Stop Passing Up This Great Deal - In Money Magazine

[Brantley]

Just read article titled "stop passing up this great deal" by Penelope Wang in Money magazine regarding Roth 401(k)s. The article is relying on a recent study by T. Rowe price compiled by Stuart Ritter. I have a few question for you folks if you have also read this article.

1) Can anyone find the source research? I looked on the Troweprice website, and under Ritter's profile with no luck. Also ran a google search on the quotes used, and checked out trowe's twitter posts and didn't see it. I also saw this research was references in articles written in WSJ, kiplinger, and CNN.

2) Some items in this article seem incorrect or misstated such as,

"Roth 401ks leave just about all workers, regardless of age or tax bracket, with more money to spend in retirement than pretax plans do"

"The Roth 401k should be considered the default investment"

"Every dollar you save in a Roth 401k is worth more than a dollar you put in a pretax account. That's because you'll eventually pay income taxes on those pretax dollars, while you get to keep every Penny in a Roth"

I'm assuming this article is relying on the fact that you can put more money in a Roth vs Traditional(which depending on tax brackets I'm not even sure that Roth would win). Is there something I'm missing here or is this article complete garbage?

[dodecahedron]

Very knowledgeable tax writer, David Cay Johnson, had a twitter exchange with Penny Wang (author of the Money article you cite) that she had made a very flawed assumption that RMDs from traditional retirement funds would reduce SS benefits by 50% to 85%. She has tweeted back that she was wrong and will publish a retraction.

Quoting Johnson's tweet:

Bad error:@Money @PennyWriter sez SS benefits reduced 50%/85% per extra $ income. WRONG. Not publish work email, either

She tweeted in reply:

@DavidCayJ You're absolutely right. My bad. Thanks for pointing it out. We'll run a retraction.

Pretty bad quality control on Money Magazine's part to let such an egregious mistake get out there. Such is financial journalism these days.

[ralph124cf]

It depends on ones current and predicted future tax rates.

It also depends on how much you can afford to save.

Since the limit that you can put in for yourself is $17,500 for either the pre-tax or the Roth, the $17,500 is worth quite a bit more in a Roth account. This assumes that you have enough "extra" money to pay the taxes on that much extra income.

If you expect your future tax rate to be higher than it is now, this is an especially good move.

If you expect your future tax rates to be lower, then pre-tax may be better.

Ralph

[pkcrafter]

Here's a link to a TRP article that contains the quote you show.

http://www.thinkadvisor.com/2014/06/23/ ... -should-to

In the article is a link to a study. Ritter is mentioned in both, but I don't think he did the study, although he may have been involved.

Paul

[Brantley]

It depends on ones current and predicted future tax rates.

It also depends on how much you can afford to save.

Since the limit that you can put in for yourself is $17,500 for either the pre-tax or the Roth, the $17,500 is worth quite a bit more in a Roth account. This assumes that you have enough "extra" money to pay the taxes on that much extra income.

If you expect your future tax rate to be higher than it is now, this is an especially good move.

If you expect your future tax rates to be lower, then pre-tax may be better.

Ralph

Penny seems to be implying that even if you tax rate is lower in retirement, Roth is still likely to come out ahead...

[ResearchMed]

Very knowledgeable tax writer, David Cay Johnson, had a twitter exchange with Penny Wang (author of the Money article you cite) that she had made a very flawed assumption that RMDs from traditional retirement funds would reduce SS benefits by 50% to 85%. She has tweeted back that she was wrong and will publish a retraction.

Quoting Johnson's tweet:

Bad error:@Money @PennyWriter sez SS benefits reduced 50%/85% per extra $ income. WRONG. Not publish work email, either

She tweeted in reply:

@DavidCayJ You're absolutely right. My bad. Thanks for pointing it out. We'll run a retraction.

Pretty bad quality control on Money Magazine's part to let such an egregious mistake get out there. Such is financial journalism these days.

That's a HUGE "error".

How in the world could someone be "qualified" to write for a real financial publication and get this so, so wrong?

Problem is, all too many people will read that (well, whoever does make it through any such article rather than "intending to read it", etc., or starting and getting bored or confused) and RETAIN the specific "50-80% loss" or just "a really huge loss"?
They won't know to check it elsewhere.

