Roth Estimator

I’ve written a lot about Roth conversion especially partial Roth conversion. I’m in the middle of Roth conversion which is why its on my mind. I was playing with a simple FV calc to look at conversion over time. This is a non-spreadsheet method of analysis that is pretty good.

My previous analysis says that something like a 500K TIRA is OK to own when RMD happens. It’s payout is small enough to keep you in the 12% bracket for a long time. So lets’ say you’re age 65, the SECURE act passes, have a 1.5M TIRA and 7 year to convert it (post SECURE act RMD won’t start till age 72). You want to clean out the TIRA such that 500K is left in the IRA after 7 years of conversion. Do you just divide 1M by 7? That would be the crudest estimate but wouldn’t get the job done. The TIRA will continue to grow over the course of the 7 years (presumably) so you will have to transfer more than 1/7 per year (142,000/yr). Lets look at a FV calculation

This calculation says at 4% return (return above inflation) you would want to convert 185K/yr not 142K /yr. Taxes on 185K would be \$36,538 and \$ 25,106 on 142K MFJ standard deduction one spouse over 65. So now you know how much to transfer and what it’s going to cost. What’s the end value of the Roth?

So at the end of conversion you would have 1.461M in the Roth (nearly as much as you had in the TIRA, and 512K left in the TIRA for a nearly 2M total. Your taxes would be 7 x 36,538 or \$255,766 or a cost of conversion of .13 cents on the dollar, a pretty good deal IMHO. Well below 22% or 24% marginal costs.

I pay my taxes not from the Roth conversion but from cash I free up from my brokerage account mixed with long term cap loss, so my taxes money comes out tax free. A good reason to consider learning how to tax loss harvest.

The results may not be perfect since they represent averages but a rational estimate of both conversion amount and cost. As you convert you can adjust the rate of conversion if the market happens o hit a home run or has a crash. My goal is to wind up with 500K in the TIRA and more than 1M in the Roth prior to RMD. Simple quick no muss no fuss, no Monte Carlo.

This analysis presumes you are living on cash for the conversion period for max conversion efficiency, but even if you’re living on side gig income or dividends this is how to do the analysis, you would just have higher taxes if not living on cash, but you would be paying taxes anyway.

A word on how assets come out of the TIRA. My goal is to store my blonds in my TIRA. Your best placement of bonds is in a pretax or Roth account. My asset transfer is to get the highest return asset out first and into the Roth. Second highest next. The idea is to get the growth into the Roth to avoid paying more taxes on the appreciation. Moving growth first may change the calculus slightly but it’s a trivial matter to adjust every year now that you have a method to judge. Simply place the assets on the efficient frontier plane and read off which one pays the most. Alternatively you can move the riskiest first but depends on your goal. My goal is to have the bonds in the IRA and enough stock to own a “tangent portfolio” which is the AA which pays the most return for the least risk. If you owned BND and VTI the AA of the tangent would be 12% VTI and 88% BND. You may want higher growth but higher growth brings more taxes and moving out of 12% bracket sooner since this account RMD’s, so I’ll get my growth in the Roth which grows tax free. You get the tangent portfolio from the efficient frontier curve of the assets

The really good news is \$185K/yr avoids all the tax cliffs and surcharges built into the tax code. If you convert to the top of the 24% you go off the cliff and pay more taxes than need be, but that’s a subject for another discussion.

My brokerage (FIDO) allows transfer of assets whole or pratial between Roth and TIRA so I don’t even need to convert to money, just transfer and pay the taxes.

6 Replies to “Roth Estimator”

1. GasFIRE says:

Nice discussion on Roth conversion but the devil is definitely going to be in the details. I suspect a lot of the maximum pretax savers are going to run out of room and time if they are interested in significant Roth conversions. I haven’t made a pretax retirement contribution in 7 years since I went part time but bull market account growth has been significant. I started my conversions last year but it was like pissing in the wind as my account balance is still higher than when I started. Most of my bond position had been in munis in the taxable but I now see the value of having your bond position in the pretax to slow the growth there rather than in taxable or Roth. As I increase my bond portfolio percentage, it’s by selling pretax equity positions. Since both my wife and I are still working there is only a modest amount of space in the 24% bracket for conversions. As a one time boost, I am considering pairing a significantly larger Roth conversion amount with opening a DAF worth 5-8 years of my typical charitable contributions. Last year was the first time I hadn’t itemized my tax return since I entered practice, so this seems like a reasonable way to reclaim a deduction at least for one year. As always I like your detailed analysis and the more global view of the entire accumulation/retirement spending process.

