Up to PRC2022 the tool showed that Roth conversion is always beneficial for me. The more I convert the more I will save by EOLSV (end-of-life savings value). The reason was that my Roth allocation was 100% equity, and the IRA was 60/40 Equity/Fix. Moving money from IRA to Roth also meant increasing my overall equity %. Then PRC2022 introduced mode 2, an overall allocation that spans all accounts for stocks and bonds. Mode 2 keeps the desire Equity/Fix ratio during Roth conversion. To my surprise Roth conversion stopped to be beneficial in mode 2.
However, I am not convinced that Roth conversion in my case is a bad idea. It depends on the goal.
PRC optimization goal is maximizing EOLSV. The tool is trying to maximize total saving for all accounts (Taxable, IRA and Roth). I assume that many PRC users will not deplete their accounts by end of life and their goal become to maximize these accounts for their heirs. Let’s take EOLSV example: $1M in taxable, $4M in IRA and $1M in Roth, the total is $6M. It is very different from $1M-in taxable, $1M in IRA and $4M in Roth which is also $6M total. PRC assumes that a Roth dollar is equal to an IRA dollar (and equal to taxable $) and simply add these 3 savings dollars and try maximizing the total number. Paying taxes on inherited IRA can be a significant burden on the heirs especially if you need to deplete the account within 10 years of the inheritance. Roth inheritance also require depletion within 10 years but does not require to pay taxes. Assuming the marginal tax rate of the heirs is 25%, then they will pay at least $1M in taxes on the $4M inherited in IRA.
It is possible that the conversion itself will require paying higher tax (rate of 28% or higher). But because Roth continue compounding and growing tax free, the overall benefit (original owner + heirs) of conversion may be large.
It will be nice if the tool be able to calculate this and optimize Roth conversion (and other optimizations) not just based on simple sum of EOLSV accounts but something like EOL+10SV. Can the tool have a programmable goal that the user can configure?
You were mis-specifying the problem prior to 2022. As we learned in science class, you need to isolate variables to learn what causes what. Instead, you mixed a far more powerful factor, the 40% change in asset allocation, in with a smaller factor - the several percent change in value when you do tax arbitrage with a Roth Conversion. You swamped the effect of the Roth Conversion with the asset allocation change.
To do a fair comparison in Mode 1, you should have used the same asset allocation in all accounts (using an average if they are actually different).
Note that Mode 2, where you prioritize where to hold stocks vs. bonds, will always show an improvement in the final savings when you use a constant rate of return, but if you examine a bear market, like the historical sequence starting in 1965, you will find it makes you worse off than if you had held everything at equal allocation in all accounts.
In fact, preferentially stuffing bonds into your t-IRA is really just a sneaky shift in asset allocation towards stocks. That is, the government owns a share of your IRA (based on your marginal tax rate when you withdraw), so stuffing bonds in there forces the government to hold more bonds, so you get to hold more stocks.
I did an experiment and for our situation, an 80/20 asset allocation using Mode 2 gives the same result as Mode 1 at 87.2/12.8 (the equivalent allocations will be different for everyone depending on tax bracket, longevity assumptions, size of accounts, etc. but by continuity we know one exists). To find your equivalent numbers between Mode 1 and Mode 2, just set an allocation in Mode 2, get the final savings value, switch to Mode 1 and guess at an allocation, using the same guess in all accounts and then look at the final savings value and make a new guess. Or do the reverse and start with Mode 1 and then guess and check in Mode 2.
What was interesting was that the same allocation for Mode 1 at 87.2/12.8 also gave very similar answers to Mode 2 at 80/20 in historical sequences starting in 1965 (a bad bear) and 1982 (a big bull). The differences between the two modes seem explainable based on how full of bonds my IRA happened to be when the bear growled or the bull ran. Another way of looking at it is that with Mode 2, your effective asset allocation keeps changing to greater and greater stock percentages as you fill your IRA with bonds and that's not what I intended to do.
While I was one of the people badgering Stuart to include Mode 2, I now realize that it doesn't provide a free lunch, it's just the same risk/reward as greater stock allocation.
I don’t want to derail the discussion from the title of this chat (Is maximizing EOLSV the goal for everyone?). Your analysis is nice, but the ways to manipulate mode 1 to behave like mode 2 are not part of the requested new feature. Nonetheless thank you for the ideas.
Some thoughts on optimizing for heirs:
The finances for your heirs, including incomes, expenses, taxes, SS, ACA, IRMAA, etc. are just as complex as yours, seems like it would require complete inputs of their situation before Stuart ever started thinking about optimization.
An alternative you can do yourself is to set up a case for each heir and manually put the values at inheritance into the heirs' respective taxable, inherited Roth and inherited IRA accounts, along with your estimates of their income and other account balances at your passing and run it that way. Tedious, but might be worth doing once to get a feel for their brackets.
