Roth conversions

^^^^^ I would have no problem with tamping down the stretch IRA... the original intent was to provide tax benefit to the worker/saver... not to their heirs.... and a $450k exemption is pretty generous. Hardly a "money grab".

If I were king I would make it $250k exemption and the rest over up to 10 years.
I would think it most equitable to make this and other changes in such a way that folks don't get penalized for making decisions based on the old rules. So, say, the ability to pass along unlimited IRA holdings and proceeds from them that exist on the date the legislation passes would remain in place, new monies could be passed along under the new, more restrictive rules. Or, at very least, give people a significant window to adjust to the new rules. One person's "money grab" is another's "fair and reasonable adjustment," but making the rules "prospective" rather than affecting dough that has already been committed in a certain way under the old rules has merit, IMO. Nobody truly knows what the "intent" of the government was when the original law was passed, every legislator surely had a slightly different intent. All we can do is look at the law and the implementing rules and try to work with them.
 
Last edited:
ignore at will, but I suspect you might agree with the simple concept that the more you spend, the less is left to compound.
It is hardly incomprehensible or "of no substance".
The amount of X paid today in annual tax is much more important than the same amount paid some 15 years later. It kind of hurts today's portfolio while it would barely move the needle later on.

samclem gets it (see post #71). He has drawn a line on how much to convert, as higher annual conversions (and tax paid on these) reduce the utility of the current portfolio to a uncomfortable level, while in later years (after starting the RMDs) the tax rate may be higher, but the overall utility of the portfolio is lessened (due to its increased size and retiree's lower needs) and the tolerance to spending a bit more in income tax at that stage is much more relaxed.
This utility (or value as I called it earlier) of keeping more money in the portfolio is much more pronounced during the first few years in retirement, as those years carry the highest exposure to the sequence of returns risk.
samclem's decision is not based merely on the tax rate arbitrage criteria, but also on factors like portfolio size and chances of survivability (in the military sense of remaining "mission capable")
So is mine.
Quantifying this though is a bear, and I admit I'm having hard time with the details.
 
^^^^^ I would have no problem with tamping down the stretch IRA... the original intent was to provide tax benefit to the worker/saver... not to their heirs.... and a $450k exemption is pretty generous. Hardly a "money grab".

If I were king I would make it $250k exemption and the rest over up to 10 years.


Well....ummmm...of course you have no problem with it because most likely it does not affect you or your family. Just like I have no problem if welfare benefits are cut since I don't collect welfare ( but I do pay for them). We are all concerned primarily with those financial scenarios that most affect our own situation.


And I disagree. A $450,000 exemption is paltry if one has millions in an IRA. Should be more.
 
Actually, it may affect me and my family depending on how long I live.

I doubt that many here expected the tax deferral benefit to extent beyond the grave to future generations... that is why they called it an individual retirement account. While I expected that tax deferral benefits would extend to my spouse, I didn't expect it ti extnd to my kids or grandchildren.

Pigs get fat, hogs get slaughtered. Don't be a hog.
 
ignore at will, but I suspect you might agree with the simple concept that the more you spend, the less is left to compound.
It is hardly incomprehensible or "of no substance".
The amount of X paid today in annual tax is much more important than the same amount paid some 15 years later. It kind of hurts today's portfolio while it would barely move the needle later on.

samclem gets it (see post #71). He has drawn a line on how much to convert, as higher annual conversions (and tax paid on these) reduce the utility of the current portfolio to a uncomfortable level, while in later years (after starting the RMDs) the tax rate may be higher, but the overall utility of the portfolio is lessened (due to its increased size and retiree's lower needs) and the tolerance to spending a bit more in income tax at that stage is much more relaxed.
This utility (or value as I called it earlier) of keeping more money in the portfolio is much more pronounced during the first few years in retirement, as those years carry the highest exposure to the sequence of returns risk.
samclem's decision is not based merely on the tax rate arbitrage criteria, but also on factors like portfolio size and chances of survivability (in the military sense of remaining "mission capable")
So is mine.
Quantifying this though is a bear, and I admit I'm having hard time with the details.
I have a question for you.
Would you say conceptually that if one believes they will at least be in the 22% future bracket, then one should fill at least the 12% bucket now?
I guess assigning a ROR to the taxes saved by not converting now vs. the 10% tax differential brackets in the future would be difficult to quantify?
Of course, this assumes that tax rates remain the same (can't imagine them going lower in the future).
 
