Quick Question

[pjddjp]

Hi Jim,
Thank you again for your excellent program, and for your patience with questions.
The attached image was produced with investment return at 3.5%, no standard deviation, and inflation at 2%, no standard deviation.
In the tax-free column, can you please tell me what formula the program is using to get from $257,923 to $261,792, in the first two rows?
There should be nothing affecting this other than inflation and investment returns, since the tax-free account is not being touched.
When I calculate this, I use $257,923 x (1+.035) / 1.02, and the result I get is close, but not exact, $261,716, instead of the program's $261,923.
Thanks very much. I'm pretty sure this will be my last question. (o:
DDD

[jimr]

I think you may have found a kind of weird bug.

By any chance could you have originally entered 3.53% for the return instead of 3.5%? If you did, this would explain what you're seeing.

If the return is 3.53%, then $257,923 x (1+.0353) / 1.02 = $261,792

The reason this is confusing (and technically a bug) is that the "return avg" field should only be allowed to have one digit after the decimal point. The input verification happens when you hit run. The program goes through each input to validate it and also rewrites each input field according to some preset formatting rules (including things like the number of digits after the decimal point).

The bug is that the original value you input erroneously gets captured before the extra digits are removed and is sent into the simulation. So even though the input looks like it was truncated to one decimal place, the value sent to the simulation has as many decimal places as you original entered.

If you hit run a second time without changing any inputs, the correct value (3.5 instead of 3.53) will be used the second time.

Anyhow, I think that's likely what you're seeing. Does this explanation seem plausible?

[pjddjp]

Thanks Jim,
Well, your explanation makes good sense, but unfortunately the program does not reset as you suggested, and, worse yet, I never did enter 3.53%. I would have no reason to enter that kind of precision for investment returns.
I'm curious if that's a bug that can be fixed, but now that I know I'm not insane, I the results are certainly close enough for government work, as they say.
Thanks so much for getting back to me - and let me know if you ever update the code.
Warmly,
Danny

[jimr]

Actually, it sounds like something different must be going on in your case then. The issue I was talking about would only occur if you actually did originally type in 3.53%. Since you didn't do that, this isn't the issue and I can't explain the numbers you're seeing.

Do you have anything configured in the upper table in additional inputs by any chance?

Interestingly, the tax deferred portfolio also looks like it's using 3.53% for return instead of 3.5%
For example, in excel using =2091722*1.0353/1.02 -> 2,123,098 which matches the second entry in for the tax deferred portfolio in your screenshot.

So aside from you actually entering 3.53% for the return, I can't explain what you're describing and I can't reproduce it on my end. When I set things up according to your screenshot, I get the expected amount for the second year tax free portfolio value instead of the amount you're getting.

If you're up for it, you could email the .frp file and I can look into it further. The only caveat is that email isn't secure so it's important to remove any personal information from the file and also to send it to me using an anonymous email account that doesn't have your name associated with it.

My email is info@flexibleretirementplanner.com

Jim

[pjddjp]

Oh, but there is another, more substantial problem. I get the same results initially, but for that small error, but after 42 years, the taxable results from the program show a taxable account value of $1,778,438, while I get a substantially different result, $2,581,597. I'm 90% sure my coding is correct. I imagine it would be really hard for your to understand the discrepancy without knowing all of my inputs, so I'll just attach them here. I would love to hear any insights.
Let me know if you need to see my calculations, and how I can send you the XLSM file.
Attachments are low-res due to web page's size constraints, and must be delivered one per post, so expect several. Sorry for the message bomb.
PAGE 1 IS ATTACHED.
Danny

[jimr]

I guess there is some stuff in additional inputs, so I'll need more pages from the report to reproduce what you're seeing.

[pjddjp]

PAGE 2 HERE
(Web site blocks frequent resubmissions.)

[jimr]

Also, are you using the latest version of the program (version 04.03.03 Built on 7/10/2020)?

You can get the version by clicking the Help->Release Notes menu option at the top of the main window.

[pjddjp]

Hi Jim,
I didn't have latest version. I'm so glad you fixed the discrepancy between initial vs. retirement portfolio value, when the retirement year is in the past.
Regarding the discrepancy I mentioned, it's unchanged.
I think the web site is a tough medium for communication about subtleties and for exchanging documents. Let me know if there's another way to correspond.
In the mean time, I'll try to have a look at the source code to try to iron out the inconsistency between the program's results (mine are 22% higher), and will let you know if I think there's something worthy of your attention.
Thank you!
Danny

[jimr]

You can correspond by email at info@flexibleretirementplanner.com