Basic question about the Detailed View Table

Post questions about how to use the planner, user inputs, how the planner works, and comments and suggestions.
vlads
Posts: 2
Joined: Mon May 23, 2016 3:01 pm

Basic question about the Detailed View Table

Post by vlads »

Apologies if this has already been answered elsewhere (a quick forum search didn't produce an answer)
I'm a bit confused about the treatment of taxes in the Detailed View Table.
If I understand correctly, "Planned Expenses" do not include any taxes: these are just anticipated net expenditures.
The "Median Total Withdrawal" clearly does not include taxes, since it never exceeds Planned Expenses, and is equal to Expenses when these are being funded at 100%.
Is the payment of taxes incorporated into the Portfolio Value? I.e., the Value is already net of taxes?
Thanks very much.
jimr
Posts: 821
Joined: Thu Feb 28, 2008 6:48 pm

Re: Basic question about the Detailed View Table

Post by jimr »

The short answer is yes, the taxes have already been taken into account. The planner logic assumes the basis of the taxable portfolio is 100% (eg all taxes already paid) and the basis of the tax deferred portfolio is 0% (eg all distributions fully taxable).

On the taxable side, portfolio gains are generated each year based on the return/std dev. Taxes are subtracted from the gains using the investment tax rate. The after-tax gains are then reinvested back into the taxable portfolio, keeping the taxable portfolio basis at 100%. These investment taxes are not reflected in any of the planner cash flow outputs.

On the tax deferred side, all portfolio gains are reinvested in full without any deduction for taxes. This keeps the tax basis of the deferred portfolio at 0%. Taxes are paid on deferred withdrawals at the income tax rate. These taxes are computed automatically and shouldn't be added by the user as expenses. These taxes are shown in the 'Taxes on Withdrawal' column in the Detailed View Table (with Show More detail option selected).

Also, RMDs and the default withdrawal order (taxable, deferred, tax free) are also important to consider when trying to understand the detailed view output.

Finally, if you right-click on any column header in the table, you'll see an option to 'show all columns' which adds even more information than the 'show more detail' option. If you can get the info to fit on your screen, you can right-click on any cell in the table to export it to excel.
vlads
Posts: 2
Joined: Mon May 23, 2016 3:01 pm

Re: Basic question about the Detailed View Table

Post by vlads »

Thank you so much for the reply. Is there perchance any documentation that defines what each column in the full detailed display represents? The column names are suggestive, but it would be good to confirm that my interpretation is correct.
jimr
Posts: 821
Joined: Thu Feb 28, 2008 6:48 pm

Re: Basic question about the Detailed View Table

Post by jimr »

That would be a good addition to the documentation, but unfortunately I haven't gotten to it yet.

I did add it to my to-do list.

Jim
Paul O
Posts: 9
Joined: Tue Oct 30, 2018 5:22 pm

Re: Basic question about the Detailed View Table

Post by Paul O »

Jim,

As stated in so many other posts, thank you for sharing such a rigorous program incorporating so may options. I have a question question regarding the withdrawal rate summarized in the Detailed View Table.

If I set the plan so it starts before retirement, the withdrawal rate for the first year of retirement is the Median Total Withdrawal (first year of retirement) divided by the LAST year's ending median portfolio value. This also applies to all subsequent years as long as there is a balance for the previous year. This is consistently followed through all years of retirement.

If I now set the plan so the first of retirement corresponds to the first planning year (i.e., plan starts at 67 and I retire at 67), the withdrawal rate for the first year is not calculated in the same way. The withdrawal rate for the first year of retirement is the Median Total Withdrawal (first year of retirement) divided by the CURRENT year's ending median portfolio value. As with the first case, the withdrawal rate for subsequent years of retirement is the Median Total Withdrawal (first year of retirement) divided by the LAST year's ending median portfolio value.

In this second case when the first year of the plan corresponds to the first year of retirement, why is the withdrawal rate not calculated as the Median Total Withdrawal (first year of retirement) divided by the initial portfolio balance (equivalent to LAST year's ending portfolio value)?

Regards,
Paul
jimr
Posts: 821
Joined: Thu Feb 28, 2008 6:48 pm

Re: Basic question about the Detailed View Table

Post by jimr »

The short answer to your question is that I was a bit lazy when I wrote that code.

The code where that's calculated only had access to the yearly ending portfolio value, not the starting value. To avoid triggering an out of bounds access error, it checks to see if it's the first year or not. If it's not the first year, it uses the previous year's portfolio balance for the calculation. If it's the first year, it uses the current year instead. This isn't in the simulation logic, but rather it's code that populates the detailed view table after the simulation is done running.

It turns out that getting access to the starting portfolio balance wasn't a big deal, so this will be fixed the next time I release the software.
Paul O
Posts: 9
Joined: Tue Oct 30, 2018 5:22 pm

Re: Basic question about the Detailed View Table

Post by Paul O »

Thank you for your prompt response and explanation Jim.

Again, thank you for sharing the simulation. I happened upon the questioned circumstance when I was manually checking output just to follow all the intricacies and logic in your simulation.

Paul
Paul O
Posts: 9
Joined: Tue Oct 30, 2018 5:22 pm

Re: Basic question about the Detailed View Table

Post by Paul O »

Jim,

A simple question about "Median RMD" in the detailed view table. Does the "Median RMD" represent the RMD associated with the "Median Portfolio Value" or does the "Median RMD" represent the median of the RMD used for the 10,000 simulations? I imagine that with 10,000 simulations the "Median RMD" would be similar no matter how it is calculated. Similarly, would your same answer apply to the "Median Total Withdrawal"?
jimr
Posts: 821
Joined: Thu Feb 28, 2008 6:48 pm

Re: Basic question about the Detailed View Table

Post by jimr »

The median RMD is the actual median RMD from the 10,000 iterations. It's directly mapped to the tax-deferred portfolio value, but only indirectly mapped to the total portfolio value (assuming taxable and/or taxfree are non-zero).

The total withdrawal amount is also the actual median value.

For most simulation variables, the sim saves off a copy of the amount from each year of each simulation iteration so it can later determine the median value.
Paul O
Posts: 9
Joined: Tue Oct 30, 2018 5:22 pm

Re: Basic question about the Detailed View Table

Post by Paul O »

Thank you Jim for the explanation regarding median RMD and median total withdrawal. I am evaluating the tax impacts of IRA to Roth conversions at various ages. Looking at the detailed view output, I was trying to figure out why in the detailed view table that the total balance is exactly equal to the sum of taxable, tax-deferred, and tax-free balances. I assume that if the taxable, tax-deferred, and tax-free balances are medians representative of probably different simulations in the sequence of 10k simulations, that the sum would be different from the median portfolio value (also likely a different simulation). For the component medians to exactly equal to the median total, I assume either:

1. That the taxable, tax-deferred, and tax-free components reported in the detailed view table are not the median of 10k simulations but that the components correspond to that of the median total portfolio value.

or

2. That the taxable, tax-deferred, and tax-free values are medians (representing different simulations) and that the total portfolio value in the detailed view table is the sum of the taxable, tax-deferred, and tax-free medians.

Looking at the portion of code that you posted, it looks like case #2 is maybe what the FRP utilizes. Is this the case or is there another reason why the component medians exactly equals the median total?

Regards,
Paul
Post Reply

Who is online

Users browsing this forum: No registered users and 27 guests