{Stuart Matthews here...I'm adding this topic on behalf of another (unknown) user because, unfortunately, I experienced a website failure and had to restore from a backup copy and lost that user's post (luckily I received an email containing the same question and have posted it here for all to see).}
I understand due to the American Rescue Plan Act (ARPA) that the ACA “subsidy cliff" has been eliminated for 2021-2022, with a maximum 8.5% of income payable for premium when Modified AGI (MAGI) exceeds the cliff threshold of 400% Federal Poverty Level. I am curious what priority PRC is placing on maintaining ANY amount of subsidy in these two years, particularly with regard to Roth conversion optimization.
In my situation, I am insured under the ACA and qualify for the premium subsidy throughout the ACA period. I want to do Roth conversions until age 70 whenever advantageous. With constant (inflation-adjusted) income and expenses throughout the ACA period, PRC is calculating ginormous Roth conversions for 2021-2022 then going back to modest conversions for the remaining period years, keeping my MAGI exactly at the cliff as expected. In Tabular Projections, I see “ACA Subsidy Amount” as blank for 2021-2022 and “Final Margin to the ACA Cliff” of huge negative values (the projected Roth conversions). Then for 2023 and beyond I see the expected modest values for “ACA Subsidy Amount” and “Final Margin to the ACA Cliff” values of blanks, the Roth conversions doing their work right up to the cliff.
I’ve been wrestling in my own mind how to approach these two years of cliff reprieve but forgoing 100% of ALL subsidy and Roth converting to the moon was not something I had been considering. I was thinking I’d just pay a little extra ACA premium if I miscalculated my MAGI and exceeded the cliff by a few dollars but not intentionally go substantially above it. It would seem that PRC is otherwise placing an extremely high value on maintaining the ACA subsidy for as long as possible since it does not give us the option of allowing MAGI to exceed the cliff when projecting Roth conversions. If it’s normally such a bad idea to exceed the cliff and lose the subsidy, why is PRC seemingly forgoing all subsidy dollars in 2021-2022 by allowing such a stratospheric MAGI (when it could be at least limiting healthcare expenses to 8.5% of a less-than-stratospheric MAGI and holding onto some subsidy)? The subsidy post-cliff still reduces to zero as MAGI continues to increase, no?
When I purchased PRC I was actually eager to test out whether it might be wise to forgo the ACA subsidies entirely, pay 100% of the premium, and instead focus on optimizing Roth conversions to the max in the pre-Medicare years. But when I discovered that PRC placed such a high value on the ACA subsidy dollars that it didn’t even allow the option of Roth converting above the cliff, I concluded further investigation was likely unmerited. Now, with PRC making no attempt to limit MAGI in 2021-2022, I am wondering whether there may indeed be some valid tradeoffs between ACA subsidy vs. Roth conversion prioritization.
I am not necessarily questioning PRC nor its approach for 2021-2022. I have done a good bit of reading on the ARPA provisions but still do not fully understand them; the technicalities are confusing to me. I am likely misunderstanding some fundamental implications of the cliff elimination for these two years and am admittedly feeling a little fuzzy about it all. I’m so accustomed to tracking my MAGI to the penny to avoid that cliff that I haven’t yet been able to rejigger my thinking! Does this temporary mitigation of the ACA tax - normally exceedingly high when falling over the cliff - truly offer a rare opportunity for unthinkable (for me) AGI or should there still be some limit imposed to salvage at least some subsidy? I would just appreciate, if you could, a little explanation and education on the thought process behind PRC's ACA implementation for 2021-2022. Also, in Boglehead-style if you are familiar with that site, welcomed is enlightenment/strategy/insights from anyone else on this Forum who understands the ARPA implications better than I do and may have considered how best to capitalize on this opportunity.
Thanks all.
To the statement "I am curious what priority PRC is placing on maintaining ANY amount of subsidy in these two years, particularly with regard to Roth conversion optimization", I can say that PRC gives no priority to maintaining subsidies in 2021 and 2022 when trying to optimize Roth conversions. While the algorithm to optimize Roth conversions is quite complex, it lacks the complexity to simultaneously optimize ACA subsidies. My suggestion would be to run the Roth optimization with a start year of 2021 and then adjust the starting year to 2023 (to prevent the Roth conversions from driving your AGI so high that you lose the ACA subsidies) and then see what long-term effect it has. I just set up and ran such a test and that approach did yield a slightly better long term result.
To the statement "If it's normally such a bad idea to exceed the cliff and lose the subsidy, why is PRC seemingly foregoing all subsidies in 2021-2022 by allowing such stratospheric MAGI (when it could be at least limiting healthcare expenses to 8.5% of less-than-stratospheric MAGI and holding onto some subsidy)", I'll just echo what I said above: PRC's algorithms simply don't contain the complexity to simultaneously optimize Roth conversions and ACA subsidies. With that said, I believe your thought process is very sound and you can explore that as I suggested above. I think there may very well be some valid trades between ACA subsidies and Roth conversion optimization.
With that said, I'll leave it for other users to comment further. Thanks for the thought-provoking post!!