There is already too much confusion about the "up to 85% of your SS benefits will be taxed [depending upon total income, etc.]" and "you'll lose 85% of your SS benefits to taxes if you earn 'too much' ".

But it really bothers me that someone like this thinks she/he is qualified to write about such things.
What other mistakes has this writer made, and communicated to others?

RM

[Brantley]

Here's a link to a TRP article that contains the quote you show.

http://www.thinkadvisor.com/2014/06/23/ ... -should-to

In the article is a link to a study. Ritter is mentioned in both, but I don't think he did the study, although he may have been involved.

Paul

Read this line from the study,

"Millennials in particular stand to benefit from the tax-advantages of Roth IRAs, because the longer their contributions have to compound tax-free, the more those contributions could be worth in retirement."

Does the author not understand that TIRA and Roth IRA withdrawals are equivalent assuming same tax rate during accumulation and withdrawal phase? Also, not sure why the article is comparing maxing a Roth IRA vs max TIRA and investing difference in taxable account. In regards to 401k, I would bet that most Americans are not maxing out contributions. Not an equivalent comparison.

[dodecahedron]

It depends on ones current and predicted future tax rates.

It also depends on how much you can afford to save.

Since the limit that you can put in for yourself is $17,500 for either the pre-tax or the Roth, the $17,500 is worth quite a bit more in a Roth account. This assumes that you have enough "extra" money to pay the taxes on that much extra income.

If you expect your future tax rate to be higher than it is now, this is an especially good move.

If you expect your future tax rates to be lower, then pre-tax may be better.

Ralph

Penny seems to be implying that even if you tax rate is lower in retirement, Roth is still likely to come out ahead...

I think her reasoning for that faulty conclusion is based on her flawed assumption that traditional withdrawals will drastically reduce SS benefits. It is true that you need to consider more than just your marginal statutory income tax bracket in retirement in making this call, however. Your effective marginal rate will often be higher than your statutory marginal rate due to indirect effects on taxable SS, Medicare premiums based on AGI, and possibly other effects as well (e.g. qualifying for local property tax exemptions in some places.) On the other hand, in recent year Congress has allowed charitable contributions directly from IRA, reducing AGI as well as taxable income. That provision expired in 2013 but has been renewed several times already and may well be renewed again.

In any event, there is lots of uncertainty about future tax law and my inclination is to diversify risk by having some of both types of funds. Still no excuse for Penny Wang's sloppy statement of Interaction between traditional retirement withdrawals and SS benefits.

[Taylor Larimore]

RM:

Everyone makes mistakes (particularly when it involves the U.S. Tax Code). Penny Wang quickly admitted her mistake and has published a retraction.

I know Ms. Wang. Penny is a very knowledgeable financial writer and a friend of the Bogleheads. This is one of her articles about the Bogleheads published in Money magazine in 2006:

Where You Can Turn for a Helping Hand

Best wishes
Taylor

[Brantley]

Here's a link to a TRP article that contains the quote you show.

http://www.thinkadvisor.com/2014/06/23/ ... -should-to

In the article is a link to a study. Ritter is mentioned in both, but I don't think he did the study, although he may have been involved.

Paul

Isn't this 100% wrong?

“A significant drop in tax rates between when the investor made her IRA contribution and began retirement withdrawals can often be offset by the power of tax-free compounding,” Mr. Ritter says. “But for investors nearing retirement, there isn’t enough time for the money to compound at a rate to counter the significant reduction in their tax bracket during retirement.”

[dodecahedron]

RM:

Everyone makes mistakes (particularly when it involves the U.S. Tax Code). Penny Wang quickly admitted her mistake and has published a retraction.

I know Ms. Wang. Penny is a very knowledgeable financial writer and a friend of the Bogleheads.

I agree that tax code is very complex and anyone can make a mistake, including an award winning veteran financial journalist like Penny Wang. But Money Magazine should have realized that such errors about our complex tax code are so easy to make in advance and should have had a fact checking protocol/editorial review process in place before allowing such egregious misinformation and her extremely sweeping broad-brushed conclusion to be published. Relatively few people may see the retraction unless it is given a great deal of prominence.

