Firecalc says I'm good, Fid. RIP says no

[mrWinter]

I've got a long time before retirement, 20 years according to current plan.

When I run firecalc it generally says my plan is good, as does the other popular historical backtesting calculator that shall not be named (the Voldemort calculator?). However when I run things in Fidelity RIP I get a much more pessimistic outlook. For example firecalc says I can spend 120k a year upon retiring with my current plan with 100% success. FRIP says "You may have 71k a year".

I'm wondering if others have seen similar results with RIP being more pessimistic or not. I know RIP uses Monte Carlo simulation, (which I don't like as much) and wonder if perhaps this particularly long time frame of savings before retirement somehow makes this outlook more pessimistic.

I looked at the methodology of RIP but wasn't 100% clear on the logic they use. I'm not sure if the simply roll the dice for each subsequent year of the simulation run, and thus sometimes get a ton of very negative return years in the simulation, or of they also do some kind of analysis of the previously rolled numbers and make it less likely to roll more bad years if all recent past ones have been bad too, and thus avoid simulating a whole decade of negative returns for example.

 

[nwsteve]

Take a look at the RIP cash flow table--it is an option on the results page with the graph. This table will show what your balances are doing and how the model is treating your investments/spending. Also RIP incorporates a much higher rate of inflation for healthcare expenses than most other models that tend to just use a historic overall average

 

[Spock]

I pounded on various retirement calculators for years. Out of 5 popular, well know websites, 2 would say I'd be eating out of a trash can in 10 years, 2 would say I'd die with more money than I have now, and 1 would goldilocks spending the last dime just as I reach projected life span.


I ended up building my own spreadsheet that had different tabs for different calc methods, budgets, FV and PMT functions, projected spend each year, etc.

With 20 years to go, I suggest this:
1. Live below your means, save as much as you can (after tax is important if you are going to RE, look at the national debt-do you really think tax rates will be lower long term?).

2. "Live a little" now. You could be dead tomorrow and then what will your savings get you?

3. Friends will be a scarce resource later. Grow and cultivate a large inventory OUTSIDE OF WORK now. (Friends are people you share meals with, not chat with on Facebook).

 

[Dtail]

I use both calculators and find that Fidelity is about 11-12% lower. At 120k in Firecalc, something seems off with RIP.
What tax rate are you using with RIP, plus what medical expenses did you input?

Edit - also not sure of the specific methodology, but the results are not truly random Monte Carlo wise, coz if you run the results 10 different times without changing the inputs, the results are exactly the same.

 

[MRG]

Of all the calculators I've run RIP is the most conservative.

 

[mrWinter]

I use both calculators and find that Fidelity is about 11-12% lower. At 120k in Firecalc, something seems off with RIP.
What tax rate are you using with RIP, plus what medical expenses did you input?

I didn't explicitly enter any medical expenses, I just input an estimated monthly retirement expense, breaking out essential and discretionary, I didn't do the detailed worksheet. Maybe I should clarify that I'm actually running the "Planning & Guidance Center" which apparently "includes functionality previously available in the Retirement Income Planner, Retirement Quick Check, and Portfolio Review tools", not sure if any of this is different from what others are using or have used recently.

Edit - also not sure of the specific methodology, but the results are not truly random Monte Carlo wise, coz if you run the results 10 different times without changing the inputs, the results are exactly the same.

Agreed, I noticed that too and found it fishy, whenever I revisit the page it says "Now we're running approximately 250 simulations" and then comes up with the exact same numbers. Either they are faking the 'rerun' and making me wait for the webpage to load for dramatic effect, or there is no randomness in it.

 

[Dtail]

I didn't explicitly enter any medical expenses, I just input an estimated monthly retirement expense, breaking out essential and discretionary, I didn't do the detailed worksheet. Maybe I should clarify that I'm actually running the "Planning & Guidance Center" which apparently "includes functionality previously available in the Retirement Income Planner, Retirement Quick Check, and Portfolio Review tools", not sure if any of this is different from what others are using or have used recently.



Agreed, I noticed that too and found it fishy, whenever I revisit the page it says "Now we're running approximately 250 simulations" and then comes up with the exact same numbers. Either they are faking the 'rerun' and making me wait for the webpage to load for dramatic effect, or there is no randomness in it.

It sounds like you are running the same calculator as me. Are you a current Fidelity customer? I believe the programs might be different otherwise.

Did you enter a specific tax rate?

So using my made up term, the methodology is kind of a "fixed result run in a random Monte Carlo simulation". It is running random simulations (vs. running them in historical actual stock mkt sequences) but in the same sequence each time so same results. Who knows, but more important to figure out why your number appears artificially low.

 

[DEC-1982]

I use RIP and I think being conservative (for financial planning) is good. YMMV.

 

[mrWinter]

It sounds like you are running the same calculator as me. Are you a current Fidelity customer? I believe the programs might be different otherwise.

Did you enter a specific tax rate?

So using my made up term, the methodology is kind of a "fixed result run in a random Monte Carlo simulation". It is running random simulations (vs. running them in historical actual stock mkt sequences) but in the same sequence each time so same results. Who knows, but more important to figure out why your number appears artificially low.

Yes I am a fidelity customer. I did not enter a specific tax rate, just left it at the default, which is apparently 18%.

I use RIP and I think being conservative (for financial planning) is good. YMMV.

Agreed, but only to an extent. A 40% reduction in potential spending is more than conservative.

