iORP and Roth Conversion questions

[newellcr]

It’s been a while since I engaged on the forums. I’ve been lurking more seriously over the last two years and have been interested in iORP for the IRAtoROTH ‘optimization’. I only put that in scare quotes because highest level of spending isn’t necessarily optimizing depending on the user. ACA isn’t an issue. I’m out of the workforce and DW is out in June.

Thanks,

Chris

My questions/discussion items.

At one time I thought I saw that IORP was a paid service. I ran a dozen scenarios yesterday and today. There also seemed to be fewer options (required income vs just disposable income is one that comes to mind) than what I recall reading about. Is there a fee service with additional inputs or controls? In general, I’m happy with the output, but I’m an ER number crunching geek.

The IRAtoROTH numbers optimize to the top of the 24% bracket and that makes sense to me for our situation. We’ve already converted over $400k in the last 6 years, so the shock value of six figure conversions isn’t there for me. It looks like iORP converges on Roth conversions being complete by age 63 for us (current ages in low 50’s) and even well before 63 if we can really stomach, or find it favorable, to follow iORP to the top of the 24% bracket. I assume this convergence at 63 is for the IRMAA medicare bump. Have folks been comfortable with the way iORP optimizes the IRAtoROTH?

The Monte Carlo numbers are similar to our Vanguard plan and are similar to FIREcalc. The 3-PEAT numbers for discretionary (total, in iORP’s case) spending seem inflated by about 25% which I attributed to flat application of interest rate for stocks and bonds. Does this seem right? I readily admit that I haven’t chipped through to what the 3-PEAT is, but I’m not sure it’s necessary.

My zero IRAtoROTH conversion and to the top of the 24% bracket ROTH conversion runs are only 2.5% different in discretionary spending (range of $190k/yr). This seemed hard for me to believe. If it is accurate, and it probably is, it lends some credence to the comments about ‘Roth Conversions don’t matter as much as you might think’. It also supports leaving additional amounts in the tIRA for late life healthcare costs and/or charitable giving. We do have long term care insurance. Not quite sure how to ask the question, so how’s this – It doesn’t seem like a bad idea to leave $100 – 200k in the tIRA at age 60, ignore it in the AA, set the one fund timeframe to 30yrs to retirement, and let it grow as a separate insurance policy/goodie fund. Is this a bad idea? Are there other uses for late in life tIRA funds?

 

[SevenUp]

AFAIK I-ORP is and always has been free.

I-ORP's tax calculations are decent if one has a Goldilocks income: not too high, not too low.

If you compare your tax calculations to I-ORP's and they match well, I-ORP's answers are probably applicable. Caveat user if your tax situation doesn't mesh well with I-ORP's assumptions.

 

[sengsational]

Good question about not driving tIRA assets 'to zero' through conversions. I'm surprised i-orp would do that, given it could do just enough converting to make sure RMD's were pulling at a tax rate the same or lower than the conversions. But I think there's a tweak in there to presume tax rates will be higher in the future.

 

[RunningBum]

Good question about not driving tIRA assets 'to zero' through conversions. I'm surprised i-orp would do that, given it could do just enough converting to make sure RMD's were pulling at a tax rate the same or lower than the conversions. But I think there's a tweak in there to presume tax rates will be higher in the future.

Future pension and social security income might be enough to warrant fully converting.

 

[newellcr]

Thanks, SevenUp. 2019 tax year data will help a lot based on some rental carry over losses. Nothing is easy for me (that's a blessing). I really do like the iORP interface that allows multiple one time life events.

 

[newellcr]

Sengsational and Runningbum. Thanks for the replies. Here is the reduction of the TaxDef for IRAtoRoth that is identical for the no limit and top of 24% bracket. All of the yr 1 to 6 reductions go 100% to IRAtoRoth conversions.

940 yr 1
730 yr 2
505 yr 3
263 yr 4
169 yr 5
72 yr 6

It then accounts for one time capital gains for several years which disrupts IRAtoROth progression but forces the TaxDef balance to zero by age 62. I mentally questioned if it was forcing zero at 62 or if it was a coincidence, so I checked the iORP run for conversions to the top of the 22% bracket. The taxDef balance for 22% at 63 is 110 and 64 is 12. It dabbles along and is zero by 71. The last IRAtoRoth is done at age 62. So iORP does seem to want to zero the TaxDef by age 62 or 63.

His and her pension and some residual rental profits continue/combine to make for a nice income stream before retirement savings pad that number.

My plan (prior to running iORP) for IRAtoRoth numbers are in the $120-150k until the one time capital gains come into play then the conversions will drop to $40-60k per year.

As I looked at the downward progression of TaxDef money on my own plan, it makes me question taking the balance to zero.

 

[atmsmshr]

iORP is a nice and free software, but I think it has a few limitations for my specific situation in LTCG conversion.

My preliminary optimization outcome is to convert 40% of my IRA to Roth. That gives me a 17% tax rate during Roth conversions, 4% effective tax rate during LTCG conversion, and 8% effective rate until the bitter end, meaning I will not be in any current tax rate above 15% after LTCG harvesting.

Setting aside IRA money to fill the remaining MFJ tax bracket less than $80,000, and over that will be Roth.

Some extra IRA money for health shock deductible medical expenses (some LTC may remain deductible). Even with this amount, RMD should not hurt if DW has to file single upon my shuffling off this mortal coil.

My Suspected Limitations of iORP

iORP does not model the connection between your capital gains tax rate and your personal income tax to compute your capital gains tax rate! Proves the point to understand the model and tax code; however you can artificially add a tax rate.

Additionally, iORP harvests LTCGs at the same time of Roth conversions, which is tax inefficient – these activities should be sequential not concurrent.