Ironically, just a few months ago Penny Wang tweeted about a far more balanced and nuanced discussion of the Roth vs. Traditional question in the WSJ:

http://online.wsj.com/news/articles/SB1 ... reno64-wsj

Some nice links in that discussion.

[dodecahedron]

Here's a link to a TRP article that contains the quote you show.

http://www.thinkadvisor.com/2014/06/23/ ... -should-to

In the article is a link to a study. Ritter is mentioned in both, but I don't think he did the study, although he may have been involved.

Paul

Read this line from the study,

"Millennials in particular stand to benefit from the tax-advantages of Roth IRAs, because the longer their contributions have to compound tax-free, the more those contributions could be worth in retirement."

Does the author not understand that TIRA and Roth IRA withdrawals are equivalent assuming same tax rate during accumulation and withdrawal phase? Also, not sure why the article is comparing maxing a Roth IRA vs max TIRA and investing difference in taxable account. In regards to 401k, I would bet that most Americans are not maxing out contributions. Not an equivalent comparison.

There is a sense in which he is right. With a traditional, you are FORCED to take RMD starting at age 70.5, even if you would prefer to keep the funds invested and let them grow tax deferred even longer. With a Roth IRA, you are allowed to let the money grow inside the tax-advataged wrapper for the rest of your life, if you want. So you can potentially get a longer tax-advantaged compounding period with a Roth IRA. (Roth 401k does NOT directly give you that additional compounding opportunity, of course, because it does have RMD at 70.5, but you can roll into a Roth IRA and indirectly get same benefit under current tax law.)

I do agree with you that most Americans are not maxing their accounts though. Indeed, many are withdrawing what little they contribute far too soon, long before retirement.

[ResearchMed]

RM:

Everyone makes mistakes (particularly when it involves the U.S. Tax Code). Penny Wang quickly admitted her mistake and has published a retraction.

I know Ms. Wang. Penny is a very knowledgeable financial writer and a friend of the Bogleheads. This is one of her articles about the Bogleheads published in Money magazine in 2006:

Where You Can Turn for a Helping Hand

Best wishes
Taylor

Yes, but this is such a common - and significant - misunderstanding that someone writing a column like this should be planning to explain it away, not perpetuate it.
For some people, the difference between the reality and the feared "reductions" could make a big difference in their planning... and fears.

RM

[Brantley]

Here's a link to a TRP article that contains the quote you show.

http://www.thinkadvisor.com/2014/06/23/ ... -should-to

In the article is a link to a study. Ritter is mentioned in both, but I don't think he did the study, although he may have been involved.

Paul

Read this line from the study,

"Millennials in particular stand to benefit from the tax-advantages of Roth IRAs, because the longer their contributions have to compound tax-free, the more those contributions could be worth in retirement."

Does the author not understand that TIRA and Roth IRA withdrawals are equivalent assuming same tax rate during accumulation and withdrawal phase? Also, not sure why the article is comparing maxing a Roth IRA vs max TIRA and investing difference in taxable account. In regards to 401k, I would bet that most Americans are not maxing out contributions. Not an equivalent comparison.

There is a sense in which he is right. With a traditional, you are FORCED to take RMD starting at age 70.5, even if you would prefer to keep the funds invested and let them grow tax deferred even longer. With a Roth IRA, you are allowed to let the money grow inside the tax-advataged wrapper for the rest of your life, if you want. So you can potentially get a longer tax-advantaged compounding period with a Roth IRA. (Roth 401k does NOT directly give you that additional compounding opportunity, of course, because it does have RMD at 70.5, but you can roll into a Roth IRA and indirectly get same benefit under current tax law.)

I do agree with you that most Americans are not maxing their accounts though. Indeed, many are withdrawing what little they contribute far too soon, long before retirement.

Ok gotcha. That makes sense. The author was not very clear in this respect. BUT, does this argument take into consideration conversion to Roth IRA early on in retirement, in which you could likely pay 0-15% tax? That would negative this argument completely.

[dbltrbl]

One area Roth shines is no RMDs at age 70. Big plus for most on this forum. So even if Traditional and 401k Roth come out same, something to consider.

