Re: "Brian, I think it may be a fluke of your posting style, but your posts come off like you are attacking, almost accusing, Pralana of being deliberately cryptic."
If so, I apologize.
In a coorporate email etiquette class in the 1990's we were told to explain precisely what was meant by the following in an email:
"I did not say you stole the idea."
The meaning changes completely depending on which word gets the accent, if there is an accent. There is no way to know what the sender meant.
However, thank-you for bringing this to my attention. I will try to be mindful. When dealing with irreversible monetary decisions like Roth Conversions, I admit to some vexation and frustration at the steepness of the learning curve and unexpected behaviors.
This also reminds me of Scottie in an original Star Trek episode to the insulting Klingons - Scottie stood fast when they trashed Captain Kirk. When they trashed his ship, however, Scottie threw his punch 😉
-- Brian
1. At each bracket for each year for each pass, how many data points do you use within each bracket, how are they selected, and why?
Only one - there's no sampling. For a given set of parameters, Pralana is deterministic.
2. Once Pralana optimizes for Year N, Iteration Y, the algorithm will never revisit year N until Iteration Y+1. Is this a correct understanding?
I believe so, yes.
3. If the preceding is true, there is no way to know if a decision for Year N, Iteration Y precluded better subsquent outcomes. Is this a correct understanding?
I think you mean, having changed a later year, there's no way to know that a different choice for an earlier year would now be better. I think this is technically true, although that's what pass 3 is for.
a) When Pralana compares the regular investment account balance with other balances, is it using an effective balance or nominal balance for the investment account? How is this effective balance defined or computed over time?
There is no comparison between balances of the different accounts. The overall Effective Savings value is the sum of all the account balances, where tax-deferred balances are discounted by your specified Effective Tax Rate.
b) Do I understand correctly that Pralana forces the regular investment account to liquidate and buy assets to maintain the fixed stock/bond ratio? I infer this because of Capital Gains that are reported with no Investment withdrawals. If so, when is this balancing done? There is a toggle for account growth at end of year or middle. Is this also when rebalancing is done?
To maintain your overall asset allocation, Mode 2 will rebalance accounts in the order specified under "Mode 2: Account Prioritization". If your taxable account is high enough on the list, then yes, it will be rebalanced if necessary, possibly generating capital gains. I believe this rebalancing is always done at the end of each year.
c) When doing Roth optimizations for a given year, is Pralana looking at the balance before or after such forced adjustments, if such adjustments occur?
I'm pretty sure the changes in balances due to the conversion happen before the rebalancing.
d) Financial Assets->Asset Class Taxation refers to the Investment Account. I still grapple with what is meant by the cash account, what its portfolio is, etc. I tried setting the ceiling and floor to 0, but this is ignored. I tried setting it to 1, and get the same result as a blank. I prefer to use my investment account rather for all my cash needs. Is the cash account portfolio the same as the "regular investment account"?
The cash account is like the clearinghouse for all transactions; you need to have a sufficient minimum balance in order for expenses to be paid efficiently. I use a floor of $30K and a ceiling of $50K. If you wish, you can specify a separate cash allocation within other accounts (for *holding* cash). I'd suggest reading up about the cash account in the manual.
e) I can manually order my liquidations in much more elaborate ways than allowed in Financial Assets->Management->Regular Savings Withdrawal Controls. I can leverage LTCG, STCG, dividends, interest, capital losses, various tax exempt bonds. Pralana does not give me sufficent fine grained control over liquidations to test strategies manually against Roth Conversions. Nor does Pralana run simultaneous optimizations between liquidation strategies and Roth Conversions. Is this correct?
Yes, Pralana has limitations. If you really need that degree of flexibility, I think you'll need to develop your own spreadsheet/software (or hire someone to do it).
f) If Mode 1 can solve this problem, then I will need to master Mode 1.
Mode 1 is problematic for optimizing Roth conversions unless each account has the same asset allocation (which is generally not tax-efficient). If your Roth account has faster-growing assets than your tax-deferred account (which typically *is* tax-efficient), the optimizer will recommend massive Roth conversions to take advantage of the faster growth, which will increase your portfolio's risk substantially. The "benefit" comes from this increase in risk, not from the conversions per se.
d) Financial Assets->Asset Class Taxation refers to the Investment Account. I still grapple with what is meant by the cash account, what its portfolio is, etc. I tried setting the ceiling and floor to 0, but this is ignored. I tried setting it to 1, and get the same result as a blank. I prefer to use my investment account rather for all my cash needs. Is the cash account portfolio the same as the "regular investment account"?