1. mdonfire says:

Unfortunately it is like pissing in the wind, and the more aggressive you get in converting the more it costs you to convert. It is a true study in optimization and I have 100 spreadsheets to prove it! I had to optimize each dimension since my skills aren’t good enough to write multivariate sheets and trust the outcome to not be garbage and I wasn’t up to speed on spread sheet proof of performance for what amounts to a single calculation, but I did learn a ton about the tax code and the traps set to separate you and your money. Despite the rhetoric about “fair share” the tax code is designed to soak the rich, but not the super rich. It levels out for the super rich to 43 cents on the buck but I’m sure there are ways around that like the DFA. The psychology involved understanding my retirement accounts did not belong to me but to US, me and Uncle Sam and he meant to collect his part as aggressively as he could during my dotage and then double his efforts to fleece my wife once I passed. I don’t look at the government as my enemy in this because I entered into the pre-tax bargain willingly. But it is what it is and I decided it was worth the cost to extricate myself from the bargain in order to regain some tax flexibility, so my perspective is this is just the cost of doing business. IMHO to do this effectively you have to reach a bed rock reality about who owns what and what power can be wielded to get it and the power is considerable. It does not pay to be stupid or pretend it’s something different than what it is. Once you get to that place is when a plan can be made. The government is cash starved and came up with the Roth conversion as a means to buy you out early with a deal that benefits both over the decades. The rub is you have to live long enough to realize the benefit. My analysis is if I’m dead I don’t care anyway and my only concern was I don’t screw my wife in the mean time. Roth conversion will pay off for her as well in the long run so for me it was a green light decision once I gamed out the consequence. Roth also had some legacy advantages but those will probably go away with the SECURE act. The SECURE act will have a time period during which your heirs MUST empty the accounts and pay those taxes. My goal is to cover me and my wife and I’m less plussed about covering my kids. If things go as planned I can transfer some to them while I’m still alive. If not there still should be millions to divvy up with Uncle Sam after my demise.

A strong impetus for me to retire was to do Roth conversions. I liked my job well enough but I knew if I didn’t make a plan to clean out the TIRA I’d be on an ever increasing tax treadmill post RMD and I wanted to avoid that if I could. I made a plan in my 50’s to contribute heavily to my brokerage while I had the money to pay the taxes. I stopped contributing to pretax all together and I’ve been tax loss harvesting since my 20’s so I had a big wad of tax loss harvested and documented. I still contributed a matching amount to a 401K when we sold our practice out to corporate medicine and they offered a matching program and I also invested is a HSA with them. Hard to pass up free money. When I turned 65, I worked enough to make 1 more and final year toward SS eligibility, went on medicare and started Roth conversions. SS won’t happen till age 70. I cashed in some stock so I had cash to live on plus cash for taxes during Roth conversion which makes all of my conversion equivalent to ordinary income but that also is the limit of my ordinary income so my tax bill is optimized. It wasn’t hap hazard but completely planned. I also have the TLH so I could turn the stock into cash tax free, much like using my brokerage as a Roth from a tax perspective. Worked like a charm.

When this SECURE act came up I rejiggered my cash to extend my conversion another 2 years allowing me to convert a little less over a longer time. My original time frame was 5 years, now its 7 years total and I’m 2 years into it. The smaller yearly conversion (185K vs 250K) saves quite a bit in taxes (about 100k) and I still end up where I want to be when the final trigger is pulled. So the key to this plan was retiring and NOT working to 70 giving me time to pull the levers and minimize the tax burden. You may not have TLH for example but you can still optimize, save up some cash for when you do pull the trigger. My wife had a little business but I asked her to retire too since I needed close control on expenses and ordinary income.