[LadyGeek]

RM:

Everyone makes mistakes (particularly when it involves the U.S. Tax Code). Penny Wang quickly admitted her mistake and has published a retraction.

I know Ms. Wang. Penny is a very knowledgeable financial writer and a friend of the Bogleheads. This is one of her articles about the Bogleheads published in Money magazine in 2006:

Where You Can Turn for a Helping Hand

Best wishes
Taylor

That article is in the wiki: The Bogleheads® ("External links")

Thanks, the original link was broken. It's fixed now.

[nisiprius]

Truth: the tax rate on Social Security is the same as your other income, except that you only have to pay that it on a portion of your Social Security income, not all of it. The formula for what the portion is uses the numbers "50%" and "85%" in it.

Crazy idea: the tax rate on Social Security income is 50% to 85%.

The crazy idea seems to be a "meme on the rise." It's spreading like wildfire.

[LadyGeek]

To expand on Nisiprius' post, it's in the wiki: Taxation of Social Security benefits

[nisiprius]

To expand on Nisiprius' post, it's in the wiki: Taxation of Social Security benefits

GOOD ARTICLE.

Now what someone needs to do is to create an online calculator that lets you enter the total amount of your Social Security income, and produces a CHART with X = total other income and Y = amount left after taxes...

[sambb]

Is a roth 401k really advantageous if one is in the top tax bracket of 39%? Does AMT play a role? Assume a 20 year or longer time horizon

[Taylor Larimore]

Is a roth 401k really advantageous if one is in the top tax bracket of 39%?

The general rule is that a Traditional 401K or Traditional IRA is advantageous for taxpayers in a high income tax bracket--especially if they expect to be in a lower income tax bracket at retirement.

This Morningstar tool may be helpful:

http://screen.morningstar.com/IRA/IRACalculator.html

Best wishes.
Taylor

[denovo]

Really bad article, even if well-intentioned.

[LadyGeek]

If you're referring to the wiki article, suggestions for improvement are always welcome. Post in Suggestions for the Wiki or start a new thread to discuss.

[tainted-meat]

For people not covered by a Traditional Defined Benefit Pension Plan it will be difficult to justify using the Roth 401k over the Traditional 401k in most cases where a person is filing as single or a dual-income married filing jointly situation.

Any blip in employment creates opportunities for converting Traditional dollars to Roth dollars. If you have kids and one parent stays home, it also creates additional opportunities to convert dollars.

If you really end up with "too much" in pre-tax savings, you have the opportunity to retire earlier and convert the monies at insanely low effective tax rates.

Food for thought...

[dodecahedron]

For people not covered by a Traditional Defined Benefit Pension Plan it will be difficult to justify using the Roth 401k over the Traditional 401k in most cases where a person is filing as single or a dual-income married filing jointly situation.

Any blip in employment creates opportunities for converting Traditional dollars to Roth dollars. If you have kids and one parent stays home, it also creates additional opportunities to convert dollars.

If you really end up with "too much" in pre-tax savings, you have the opportunity to retire earlier and convert the monies at insanely low marginal rates.

Food for thought...

Excellent point. Another good midlife conversion opportunity comes if you or spouse decide to take time off from your job to pursue a graduate degree or start up a business or even just move to a different career where the rewards are primarily nonfinancial.

[grabiner]

2) Some items in this article seem incorrect or misstated such as,

"Roth 401ks leave just about all workers, regardless of age or tax bracket, with more money to spend in retirement than pretax plans do"

"The Roth 401k should be considered the default investment"

"Every dollar you save in a Roth 401k is worth more than a dollar you put in a pretax account. That's because you'll eventually pay income taxes on those pretax dollars, while you get to keep every Penny in a Roth"

These are true but misleading. If you save $10,000 in a Roth 401(k), you will be better off than if you save $10,000 in a traditional 401(k), because the dollars are more valuable. However, if you are in a 25% tax bracket, you don't have this choice; if you can save $10,000 in a Roth 401(k), you have the option of saving $13,333 in a traditional 401(k) because you will reduce your taxes by $3333. (And you don't have to wait until next April to get your tax savings, as your employer will reduce your withheld taxes when you increase your 401(k) contribution.)