The cash account is like the clearinghouse for all transactions; you need to have a sufficient minimum balance in order for expenses to be paid efficiently. I use a floor of $30K and a ceiling of $50K. If you wish, you can specify a separate cash allocation within other accounts (for *holding* cash). I'd suggest reading up about the cash account in the manual.
I'm not sure if there's a better way, but what's worked for me for use of the Cash account is:
- using a created "Money market" as the asset class for actual cash within my various accounts,
- reserving the (required) "Cash" asset class for the "clearinghouse" function, with its returns set at 0, and a low ceiling.
This way the money needed for debits ends up coming from my actual accounts and any credits over the ceiling gets reinvested back into them.
e) I can manually order my liquidations in much more elaborate ways than allowed in Financial Assets->Management->Regular Savings Withdrawal Controls. I can leverage LTCG, STCG, dividends, interest, capital losses, various tax exempt bonds. Pralana does not give me sufficent fine grained control over liquidations to test strategies manually against Roth Conversions. Nor does Pralana run simultaneous optimizations between liquidation strategies and Roth Conversions. Is this correct?
I don't think any of the products do this kind of optimization, even the ones professional advisors use like rightcapital and emoney. There is no substitute for excel sheets IMO for short term optimization. The closest I've come within Pralana is to force certain types of specific withdrawals (using "scheduled withdrawal"), and then compared the two scenarios with consumption smoothing (as a means of comparing net outcomes). For instance, I'll force some scheduled withdrawals in certain years from taxable that I know will be at the 0% capital gains rate as scenario 1, and compare to the recommended roth conversions scenario 2. That kind of thing.
I made a feature request that Pralana build in a tax-gain harvesting optimizer similar to the conversion optimizer. But even that wouldn't really be able to mix and match the two approaches, which is usually optimal.
Having said all that, I do encourage you to explore iterating the Withdrawal Priorities Optimizer (order of taxable/tax-deferred/tax free) with the Roth Optimizer. Going back and forth, two to four times, between the two has more power than you might expect. The results of that process will often guide you toward more nuanced explorations you can do on your own.
2. Once Pralana optimizes for Year N, Iteration Y, the algorithm will never revisit year N until Iteration Y+1. Is this a correct understanding?
Correct.
3. If the preceding is true, there is no way to know if a decision for Year N, Iteration Y precluded better subsquent outcomes. Is this a correct understanding?
An early conversion always beats a later one. This is both because when you pay taxes from taxable, the early conversion reduces your tax drag and as long as you assume that growth exceeds inflation, then getting money moved out of tax deferred early is more powerful than doing so later later as it stops the exponential growth of tax deferred that eventually pushes up tax brackets or IRMAA tiers.
I recently discovered that Financial Assets->Asset Class Taxation is the proverbial tail wagging the Roth Conversion dog in my case. Seemingly minor tweaks can cause the Roth Conversion amounts to change by 2x, 3x, 4x.
Please post more about your results, lots of folks have used the Roth Optimizer without problems. Note that because it is checking the tops of the ordinary income brackets only, there are times where the best answers are other breakpoints in the tax code. Some common ones are available for you to select in a manual search, such as the phase-in of LTCG taxes, IRMAA tiers and ACA FPL. There are other phase-ins that are possible breakpoints too that are harder to find such as the start of NIIT, the OBBBA senior standard deduction phase-out, SS benefit taxation, and at the high end, the OBBBA SALT deduction phase-out, AMT, the standard deduction phase-out. My understanding is that because many of those are based on AGI (actually various MAGI terms), Pralana doesn't inherently know the order relative to ordinary tax brackets, so it doesn't try to search all those options. If you care to explore them, you can set a fixed conversion amount
What is surprising to me is that just checking ordinary income brackets is usually very close to the optimum. I sometimes find that when a tax bracket is halfway between IRMAA tiers that it's better to back off to that lower IRMAA tier.
When Pralana compares the regular investment account balance with other balances, is it using an effective balance or nominal balance for the investment account? How is this effective balance defined or computed over time?
By "regular invesment account" you mean taxable? There is no adjustment made to taxable. Perhaps you are confusing terminology with the "Effective Tax Rate" that Pralana allows you to set for tax deferred accounts (traditional IRAs and inherited traditional IRAs) to try to correct for the embedded taxes that your heirs will pay when those accounts are liquidated. If you don't make some kind of correction for the embedded tax liability, then you would be comparing apples and oranges and get spurious answers for Roth Conversion studies.