2. Interesting discussion and I didn’t even think about throwing in future value to figure out what yearly amount needs to be taken out (I would have done the simplistic divide by # of years)

Is there a reason you are hoping to leave \$500k in the TIRA instead of a complete withdrawal?

What is the downside of trying to empty it out completely?

1. mdonfire says:

I did a protracted analysis of RMD and various TIRA sizes and tracked the tax consequences. I also read an article that the average (or maybe median) TIRA of those with IRA’s is 500-600K, and I looked at SS added to various RMD disbursements for different portfolio amounts with a constant AA. My conclusion is the government is satisfied leaving the SS + 500K RMD retiree alone and not screw them with soak the rich. This is why there is a 12% bracket, a whole bucket load of voters live in the 12% bracket. Once you hit 22% things change dramatically. You pay a higher marginal rate and a higher cap gains rate (15% vs 0%) once you hit 22%. The next leg, 24% is where the tax cliffs and surcharges occur so 22% is he beginning of soak the rich and past 22% is the beginning of really soaking the rich with surcharges and elimination of special savings for pretax IRA contributions etc. If you make 165K you can take advantage of some little tax write-offs like education and Roth contribution, 300K you’re hosed. My final year I made like 165K and for the first time ever I had things like college credits and the ability to fund a Roth happen to me, so I got a little insight into what it’s like to be middle class from a tax perspective so I studied that for loopholes and optimization.

So my analysis is best make your ordinary income look like 500K + SS. If your TIRA is not super growth oriented it will throw of a little bit say 20K to 25K/yr for a long time, my calculation for a couple decades, keeping you in the 12% bracket for a long time with its 0% cap gains rate. If the TIRA is aggressively risked for growth it will knock you into 22% in jut a few years and you will pay more taxes but not really benefit much from the RMD because a lot goes to higher taxes. I want my bonds in a pretax vehicle like a TIRA anyway so that gives me slow predictable growth and a couple thousand a month monthly income beyond SS. I make up the difference in my cash flow by selling stock from the brokerage at 0% cap gain. If you look at SS + TIRA as a fixed income, I need only add 50k or 60K from my brokerage to even out my cash flow. My brokerage WR is small enough such that I can do that forever and still get growth over all.

I can risk the Roth however I like and it can hold more bonds or not since the tax consequence is none. I choose the tangent portfolio in the 500K TIRA because it’s entirely predictable and pretty impervious to market crash 12/88 could loose half its stock value and sill be 6/88 hardly noticeable to the RMD. If it drops I can re-balance anyway and regain he loss on the way back up. The tangent is the most efficient portfolio so it pays me best for the least risk. This 500K essentially becomes a quasi inflation adjusted annuity since it throws off a mildly increasing amount of money each year because of the progressive nature of the RMD schedule. So that’s how I got to 500K

The downside to emptying it completely is complete conversion is going to cost you a lot more in taxes which pushes your break even point out quite a bit and you may be dead before you break even. These are end of life optimization strategies which help assure you run out of breath before you run out of money, even in old age when you and the old lady might be living in a nursing home. Optimized folks get the room with a view. Also Roth money is your best money since it’s outside the tax loop so it’s better to not spend it and spend it last if it needs spending, giving it a longer time to grow. At RMD my analysis is TIRA > Brokerage > Roth when it comes to desirability for spend down. In other words spend the TIRA first then the brokerage to fill in the gaps then the Roth when all else fails. Ordinary income can carry a 22-24% marginal rate and Brokerage usually only 0 or 15% Roth tax free. Homey likes tax free so he spend that one last and spend that other trash first

3. The comments are as valuable as the post, and both are wonderful. Would also have pulled the hare-brained move of not factoring interest into the Roth conversion ladder, so this is a helpful reminder to adjust accordingly.

Like you and GasFIRE, we have an extremely pre-tax heavy portfolio that will absolutely require Roth conversion, but we both enjoy what we are doing now sufficiently to not be ready to decide precisely when we plan to stop working.