[denovo]

If you're referring to the wiki article, suggestions for improvement are always welcome. Post in Suggestions for the Wiki or start a new thread to discuss.

No, no, of course not, the money magazine article!

[Onslow]

I've found this link handy in determining my taxable ss based on varying inputs:

http://www.calcxml.com/calculators/how- ... n=#results

[denovo]

2) Some items in this article seem incorrect or misstated such as,

"Roth 401ks leave just about all workers, regardless of age or tax bracket, with more money to spend in retirement than pretax plans do"

"The Roth 401k should be considered the default investment"

"Every dollar you save in a Roth 401k is worth more than a dollar you put in a pretax account. That's because you'll eventually pay income taxes on those pretax dollars, while you get to keep every Penny in a Roth"

These are true but misleading. If you save $10,000 in a Roth 401(k), you will be better off than if you save $10,000 in a traditional 401(k), because the dollars are more valuable. However, if you are in a 25% tax bracket, you don't have this choice; if you can save $10,000 in a Roth 401(k), you have the option of saving $13,333 in a traditional 401(k) because you will reduce your taxes by $3333. (And you don't have to wait until next April to get your tax savings, as your employer will reduce your withheld taxes when you increase your 401(k) contribution.)

I've run into this bad argument before when people discussed the traditional or roth ira's. Grabiner gets it right as usual. I really can't believe a finance magazine would publish this.Where was the editor?

[HenryPorter]

I'd like it cleared up whether a Roth IRA creates any footprint in the IRS's turf at all. From what I have read and expect, all proceeds from a Roth IRA past age 59.5 have zero liabilities. That is, no taxes, no taxes, no....taxes. And no means testing currently, even for SS calcs. Hope this does not stray into tax talk third rail issues, but from what I am planning on, I see the Roth as a source to withdraw funds last in days of retirement if you can juggle your traditional IRA withdrawals and SS payments and possible part-time work income into an efficient ratio tax-wise. In other words, the Roth is a resource to round off income needs if the other revenue sources come up short.

[grabiner]

I'd like it cleared up whether a Roth IRA creates any footprint in the IRS's turf at all.

The IRS knows about it, because the custodian sends you (and the IRS) a Form 5498,

From what I have read and expect, all proceeds from a Roth IRA past age 59.5 have zero liabilities. That is, no taxes, no taxes, no....taxes. And no means testing currently, even for SS calcs.

It is counted as income for some non-tax purposes, such as financial aid eligibility, but I don't think it affects any other tax records.

[FinancialDave]

I found some of Stuart Ritter's "misguided" thinking in his profile on the TROWE site - such as:

If, like most investors, you expect to pay the same tax rate in retirement, a Roth IRA still makes sense in most cases. The longer you have until retirement, the more advantageous a Roth IRA's tax-free growth potential can be

REALLY! This kind of thinking gives CFP's a bad name. Unfortunately, he is not the only one.

[FinancialDave]

2) Some items in this article seem incorrect or misstated such as,

"Roth 401ks leave just about all workers, regardless of age or tax bracket, with more money to spend in retirement than pretax plans do"

"The Roth 401k should be considered the default investment"

"Every dollar you save in a Roth 401k is worth more than a dollar you put in a pretax account. That's because you'll eventually pay income taxes on those pretax dollars, while you get to keep every Penny in a Roth"

These are true but misleading. If you save $10,000 in a Roth 401(k), you will be better off than if you save $10,000 in a traditional 401(k), because the dollars are more valuable. However, if you are in a 25% tax bracket, you don't have this choice; if you can save $10,000 in a Roth 401(k), you have the option of saving $13,333 in a traditional 401(k) because you will reduce your taxes by $3333. (And you don't have to wait until next April to get your tax savings, as your employer will reduce your withheld taxes when you increase your 401(k) contribution.)

I've run into this bad argument before when people discussed the traditional or roth ira's. Grabiner gets it right as usual. I really can't believe a finance magazine would publish this.Where was the editor?