 

[Dtail]

Yes I am a fidelity customer. I did not enter a specific tax rate, just left it at the default, which is apparently 18%.



Agreed, but only to an extent. A 40% reduction in potential spending is more than conservative.

Okay just double checking - when you inputted your expenses in Fidelity, did you leave out your estimated taxes, since Fidelity is already inputting that number for you?
Just trying to figure out the 40% diff.

 

[mrWinter]

Take a look at the RIP cash flow table--it is an option on the results page with the graph. This table will show what your balances are doing and how the model is treating your investments/spending. Also RIP incorporates a much higher rate of inflation for healthcare expenses than most other models that tend to just use a historic overall average

The "Hypothetical Assets Table In Significantly Below Average Market" is telling. It looks extremely pessimistic to me (or am I being naive?). The portfolio returns for the next 20 year of savings are:

-0.19
-0.04
-0.06
-0.04
0.01
-0.03
-0.03
0.09
0.05
0.03
-0.03
-0.01
0.10
0.01
-0.01
-0.02
0.03
0.00
0.07
-0.03

Followed by returns minus withdrawals giving portfolio changes (not returns) during retirement of:

-0.03
-0.13
-0.25
-0.14
-0.20
-0.26
-0.47
-0.69
-1.00

for the next 9 years at which point I run out of money. I'm not really sure how to calculate the returns it's forecasting those years since I'm not sure what amount it's withdrawing each year, but it looks like a similar picture.


These returns for the next 20 years seem very very pessimistic to me, a 20 year CAGR of -0.7% by my calculations. Am I wrong in thinking this is too pessimistic to use for planning? This is much worse than has ever happened in history before right? (20 year CAGR of the S&P from 1960-1980 was +7.9% according to this calculator) Possible? Sure. A good 'conservative' model for planning purposes? I don't really think so.

 

[DEC-1982]

Regarding post #11 ...

This is puzzling, because my returns are quite different.

I had recently made some changes in my portfolio, so I updated FIDO RIP today and reran the simulation. In the "Significantly Below Average Market Conditions" I see

2018 -11.90%
2019 - 1.78%
2020 + 0.26%
2021 + 0.62%
2022 + 5.46%
2023 - 2.70%
2024 + 0.22%
2025 + 6.91%
2026 + 5.20%
2027 + 5.94%


Now the calculations I am doing to determine this rate is as follows:

=(D59-(D58-E58))*100/(D58-E58)

where column D has the beginning balance
column E has the withdrawals


Are you also subtracting the withdrawals from the current year balance, like I am? especially for the numerator, when calculating profit or loss for the year? because IMO the calculation needs to handle the fact that withdrawals were made.

 

[mrWinter]

Yes I believe I'm doing the same thing, except I have savings not withdrawals. Here's the raw numbers, columns A and B coming from the website, C is constant savings rate, D is just balance minus savings to give the balance due to growth. E is calculated with the formula =ROUND((D3-B2)/B2*100,1):

year|balance|savings|bal-savings|return
start|625000|||
2018|$537,868 |30000|$507,868 |-18.7
2019|$546,775 |30000|$516,775 |-3.9
2020|$541,635 |30000|$511,635 |-6.4
2021|$548,720 |30000|$518,720 |-4.2
2022|$585,880 |30000|$555,880 |1.3
2023|$597,678 |30000|$567,678 |-3.1
2024|$610,073 |30000|$580,073 |-2.9
2025|$694,372 |30000|$664,372 |8.9
2026|$759,012 |30000|$729,012 |5
2027|$812,404 |30000|$782,404 |3.1
2028|$819,336 |30000|$789,336 |-2.8
2029|$843,453 |30000|$813,453 |-0.7
2030|$956,561 |30000|$926,561 |9.9
2031|$993,837 |30000|$963,837 |0.8
2032|$1,011,888 |30000|$981,888 |-1.2
2033|$1,019,426 |30000|$989,426 |-2.2
2034|$1,083,161 |30000|$1,053,161 |3.3
2035|$1,110,656 |30000|$1,080,656 |-0.2
2036|$1,223,871 |30000|$1,193,871 |7.5
2037|$1,214,015 |30000|$1,184,015 |-3.3
2038|$1,148,197 ||$1,148,197 |
2039|$995,633 ||$995,633 |
2040|$751,410 ||$751,410 |
2041|$644,578 ||$644,578 |
2042|$515,892 ||$515,892 |
2043|$380,639 ||$380,639 |
2044|$201,703 ||$201,703 |
2045|$63,402 ||$63,402 |
2046|$0 ||$0 |

2038 is the retirement year, hence 2037 being the last with savings.

I don't know what withdrawal it was applying from 2038 onward so I didn't calculate any return for that period because I didn't know what withdrawal to subtract in column C.

I'm actually under-representing the savings number, it's a little over 30k in these simulations.

 

[mrWinter]

I may be talking to an empty room at this point, but given those simulation returns, this is inclining me more toward my opinion that a Monty-Carly simulation is only useful in looking at average performance. These edge cases of long-term terrible performance don't educate much. Why not simulate -10% returns every year? Why not -15%? Why not -80%? Could these terrible situations come to pass? I guess, but anything could happen. I don't see a lot of point in planning for things far far worse than has ever happened in history before. I guess it could happen but you can't prepare for every black-swan possible.

Hopefully this isn't just confirmation bias and I'm not just disliking Fidelity RIP because it gave me an answer I didn't like.

 

[DEC-1982]

Before you give up on the product, have you tried talking to FIDO, to ask them to explain the results?