The IRA is also spent early with no reserves for later (like standard deductions before claiming SS, filling in lower tax bracket, medical deductions).

Plus side: iORP does have the utility of ‘is your spreadsheet in the ballpark?’, it shows the time value of money, and SS scenarios.

atom

 

[newellcr]

atom,

Thanks for the reply. I do like the iORP software. An apples to apples comparison to the spreadsheet will be easier for 2021 and even better in 2022. The income streams have been in a state of flux for a number of years. DW has a salary and starts a pension this year. Last year we sold a rental and purchased another. Getting the 2019 and 2020 taxes prepared will help with identifying carry over losses and predicting a steady state.

It will be easier to fit our finances into iORP by 2022 and the current plan is for $10 to 12k monthly conversions on a schedule. iORP has given me some additional intestinal fortitude to crank and extra 'payment' or two in during a down turn.

I do like that iORP has different AA for different buckets. Our Roths are 100% stock and the TaxDef have what constitutes a 65/35 mix overall. Every monthly installment brings TaxDef closer to all bonds. I suspect that iORP moves the IRAtoRoth piece to Roth and will allocate it as 100% stock, but keeps TaxDef stock/bond AA as is - so the tIRA AA is overweight in stock compared to reality and gets worse with every monthly payment to Roth.

My spreadsheet shows that, with a calm market, all stock will be in Roth around the 33 month range. Things will probably settle out to steady state by Jan 2023. The AA of 100% stock doesn't bother me. I have a 'reserve' inherited IRA in the lowish 6 figures in 100% bonds. I chip away at that to help with the conversion taxes, by the time that account is low, there should be a similar amount of bonds in the Roth with a HUGE contribution/conversion basis.

Now what to do with leaving some in the TaxDef bucket. It really isn't a problem that I have to deal with until 2025. Then again the market might hold a really nasty surprise/opportunity.

I also still keep going back to the results show only a 2.5% spending difference between 100% conversion of TaxDef to Roth and 0% conversion to roth. The 0% conversion would be so easy, but I watched Dad's problems stemming from the change from MFJ to single with the RMDs.

Ugh, thanks for the input everyone.

 

[atmsmshr]

Good point about the impact of an inherited IRA. My parents are still living, and I have not wanted to make any assumptions about potential inheritance.

 

[newellcr]

atom,

Dad should still be alive. That reminds me that both parents had the longevity gene in their families and I bumped up the iORP default to age 95. Life throws good and bad at you in varying sized lumps.

I don't know if having a stash in the tIRA will really matter in the end. It's funny, I thought having a 30 yr plan to ER was complicated. Ha!

Thanks again for the comments. It is really helpful.

 

[newellcr]

Noodling on these iORP runs made me realize that the real impact of no conversions to Roth vs the up to 24% conversion run is not a 2.5% difference. The real difference is only based on the tIRA and Roth pieces. The LTCG and fixed income pieces remain the same. The difference is more like 5 to 6% in favor of converting up to the 24% bracket. That makes more sense to me...

 

[Montecfo]

Noodling on these iORP runs made me realize that the real impact of no conversions to Roth vs the up to 24% conversion run is not a 2.5% difference. The real difference is only based on the tIRA and Roth pieces. The LTCG and fixed income pieces remain the same. The difference is more like 5 to 6% in favor of converting up to the 24% bracket. That makes more sense to me...

How you mind explaining that in a bit more detail? Specifically, what is the 2.5% and the 5-6% you are referring to?

Thanks

 

[newellcr]

Montecfo,

Absolutely. At the top of the iORP output is the 'Projected, maximum, annual Discretionary Spending' line. With the program set to the Zero Conversion to Roth iORP outputs 189k for discretionary spending. With the program set to unlimited or to the top of the 24% bracket for Roth conversions, iORP outputs $194k. The difference is 2.5%, or thereabouts. Having thought about it. As I understand it the top number, call it $190k, is the flat spending (inflation adjusted) possible with the input as given. This includes fixed income, future one time events, part time work, etc. So the tIRA and Roth piece for me really reflects all of the 2.5% difference. My fixed income, future one time events, part time work, etc - these do not vary from my multiple iORP runs. So the difference is based only on the balances of the tIRA and Roth pieces. tIRA and Roth balances should be a bit more than 50% of our iORP output for Discretionary Spending. That's how I get to the rough figure of 5-6%.


Hope that helps.

 

[SevenUp]

I do like that iORP has different AA for different buckets.

Careful with that. A known "issue" with I-ORP is that if you assign higher returns to stocks than to bonds, and have different asset allocations in different accounts, I-ORP will "realize" that the account with the higher stock percentage is "better."

That will likely be more "valuable" to I-ORP than tax considerations. Caveat user.

 

[newellcr]

SevenUp, Thanks. I've been having an issue with getting consistent runs, not sure if it was based on the AA only, not just today, but yesterday as well. There are a lot of parameters to keep an eye on and at one point my save/cookies were partly missing data. I am getting frustrated with re-running the numbers, but the AA per bucket does seem to matter with my iORP runs not only for the Discretionary Spending numbers, but also for the optimal conversion amounts.

I am sitting here trying to convince myself it doesn't matter (ie don't get frustrated) much over the short run. Both AA runs tell me to convert up to the top of the 24% bracket for at least 2 years, then do significant conversions for a few more years to take the tIRA to either zero or to $260k by 63. BTW, the 60/40 AA shows less difference from zero Roth conversions to the top of the 24% bracket. 60/40 gives $1k/yr difference at $178k.

I wasn't going to follow iORP to the top of the 24% bracket anyway but was going to stick with my spreadsheet and convert into the 24% bracket and feel my way along this process.

Again, thank you for the input.