This is spot on -- to compare apples to apples you need to REALIZE the simple fact that you can only compare equal amounts of what I would call Marginal Income and what you should do with it --- In other words if next year you get a $10k raise, the decision between these two options IS - do I put $7500 in the Roth 401(k) or do I put $10,000 in the standard 401(k). You are not going to be able to put $10k in each, from a realistic standpoint, because you will be short $2500 on your tax bill next year (assuming a 25% marginal rate in these examples.)

Once again - on equal tax footing there is ABSOLUTELY no advantage to one over the other - in the area of returns that is. There are of course other things to consider, as we all know.

fd

[ObliviousInvestor]

Is there any way at all to access this article online?

I don't subscribe to any print publications, and my local grocery store still has the July issue of Money for some reason.

I spoke with Penny for this article with regard to how the ACA and taxation of Social Security can make a retiree's marginal tax rate much greater than just their tax bracket (e.g., due to unique taxation of Social Security, a retiree in 15% bracket can have federal marginal rate of 22.5% or 27.75%). I am worried that it somehow came through very wrong.*

*Edited to add: If it did, it's probably my fault due to communicating poorly over the phone.

[FinancialDave]

Is there any way at all to access this article online?

I don't subscribe to any print publications, and my local grocery store still has the July issue of Money for some reason.

I spoke with Penny for this article with regard to how the ACA and taxation of Social Security can make a retiree's marginal tax rate much greater than just their tax bracket (e.g., due to unique taxation of Social Security, a retiree in 15% bracket can have federal marginal rate of 22.5% or 27.75%). I am worried that it somehow came through very wrong.

Mike, I don't think you were misquoted - but I do disagree with the above and my research in TurboTax seems to confirm it - the max marginal rate that SS can create from anyone in the 15% bracket is 21.25% - in most cases less.

Mainly because the maximum that will fall out to line 43 on the 1040 is 85% of the SS and the maximum tax rate is 25% so 25 x .85 gives you 21.25%.

fd

[FinancialDave]

Is there any way at all to access this article online?

I don't subscribe to any print publications, and my local grocery store still has the July issue of Money for some reason.

I spoke with Penny for this article with regard to how the ACA and taxation of Social Security can make a retiree's marginal tax rate much greater than just their tax bracket (e.g., due to unique taxation of Social Security, a retiree in 15% bracket can have federal marginal rate of 22.5% or 27.75%). I am worried that it somehow came through very wrong.

Mike, I also agree that the taxation of SS is very non-linear, but when you add SS into a person's already existing tax return, I have never seen a case where it adds more than 85% of that income to your AGI. Am I missing something in IRC-86?

So I think the Money article has the "Tax diversification" section of the article wrong as well.

Besides the above the article says each additional $ of SS is reduced by 50 cents OR 85 cents -- the truth is it can be reduced from ANYWHERE between $0 to 85 cents depending on the persons other income.


fd

[dodecahedron]

Is there any way at all to access this article online?

I don't subscribe to any print publications, and my local grocery store still has the July issue of Money for some reason.

I spoke with Penny for this article with regard to how the ACA and taxation of Social Security can make a retiree's marginal tax rate much greater than just their tax bracket (e.g., due to unique taxation of Social Security, a retiree in 15% bracket can have federal marginal rate of 22.5% or 27.75%). I am worried that it somehow came through very wrong.

Mike, I don't think you were misquoted - but I do disagree with the above and my research in TurboTax seems to confirm it - the max marginal rate that SS can create from anyone in the 15% bracket is 21.25% - in most cases less.

Mainly because the maximum that will fall out to line 43 on the 1040 is 85% of the SS and the maximum tax rate is 25% so 25 x .85 gives you 21.25%.

fd

Financial Dave, I believe you are forgetting that the additional dollar of RMD that triggers the taxability of the additional dollar of SS benefit is itself taxed (at 15% in your example.) So it is quite complicated to determine the marginal tax rate of SS income because you have to take into account the direct and indirect tax effects and combine them in order to figure the effective marginal tax rate.