Of course, this assumes that tax rates remain the same (can't imagine them going lower in the future).
I would have said that, too, and probably did. Yet, they are going down for most people in 2018. So, ya never know.
 
Quantifying this though is a bear, and I admit I'm having hard time with the details.

But until you do you should not be forceful in defending your position. People here look for data for a constructive dialogue.

I recommend you look at I-ORP and use the extended analysis feature to optimize withdrawal and spending in retirement. You can play around with the tool to determine if Roth conversions would optimize your situation.
 
ignore at will, but I suspect you might agree with the simple concept that the more you spend, the less is left to compound.
It is hardly incomprehensible or "of no substance".
The amount of X paid today in annual tax is much more important than the same amount paid some 15 years later. It kind of hurts today's portfolio while it would barely move the needle later on. ....

The example that I provided earlier included the impact of compounding with the example where the tax is in a taxable account... if the tax is paid from taxable funds, then the tax compounds at a pre-tax return since it is in the Roth... if the tax isn't paid (no conversion is done) then the tax compounds at an after-tax return.... and the former is better than the latter... I even quantified the impact. See posts 45 and 51 for the details.

.... Quantifying this though is a bear, and I admit I'm having hard time with the details.

Then why should we believe your assertion that it is better not to convert?
 
Last edited:
I recommend you look at I-ORP and use the extended analysis feature to optimize withdrawal and spending in retirement. You can play around with the tool to determine if Roth conversions would optimize your situation.
I urge folks to think really carefully about what we are really seeking when we say "best outcome."

I-ORP or a SS breakeven calculator can show us which course of action will likely to produce the highest average monthly payout/portfolio value, and people latch on to that as the answer. It sounds like the answer. But consider this example:
I am offered the chance to take up to my entire portfolio and risk it on one a flip of a coin. If I guess right, I'll be paid 3:1. If I guess wrong, I lose my money. How much should I bet?

Going strictly by the numbers, I should bet every penny I have (and borrow more if possible!). There's a 1:2 chance I'll win, and I'll be paid 3:1. It's a fantastic offer, free money.

But in the real world, I would probably bet very little. Yes, it is a fantastic opportunity. But if I bet my whole stash and lose it, the negative results are terrible: depending on handouts for the rest of my life. If I win, the positive results are very nice (tripled my money), but that extra dough doesn't have nearly the utility I am risking (my initial stash, which was enough to keep me comfortable in my retirement).

The first dollars in one's portfolio (or monthly check) have a lot higher marginal utility than the last dollars.

So, when considering whether to pre-pay taxes now vs. waiting to pay them later, I try to consider this. It's not a simple math problem. If I'm in the 12% bracket now (MFJ taxable income of $19K-$77K) and I wind up in the 22% bracket (MFJ taxable income of $77K-$165K) in retirement, it probably means my annual take (i.e. monthly income) is a LOT higher than my income today. Lower marginal utility of those last (much more numerous) dollars is fairly low, and paying taxes at that point should sting fairly little. OTOH, if my investments do poorly, or I have to deplete my portfolio for unavoidable spending (sick child, spouse, etc), and I find that our portfolio is really in trouble, I may very much regret voluntarily reducing my portfolio in years past to give the IRS a bunch of money. That money could prove really useful to me (i..e each dollar in my portfolio and in my monthly income has very high utility to me) , but I don't have it anymore.

Now, I'm sure that a smart economist/accountant has already figured out how to quantify these things, and I'd love to see that spreadsheet/calculator.

Apologies for the wordy response.
 
Last edited:
...

So, when considering whether to pre-pay taxes now vs. waiting to pay them later, I try to consider this. It's not a simple math problem. If I'm in the 12% bracket now (MFJ taxable income of $19K-$77K) and I wind up in the 22% bracket (MFJ taxable income of $77K-$165K) in retirement, it probably means my annual take (i.e. monthly income) is a LOT higher than my income today. Lower marginal utility of those last (much more numerous) dollars is fairly low, and paying taxes at that point should sting fairly little. OTOH,



if my investments do poorly,
Sure, the assumption is that over time your investments will make money. I can't see making a long-term plan that doesn't count on this. If you want to do some quasi-market timing, do more conversions when the market is low, so the recovery gains are never taxed in the Roth, and do fewer or no conversions at a market high.

or I have to deplete my portfolio for unavoidable spending (sick child, spouse, etc), and I find that our portfolio is really in trouble, I may very much regret voluntarily reducing my portfolio in years past to give the IRS a bunch of money. That money could prove really useful to me (i..e each dollar in my portfolio and in my monthly income has very high utility to me) , but I don't have it anymore.
If you've satisfied the 5 year rule you can withdraw that Roth money your converted. It may not be ideal, but this is an exception case, right? If these are foreseen expenses, you shouldn't deplete your taxable account too much to pay taxes on conversions. How much is too much is case-by-case.
 