@smatthews51 Stuart, I'm a brand new user and coming up to speed on the PRC model. Roth conversion optimization with consideration of ACA and IRMAA AGI limits was one particular feature of interest to me, and motivated me to purchase your product. As far as I can tell at the moment, PRC currently only considers the 400% FPL cliff in the optimization process. Would it be possible to include a drop down selection for the ACA limit with the other breakpoints of 150%, 200%, and 250% and no limit similar to the IRMAA selection? (Or maybe a user defined %FPL so that the user could enter 900% FPL or any other value for 2022.) The ACA premium credit and out of pocket maximums associated with this selection/input percentage figures into the projected healthcare expenses (i.e, like the 8.5% of MAGI mentioned above), and could be entered manually for each option, particularly if there is a circular reference issue in the optimization. It may be obvious that 400% FPL is the optimal planning value post-2022 in terms of overall impact to net worth projections, and any difference between the breakpoints is probably small since ACA only comes into play for a relatively few number of years in most cases. I had in my mind that 200% FPL might be the sweet-spot, but I was hoping PRC could convince me of this or inform me of a better strategy. I think the optimal selection really depends on one's estimate of out of pocket expenses relative to the actuarial value limits/OOP maximums. I believe the user could evaluate the trade space between the ACA subsidies and Roth Conversions with the additional drop-down menu, which I think addresses the topic you discuss above. If there is already a way to modify the 400% FPL limit in the ACA subsidy calculations, please let me know. Thank you.
Digging into the topic a little more confirmed to me that the medical expenses included in the tabular results do account for the ACA subsidies at the various FPL breakpoints. However, it does not appear that the Out of Pocket (OOP) limits applicable to each FPL related tier, or the 8.5% MAGI limit, are included in the calculations. There are two OOP limits that depend not only on the AGI relative to the FPL breakpoints, but also on the number of dependents, namely, an individual OOP limit and a family OOP limit. The user would need to specify which set of maximum OOP limits (individual or family) they wish to consider in the analysis, or alternatively, allocate a percentage of the annual OOP expenses to a single individual. When I included a large ($50,000) annual OOP expense in the healthcare expenses table, it looked like all of the $50K (plus the ACA Silver Plan cost less subsidy) flowed through to the Projected Healthcare Expenses at each FPL breakpoint for the ACA years. I expected something less than the 400% FPL subsidy to be optimum in this case with large OOP expenses, but since it appeared there were no OOP limits in play (or the MAGI limit), the 400% FPL endpoint remained the optimal selection. I understand it may be too complex to simultaneously optimize ACA in the Roth conversion algorithm (I read where it's thought it would be difficult to invent a more complex financial differential equation), but I believe including the OOP limits and an option to select the FPL breakpoint would be a beneficial feature for the ACA analysis.
@gbmcbryde Gordon, there's currently no way to have PRC consider any of the other FPL's in the Roth optimization process; however, that feature is being built into the 2022 model which I expect to release in early to mid January.
@gbmcbryde Gordon, the PRC algorithms deal only with ACA premiums and the associated subsidies. The determination of these amounts definitely does take into account family size and the 8.5% limit. The out-of-pocket amount is a separate line item on the Expenses > Healthcare page and is not in any way limited by PRC. With that said, the Roth conversion optimization algorithm takes family size and MAGI into account and always limits MAGI to 400% FPL. The new algorithm being built into the 2022 model will allow you to specify FPL limit during Roth conversions, with options of 1.5x, 2x, 2.5x, 3x and 4x. Do you think that'll work for you?
Stuart: Yes, being able to specify the FPL limit should work fine. I appreciate your reply and look forward to the 2022 model.
I had another question related to this analysis and Sequence of Returns risk. Would it be possible to include a fourth "scenario" in the Tabular Projections, namely, the Bear Market Analysis results? I would be interested to look at the details of the cash flow from the Bear Market Analysis in the Tabular Projections pages using the Asset Class returns selected on the Bear Market Analysis page. I figured you might have that stored on a sheet and accessible to load into the results tables. I believe it could be useful to look at the output flows from a Bear Market case, or any other particular sequence of returns if this capability were available. Thank you for your efforts to keep improving PRC's already great capabilities.
@gbmcbryde Gordon, hmmm, that's an interesting idea! I'll give that some thought. One issue I can think of right up front is that the analysis is actually associated with one of your three scenarios with only the sequence of returns changed to the historical sequence. So, this "fourth" scenario would actually be one of the other three with a modified sequence of returns. My plate is pretty much full for the 2022 model, at least the initial release, but I'll definitely keep this idea on the list for further consideration. Thanks!
Thank you Stuart. I don't believe there is any "issue" with this capability; that is, your description of the results presented from one of the historical sequences is exactly what I believe would be beneficial to examine. The modified, or selected, sequence of returns would allow you to look at how far draw downs in your cash or investment accounts from a bad starting sequence, let's say, get to zero and force an unplanned withdrawal from a tax deferred account or any number of other unintended/unexpected consequences. It might be a little challenging to explain the functionality in the manual that the user needs to carefully pick his/her scenario (1,2, or 3) and then the sequence of returns of interest on the Bear Market Page, but overall I believe it is pretty intuitive to follow the logic. Thank you.