At age 46 I have some runway left to make those plans, and at this point the goal is to 1) avoid touching the pre-tax egg to allow it to grow, and 2) build up taxable to save for both the tax hit on the conversions and the years of bridge living expenses should we pull the trigger a bit earlier.

Dream du jour is to spend some time in the years immediately after the empty nest (a decade from now) traveling and living out a few dreams in various places. That might be the time to live off of savings, Roth convert and have those no-earnings years occur.

Then again, dreams and circumstances change so the planning may be for naught…

1. mdonfire says:

There was a David Graham guest post over on Big ERN’s site regarding the likelihood of a 4% withdrawal lasting 60 years. A really thorough job by David pointing out the pitfalls. Some of the graphs show some lumpiness as withdrawals occur and time passes. The lumps happen because of the brokerage account running out of money and switching to a TIRA for income. The income goes from cap gains dependent to ordinary income dependent when it comes to taxes. Okey Dokey if you’re in the 12% but oops if you’re in 22% or 24%. What this points out is retirement funding is not homogeneous but heterogeneous. 4 x 25 implies a kind of homogeneous portfolio and withdrawal. If you RMD by definition the WR on the TIRA is variable. David’s answer includes the heterogeneous nature of ACTUAL retirement and not the 4th grade math that often poses as “retirement planning”. It’s why I think hiring some kind of manager is important. A well trained manager can help you force your goals to become a reality, instead of an incomplete hodgepodge of fantasy. In reality to retire you need to understand total cost, tax burden, length of retirement and likelihood of success and a means to break it down for EACH YEAR. This is my epoch model. Several years in an epoch may look alike but each year needs to be accounted for, and each year if you are dead or alive because that affects how your bride will live in your absence. Each epoch needs to interface with the subsequent epoch and take into account secondary effects on other epochs.

My adviser created exactly this kind of analysis customized and granularized using as accurate of number as we could muster using a program he is alpha testing and involved in creating. I can look at any year based on my age to 93 and review the projected poop for that year, projected taxes etc. It takes into account my eventual death and how that will affect my wife, her income and tax drag. Knowledge is power. It “what if’s” various scenarios like what if I Roth convert at the 12% level vs the 24% level. I had already done these spreadsheets but my numbers and the program numbers were almost identical. Knowing what I know convinces me I’ll never run out of money and neither will my wife. The efficient structure of my portfolio including tax planning also convinces me of this. The country “could” collapse and we switch from using cash to trading bullets for currency but if we get to that level of depravity I’ll save a bullet for myself.

If you want to travel and I hope you do you need a granular plan like this and the means to fund it, but it’s completely do-able. The DIYer unless quite astute will have a hard time acquiring this level of granularity. There is just too much to know. I spent 2 years devising my plan with some input every day sometimes many hours of study. I could afford to spend the time because it was exceedingly interesting and I was retired AND I had done adequate planning decades before. My plan was already in the fine tuning stage the day I quit.

So that’s what is missed by the boilerplate approach. Nobody is preaching this instead they are preaching side gigs which are W2 substitutes as a method to reduce perceived risk, but side gigs in themselves are very risky. They are preaching portfolio leverage as the means to the end but as demonstrated leverage (owning EM for example) can be very expensive if things go badly and most people wouldn’t know things are bad before it was too late. Real Estate is the same. It presumes a stable economic environment. There is a reason REITs carry a 19% risk almost as much as EM.

Were I you I’d give David Graham a call and see what he can do for you to make your dreams come true. I am very impressed with this guy and his acumen. Worst that can happen is you spend a few bucks for nothing. You’ll make it up in gas money saved riding your bike and likely far more than that in streamlining and tax efficiency in the portfolio. I don’t know if he does tax loss harvesting but a bonus if he does. He clearly has the software necessary to do some pretty high level analysis and he’s a fixed fee for service provider not an AUM. you wouldn’t hire a Neurologist to do heart surgery much less a plumber so I can’t see this romance with DIY with something as important as retirement. To me there is a big disconnect in reliable knowledge and often the wrong thing gets focused on as if it matters (like PhD diversity or shaving 2 bp off fund cost).

Thanks for the opportunity to write the counter point I think it’s a fun and interesting exercise