Sure, the assumption is that over time your investments will make money. I can't see making a long-term plan that doesn't count on this.
Yes, I do expect that my investments will grow, at least a little better than inflation. If not, I wouldn't ER. But I'm not optimizing everything for the most probable outcome. I recognize there's a fair-to-good chance that we'll hit a rough spot, one of those lines in the FIRECalc run that dip down toward the X-axis on a worrisome slope. If I'm actually riding that line (unlike looking at the FIRECalc graph), I don't know how the story ends.
At that point, I'll appreciate having that extra gas in the tank (from not paying a lot of extra taxes early) on that retirement road of undetermined length.

If you've satisfied the 5 year rule you can withdraw that Roth money your converted. It may not be ideal, but this is an exception case, right?
Right, I'll have access to the Roth money. But I won't have access to the money I gave to the IRS years ago when I was trying to optimize for the "most probable" case.

In our particular case: If our investments do great and we are in the 22% income bracket in retirement, I'll hardly miss the extra taxes we'll pay at that time (and will be happy to pay them, I promise :)). Our income will still, after taxes, be higher than it has ever been. But if our investments do poorly with our portfolio headed down and we are in the 12% or 10% bracket (or lower), I will very much miss any taxes I voluntarily pre-paid when the grass was greener (and will be very sorry I paid them, I promise). At that point, not only will we be in a lower bracket but, much more important, each buck will have higher utility (i.e. will be way more dear to us) than the dollars I gave the IRS years ago when we were flush.

But, yes, I do plan to Roth convert, probably up to the top of the 12% bracket, but not beyond that.
 
Last edited:
But, yes, I do plan to Roth convert, probably up to the top of the 12% bracket, but not beyond that.[/QUOTE]

Okay so all said and done, that is where my current thinking is to go to the top of the 12% bracket. We do have most of our monies in TIRA, but my MAGI needs to managed for ACA. Not sure it will work after 65, even if SS is delayed.
DGF is currently collecting SSDI, so can't do too much there, but some is better than none.
 
Then why should we believe your assertion that it is better not to convert?

I've made no such assertion. You inferred something that simply is not there.

The only assertion I've made is that when one thinks it is better to convert, one should triple-check one's premises, and give the matter a bit more (calculated) thought.

As to believing me, I don't really care one way or another. As I have clearly stated, I post here for my own amusement. It is up to the readership to get influenced by my posts if they seem to have merit - or not to.
 
But until you do you should not be forceful in defending your position. People here look for data for a constructive dialogue.

I recommend you look at I-ORP and use the extended analysis feature to optimize withdrawal and spending in retirement. You can play around with the tool to determine if Roth conversions would optimize your situation.

How am I forceful?
I did defend my position because I believe I am correct. There was nothing forceful about it though.

As to I-ORP, it did provide me with some food for thought a few years ago, but it is of no use to me today as my own spreadsheet-based analysis is way past I-ORP's capabilities.
I-ORP completely ignores the aspect preached in this thread by samclem, to which aspect I also (forcefully :LOL:) alluded in my posts.
 
I have a question for you.
Would you say conceptually that if one believes they will at least be in the 22% future bracket, then one should fill at least the 12% bucket now?
I guess assigning a ROR to the taxes saved by not converting now vs. the 10% tax differential brackets in the future would be difficult to quantify?
Of course, this assumes that tax rates remain the same (can't imagine them going lower in the future).

You are trying to reduce a complex picture to a simple Yes-No answer.
It does not work that way, try as you might.

I can say this: if one's portfolio is roughly more than 35x one's needed annual portfolio withdrawals, and one's AA is within the 40% to 70% range of equities, then I believe one should convert to the top of their 12% tax bracket.
However, a marginal portfolio (25-30x), or one with a silly AA (too high or too low equity allocation), would call for a more careful analysis.

Mine being of the marginal sort, I think I need to be clever about too much $ spent on taxes in my first few years of retirement.
 
You are trying to reduce a complex picture to a simple Yes-No answer.
It does not work that way, try as you might.