I've also tried the assumption that their tax rate will look like ours, then extended the lifespan of the longest lived spouse 10 years and used the Scheduled Withdrawal Table in those final 10 years to drain the IRA.
Before I found Pralana, I used a homegrown spreadsheet and would have a copy open for our finances and for each of the kids and had formulas linking the sheets to pass the inheritance to them (I'm never tried it with Pralana, but it is so massive that I'm pretty sure it would choke if you had multiple copies open). Aside from my homegrown sheet being buggy, slow and hard to maintain, I eventually realized I was piling on guesses for finances I have no real insight to and no control over and projecting their income and expenses for a couple decades in the future. So I gave up, got Pralana and set a guess for the Effective Tax Rate for Conversion of Absolute Dollars to Effective Dollars on the Financial Assets - Management tab.
Thank you, Rick. You are right; it is more complex than I thought. I just believe that it is a situation of many Pralana users and looking for an easy way to calculate the benefit.
Getting back to the name of the thread, this topic was discussed at some length under Chat - Rethinking Roth Conversions in a Down Market. The relevant discussion starts around April 21, 2022 (6:51 pm). Some of the comments I made include (others made comments too of course):
For me Roth conversions are about short to mid term decisions. Max out the 12% tax bracket before it goes up to 15%, big drops in the stock market, reducing RMDs, reducing social security taxes, etc. is what I look at when considering Roth conversions.
For me, like I said before, I look at the tax savings from reducing RMD’s, SS taxes, the ability of taking money out of a Roth if needed without being pushed into a high tax bracket, as being reasons for doing Roth conversions.
My latest comment was under Chat - Study Turns Roth Conversion on its Head - Mostly on March 30, 2023, 4:47 pm if you want to take a look. It was about using my Roth account to greatly reduce my MAGI to get a large ACA tax break.
Pizza Man, I agree with your ideas. What I would like to see is that the tool will provide numbers to calculate the conversions in the situations that you mentioned and if you want to leave money for heirs.
EOL+10SV is End-Of-Life + 10 Years [heirs Inherited IRA/Roth transfer to Taxable account] Value.
It is complex to fully calculate since it requires the full heirs’ status. However, maybe there is a way to provide partial or minimum heirs’ information to receive some realistic results for Roth conversion in case of Heirs. Here are two options:
- Assume 10 yearly transfers from IRA/Roth to Taxable account starting at EOL. The heirs marginal tax rate is the same as the original owner, and their investment allocation/location is also the same. The goal is to maximize the value of the remaining taxable account after 10 years of EOL.
- To value Taxable, IRA and Roth accounts differently by providing configurable multiplier for each of them by EOL.
@pizzaman this is a good line of thinking, from my perspective. I agree that with a lot of these software planning tools the assumption seems to be solving for the maximum EOLSV. This is why you typically end up with a big percentage of the value in Roth at the end of the projection. However THIS IS NOT THE GOAL FOR EVERYONE. Some people have no heirs, or they don't prioritize leaving money for heirs. Maybe they just want to maximize their spend down while optimizing the taxes they will pay doing so. I think more discussion could be had on this point, the original question posed. I'd say "not everyone".
For me, like I said before, I look at the tax savings from reducing RMD’s, SS taxes, the ability of taking money out of a Roth if needed without being pushed into a high tax bracket, as being reasons for doing Roth conversions.
@dbradp I agree, we need more discussion on this topic. I started a thread under "Wish List" entitled "ROTH Conversions and Future Tax Brackets" where I talk about another reason to do Roth conversions now - tax rates going up in the future, possible by a lot ????. Check it out ????
If you don't have close heirs that you want to plan for, then it could be OK to enter zero for the Effective Tax Rate to use on residual IRA balances. However, if that's what you are doing, retirement researchers tell you to use a very long lifespan. If you optimize for age 100 and pass at age 80, you won't regret it because you're gone, but if you are one of the few that makes it to a great age and you didn't optimize for that, then you have regrets that you didn't plan for that possibility.
We include a guesstimate of our heir's tax rates in the Effective Tax Rate, so it's more reasonable to use something closer to a mortality table for the first to pass and a combined mortality table for the last to pass. If you die early, then Roth Conversions don't get much time to compound gains - some conversions that make sense if live to 100 years old may not make sense if you live the to the average age or less.
RE: Is maximizing EOLSV the goal for everyone?
No. Maximizing our Enjoyment Outta Life is our goal.
Read or listen to Die With Zero. Even for those with heirs and charities, maximizing EOLSV should not be the goal.
No, I don't desire to maximize EOLSV. I desire to maximize discretionary spending along the way.
Pralana tool is about the way to reach your financial goal. “Die with Zero” is about defining this goal. These are two different topics. Pralana is a high-level calculator and “Die with Zero” is a behavior philosophy.
