Post #3 of 3 To Trust or Not to Trust
longinvest wrote: Sun Sep 14, 2025 10:05 am
Let’s have fun with negative pensions and disabling robustness features of the VPW worksheet.An 18 years old investor with a $0 portfolio has just inherited a trust that will pay $100,000/month from age 60 to age 90. To investigate the situation, the investors enters a $100,000/month “Trust (100%)” pension starting at age 60 and a -$100,000/month “Trust (-100%)” pension starting at age 90. Having read everywhere on the Internet that young investors should hold a high-stocks portfolio, the investor selects a 90/10 stocks/bonds allocation:
Apparently, the investor could spend as much as $133,914/year (for taxes and expenses) by borrowing this annual amount from age 18 to 59, until the pension kicks in at age 60.
Curious, the investor investigates the possibility of using a different asset allocation and discovers that the spending amount could be increased to $234,553/year by simply adopting a conservative 30/70 stocks/bonds allocation.
It’s fun to spend future income today, isn’t it? It’s even more fun to play with negative-value portfolios that are never exposed to margin calls.
Of course, I’ll continue to be criticized for having robustness built-in into the VPW worksheet and not allowing for spending future income today. I don’t care.
(Note: I’ve addressed the sarcastic tone of this post in a previous post above, Post #1 of 3)
This example is so preposterous as to be laughable.
But before we get to the merits, there is a discrepancy to deal with.
Results Discrepancy:
I downloaded a new VPW sheet and entered your inputs from your 1st Trust example. I had to alter the restrictions of 2 cells (cells E19 and E20 to accept Age 90 and -$100,000, respectively), which I did based on my instructions written in an earlier post. Strangely, I did not get the same result that you show.
I then downloaded a new sheet and entered your inputs twice more and received the same results as the first time, which were quite different than your results.
I then entered the inputs for your 2nd example and again got very different results.
Here are the screen shots of my inputs and results:
90/10 Portfolio
30/70 Portfolio
As far as I can see, there are four possibilities, each applicable to both of us, which creates 8 possible reasons for this difference.
1. I was using a corrupted sheet
2. I misunderstood the alterations I suggested earlier
3. I entered a typo by mistake
4. I purposely entered incorrect/additional information to mislead
5. You were using a corrupted sheet
6. You misunderstood the alterations I suggested earlier
7. You entered a typo by mistake
8. You purposely entered incorrect/additional information to mislead
1. As I said, I downloaded three new sheets so I don’t see how this is likely.
2. I think it’s safe to say I didn’t misunderstand myself.
3. I’ve checked multiple times and don’t see a typo.
4. All I can do is swear that I entered the inputs exactly as shown in the screenshots, and no additional information, with no intent to mislead.
5. – 8. I cannot speak to any of these possibilities.
While I’m confident 1-4 are not the cause of the difference, I’m open to finding out I’m wrong about that. I’m also open to finding out there is an additional possible cause.
If my results are correct:
If my results are correct, then it seems the sheet is behaving exactly as it should. Telling the would-be investor that he cannot retire at 18, and instead needs to add approximately $16k or $43k (for a 90/10 AA or 30/70 AA, respectively) to the portfolio each year.
Of course the investor shouldn’t be using the retirement sheet for the accumulation phase.
If instead, it is used to test out retirement scenarios at age 60 and 80* (when the Trust starts and stops), then both methods of entering the Trust produce identical results. The results seem realistic in telling the investor, with a portfolio of $0, that they can retire at 60 on $857,235 per year, and deposit $342,765 into the portfolio from the Trust funds. Of course they would need to make enough money for taxes and expenses from 18 until age 60. This wouldn’t be my choice or recommendation but it doesn’t seem broken.
*(I changed the Stop Age to 80 so equivalent inputs could be used in both the altered (Method 4) and unaltered (Method 5) sheets)
Age 60 Altered:
Age 60 Unaltered:
Age 80 Altered:
Age 80 Unaltered:
If your results are correct:
Even if my results are incorrect and yours are correct, this example is preposterous. To compare entering a small reduction in pension, using a negative entry, with an 18 year old deciding to borrow $5,624,388 (not adjusted for inflation) to live on until receiving their Trust funds is prima facia ridiculous. It’s akin to leveraging, which goes against the Bogleheads philosophy and is strongly discouraged.
Of course, the investor doesn’t need an altered VPW sheet to give them a bad idea.
*Warning: Sarcastic tone ahead (to mimic longinvest’s tone above)*
Let’s have fun without negative pensions and without disabling robustness features of the VPW worksheet.*
(*I must credit longinvest with the inspiration for this comment, or risk being accused of plagiarizing )
Our intrepid investor enters the Trust amount using the Standard Method (Method #5 from my chart) instead of the negative entry method (Method #4 in my chart). Then, undaunted, he enters a retirement age of 60, and then, thru trial and error, quickly discovers that he could retire at 60 on $1.2 million per year by borrowing $80,416 per year (actually less, when taking market returns into account) and adding it to his portfolio.
Curious, he enters a retirement age of 53 and discovers he could retire at 53 on over $100,000 per year by borrowing only $23,810 per year (actually less, when taking market returns into account) and adding it to his portfolio. Aren’t those fun options?
*Note: End of sarcastic tone.*
So a tool can be misused whether altered or not.
longinvest wrote: Sun Sep 14, 2025 10:05 am
Of course, I’ll continue to be criticized for having robustness built-in into the VPW worksheet and not allowing for spending future income today.
Literally, no one has done this.
This began with Sspies asking a simple objective technical question. If it had remained this way, with a clear demarkation between subjective opinions and objective facts it would have been a much shorter and clearer discussion.
I think we have both made our positions clear. I remain interested and open to any objective examples of the sheet being broken (internal error messages, incorrect results) by using one of the methods listed in my chart from above. I would also be interested to discover why the results of the Trust examples were different.
If there are no objective examples forthcoming, I believe I am done with this particular discussion. My reply was delayed due to being in the middle of a medical situation, so I need to attend to that.