I can say this: if one's portfolio is roughly more than 35x one's needed annual portfolio withdrawals, and one's AA is within the 40% to 70% range of equities, then I believe one should convert to the top of their 12% tax bracket.
However, a marginal portfolio (25-30x), or one with a silly AA (too high or too low equity allocation), would call for a more careful analysis.

Mine being of the marginal sort, I think I need to be clever about too much $ spent on taxes in my first few years of retirement.

Okay. Is your marginal portfolio% net of income i.e. net expenses(WR) or gross expenses divided by portfolio
 
Now, I'm sure that a smart economist/accountant has already figured out how to quantify these things, and I'd love to see that spreadsheet/calculator.

Conceptually, this tool would be simple:

1) Build a curve/table/formula that provides a multiplier to "weight" each monthly/annual retirement dollar. In the case of DW and I, the first dollars (used to keep the electric bill paid , food on the table, etc) would get a big multiplier. Each dollar above, say,$10K per month would have a much lower weight (we'd use this money to travel business class rather than coach. Nice, but each dollar buys much less comfort than using the same $$ to keep the heat on in the winter).
2) Run FIRECalc to give a bunch of runs describing (past) investment performance of our AA. Using the weights in Step 1, convert each FIRECalc run line to a "weighted utility of income produced" graph (over time)

3) Feed all of those FIRECalc results through I-ORP using a range of different IRA-> Roth conversion strategies, see which produces the greatest area under the curve.(TIME vs WEIGHTED UTILITY OF AVAILABLE INCOME STREAM). That'll tell us which Roth conversion strategy would have the highest expected future utility (if we assume future investment returns look like past ones).


It would be a lot of work. So, we're left with TLAR.
 
I think the worry about pre-paying taxes substantially increasing the risk of ruin is misplaced. In the last 5 years I've converted about 1/4 million and paid ~7.5% in taxes or around $19k... and saved an estimated $36k (assuming I avoid paying 22% on those withdrawals).

So even if I double that it is $40k of taxes prepaid. IMO, if a $40k expenditure significantly increases the risk of ruin, then you probably didn't have enough to retire to begin with.

So for me, the savings are so significant the any second order effects are minute by comparison and of no concern. Now OTOH, if there were no tax savings, I'm not sure that I would bother with conversions.
 
Last edited:
But, yes, I do plan to Roth convert, probably up to the top of the 12% bracket, but not beyond that.

We have been converting to the top of the 15% and will be converting now to the top of the 12% even though I-ORP recommends converting to the top of the 22% bracket for maximum optimization. So I was not insinuating that people should blindly follow any optimization or planning calculators but to use them as tools to fit their situation.
 
Last edited:
Except that many people here are facing the "tax torpedo" when RMDs start, so taxes will be paid on the IRA money eventually. Roth can reduce that tax torpedo since it has no RMDs.
+1. I have almost 20yrs until R.M.D and based on even modest growth and income assumptions I'm going to pay the piper many pounds of flesh. That said, at this point each year I crank the numbers and it doesn't make sense.
 
We try to do a Roth conversion every year to bring our total taxable income up to our total deductions. We have no pensions and are living solely off our taxable investments. Our true cost is around 10% because of reduced ACA subsidies and cost sharing incentives, but we are fine with that. OTOH, we will probably never have to pay taxes on the RMDs because the conversions will probably reduce our total IRA balances to a low enough total.

We are converting all of our stock funds first. We still have 4+ more years before we will have to start converting bond funds. At that point, we will have to decide if we want to start the glide path to a higher stock allocation, or start buying bonds in the Roths.
 
Last edited:
Sorry for my ignorance. I am new on this site and am not familiar with lots of terms used in this thread. What does "filling 12%" mean? And , MFJ?
 
Last edited:
Sorry for my ignorance. I am new on this site and am not familiar with lots of terms used in this thread. What does "filling 12%" mean? And , MFJ?

It means converting funds that would be taxable up to the max of 12% bracket. You would have to figure out what your taxable income is and then figure out how much more you could rollover over up to the 12% limit, which would be more advantageous then rolling at the next, higher, bracket.
 
Another important thing to know is the interaction of ordinary income and preferenced income and the impact on taxes.... if your taxable income is within the 12% bracket then qualified income (qualified dividends and long-term capital gains) is taxed at 0%.... if your taxable income is over the 12% bracket then qualified income is taxed at 15% or more.

As a result, for some income that is just above the top of the 12% bracket the incremental tax rate can be 27% and most of us are not keen on paying 27% on those incremental Roth conversions.
 
Back
Top Bottom