momentum," which is a well-observed phenomenon. So really we are left with heuristics, a fancy word for educated guesses, as @MRG points out. (https://en.wikipedia.org/wiki/Heuristic).

The fact that different calculators deliver different results is really a teaching moment pointing out the uncertainty of the process. IOW, the "S" in SWR is bogus. The good news, though, is that we are not robots who must select a WR and leave it unchanged. We can pick one and change it every few months as the future is revealed to us. If we don't have a lot of money, we can and should pick a conservative one. If we have more money, we can start with a larger WR and, again, adjust as the future is revealed. If we have time and interest we can fiddle with the various calculators along the way, of course, but to believe them is to not understand the problem.

 

[ranchoparque]

Thanks for all of the above comments. I realize there is no real answer to this frequently discussed topic. I does help me to see how others approach this issue.

 

[43210]

Thanks for all of the above comments. I realize there is no real answer to this frequently discussed topic. I does help me to see how others approach this issue.

Not so fast. Yes there are various uncertainties. But

understanding plus uncertainty

is massively better than

misunderstanding plus uncertainty.

 

[Ready]

I had the same question and I asked my Vanguard rep about this last month. He said the monte carlo simulation includes some runs that are far worse than the worst case historical results used in Firecalc. So essentially they are planning for something even worse than the great depression, or the heavy inflation years that caused the 4% SWR to fail.

 

[pb4uski]

How many angels can dance on the head of a pin?

Vanguard calculator with 40/60 AA, 36 years and 3.6% WR.... three runs...92% in each case.

Firecalc with same crietria....97.3%

Firecalc with Monte Carlo option and 7.8% mean portfolio return... 51.2%. 74.4%, 75.8%, 79.1%, 96.3%, 89.3%.

IMO, 3.6% is sufficiently conservative for a 63 yo. YMMV.

 

[The Onceler]

First and most importantly we do not know when we are going to die.

Just for fun -- On a stand-alone basis (for retirement planning) it would be convenient if we each knew the date of our death. But can you imagine what the world would look like if everyone had that information? To eliminate the prospect of mortality uncertainty and mortality hazard would vastly increase the risk of moral hazard. Big problem and probably mass, worldwide chaos.

 

[Lawrence of Suburbia]

Is a 94.9% probability of success in Firecalc an okay result? I ran it out 30 years, to be conservative; but truly, my health is so bad I'd be stunned if I saw 80 or 85.

 

[garyt]

Is a 94.9% probability of success in Firecalc an okay result? I ran it out 30 years, to be conservative; but truly, my health is so bad I'd be stunned if I saw 80 or 85.

I certainly think so. Firecalc runs, what, 117 cycles? So that means that you would fail in only 7 of the worst scenarios in history. I like those odds.

 

[braumeister]

How many angels can dance on the head of a pin?

Well, that has been calculated for the point of a pin here, so it's safe to assume it would be a much larger number for the head of the same pin.