Do I understand correctly that Pralana forces the regular investment account to liquidate and buy assets to maintain the fixed stock/bond ratio? I infer this because of Capital Gains that are reported with no Investment withdrawals. If so, when is this balancing done? There is a toggle for account growth at end of year or middle. Is this also when rebalancing is done?
In Mode 2, you set the priority for which accounts should contain stocks vs. bonds. So the rebalancing to keep the overall portfolio on track generally only occurs in one account in any given year. If you have so many bonds that they have spilled into taxable, then the sale of stocks to rebalance will occur in taxable. Effectively, I I understand that it is rebalanced at the end of the year so you get start of year balance + growth based on initial balance + income - expenses to get a subtotal that is then rebalanced to get ready for the start of the next year. Cash and the HSA are not included in the rebalancing. The timing within a year of rebalancing is not very important to the result as annual rebalancing means the deviation from your target allocation is never large.
In our case, I've found it to be best to set the priority for stocks as taxable first, then Roth, then traditional IRA, then Inherited Traditional IRAs. That is the most tax efficient and for us, it avoids creating taxable events due to rebalancing
When doing Roth optimizations for a given year, is Pralana looking at the balance before or after such forced adjustments, if such adjustments occur?
I believe the forced adjustments you are referring to is just rebalancing, that does not change the balances, just the allocation.
I still grapple with what is meant by the cash account, what its portfolio is, etc.
Cash account is a checking or Money Market account that would not be expected to vary in value with other investments. Returns would typically be small. Pralana uses this to determine if you need to sell something to have cash to live on. The floor and ceiling ideas mimic what folks do in real life. For instance, my wife panics if we don't have tens of thousands of $ between checking and Money Market, so that's our floor. 😀
Similarly, if you have a big expense coming up, you may not want to invest everything, so may let cash build up, that would be your ceiling.
I can manually order my liquidations in much more elaborate ways than allowed in Financial Assets->Management->Regular Savings Withdrawal Controls. I can leverage LTCG, STCG, dividends, interest, capital losses, various tax exempt bonds. Pralana does not give me sufficent fine grained control over liquidations to test strategies manually against Roth Conversions. Nor does Pralana run simultaneous optimizations between liquidation strategies and Roth Conversions. Is this correct?
Yes, real life can be infinitely complicated. On the specifics, Pralana lets you reinvest dividends, interest and distributed capital gains, or more practically, it allows you to not reinvest them, in which case they go to cash, the program looks at the year's cash balance, your cash ceiling and floor and then reinvests whatever is left, just like you would do in real life.
On LTCG, yes, I wish that Pralana allowed more granularity to allow it to sell assets with the least gains instead of one big lump. In our case, I've determined that we will only ever need to sell a portion of taxable, so I only enter the gain percentage associated with that fraction. Let's say we might only ever need to sell 1/4 of taxable before RMDs kick in. So I ordered investments from least to most gains and drew a line at the 1/4 with the least gains and use the average gain in that lowest 1/4 as the input to Pralana.
On losses, you should definitely harvest any losses, Pralana allows you to enter your carryover losses and uses them the per the tax code. It doesn't try to harvest future losses.
Not sure what you mean by liquidation strategies. If you mean Spending Strategies, you select the one you want to use and then run the Roth Optimizer. The only one that requires iteration between the withdrawal method and Roth Optimizer is Consumption Smoothing. That requires that you iterate a couple of times between the Consumption Smoothing optimizer and the Roth Optimizer in order to have a solution that incorporates both. While that's not ideal, most tools don't have Consumption Smoothing as a withdrawal strategy at all. The only competitor with it that comes to mind does something wrong and gives bad Roth Conversion answers as a result.
If you mean Withdrawal Priorities, where it finds the account order to draw from first in different phases of life, I set Roth Conversions first and had assumed that using taxable was always preferred, but the program found it was very slightly better to use the t-IRA first after retirement. I then repeated the Roth Optimizer and it found the same plan as before, so my plan didn't see any change after iterating between the two optimizers.
An early conversion always beats a later one.
I think this is generally but not universally true. Sometimes it's better to reduce/postpone an earlier conversion to take advantage of being in a very low (e.g., 0%) federal and/or LTCG tax bracket. Pralana's optimizer does this under certain conditions.