NBER's TaxSim confirms that an effective marginal tax rate 27.75% is quite possible. (Software developed and refined over decades by public finance economists, widely used and respected by academics as well as policymakers at IRS, Treasury, CBO, OMB, and Congress's Joint Committee on Taxation.) Take a single person with no dependents and standard deduction. Give him $30,000 in IRA distributions and $25,000 of Social Security gross benefits and no other income (just to keep it simple). Plug it into TaxSim and you will see that the marginal tax rate is indeed 27.75% for a person who is in the 15% statutory tax bracket. Why? Because an additional dollar of RMD raises taxable income by $1.85 and raises the tax bill by .15*$1.85=$.2775.

http://users.nber.org/~taxsim/taxsim-calc9/

It is easy to construct scenarios in which the effective marginal rate of a traditional IRA withdrawal is even higher than 27.75%. (E.g., throw in some long-term capital gains or dividends into the mix as well and let the RMD be just at the point where more of those types of income would be pushed from being taxed at 0% to being taxed at 15%. Or consider effects on Medicare IRMAA premiums.)

[jsandra]

I read this article a few days ago and was quite confused by it.

Thanks, Brantley, for posing this question. And to the Bogleheads for your quick responses.

[FinancialDave]

Is there any way at all to access this article online?

I don't subscribe to any print publications, and my local grocery store still has the July issue of Money for some reason.

I spoke with Penny for this article with regard to how the ACA and taxation of Social Security can make a retiree's marginal tax rate much greater than just their tax bracket (e.g., due to unique taxation of Social Security, a retiree in 15% bracket can have federal marginal rate of 22.5% or 27.75%). I am worried that it somehow came through very wrong.

Mike, I don't think you were misquoted - but I do disagree with the above and my research in TurboTax seems to confirm it - the max marginal rate that SS can create from anyone in the 15% bracket is 21.25% - in most cases less.

Mainly because the maximum that will fall out to line 43 on the 1040 is 85% of the SS and the maximum tax rate is 25% so 25 x .85 gives you 21.25%.

fd

Financial Dave, I believe you are forgetting that the additional dollar of RMD that triggers the taxability of the additional dollar of SS benefit is itself taxed (at 15% in your example.) So it is quite complicated to determine the marginal tax rate of SS income because you have to take into account the direct and indirect tax effects and combine them in order to figure the effective marginal tax rate.

NBER's TaxSim confirms that an effective marginal tax rate 27.75% is quite possible. (Software developed and refined over decades by public finance economists, widely used and respected by academics as well as policymakers at IRS, Treasury, CBO, OMB, and Congress's Joint Committee on Taxation.) Take a single person with no dependents and standard deduction. Give him $30,000 in IRA distributions and $25,000 of Social Security gross benefits and no other income (just to keep it simple). Plug it into TaxSim and you will see that the marginal tax rate is indeed 27.75% for a person who is in the 15% statutory tax bracket. Why? Because an additional dollar of RMD raises taxable income by $1.85 and raises the tax bill by .15*$1.85=$.2775.

http://users.nber.org/~taxsim/taxsim-calc9/

It is easy to construct scenarios in which the effective marginal rate of a traditional IRA withdrawal is even higher than 27.75%. (E.g., throw in some long-term capital gains or dividends into the mix as well and let the RMD be just at the point where more of those types of income would be pushed from being taxed at 0% to being taxed at 15%. Or consider effects on Medicare IRMAA premiums.)

I think you are misunderstanding what I am saying and what the Money article said. I certainly do understand that you can add a $100 of income and it can cause both that income to be taxed and also the SS to be taxed, but the above kind of thinking is "blurring" what is really going on in that the $100 income is taxed at some percentage and then part of the SS tax is "uncovered" and taxed at some "smaller" rate.

The quote attributed to Mike (no offense Mike as maybe they got it wrong like everything else) in the Money article actually said:

Many retirees in the 15% bracket actually have a marginal tax rate of 22% or 27% when SS taxes are added in,

Having worked on hundreds of taxes, what this means to me is you put in all the income into the 1040 from all sources other than the SS and the tax payer is in the 15% bracket. THEN (and not the other way around) you add in the SS to find the marginal rate due to the SS. The most your income can go up from this step (forgetting about other non-linear credits that usually don't apply to the retired anyway) is 85% of the SS total. Hopefully we are not talking about how "other" credits completely foul up even the notion of a marginal tax rate. In this simple example the implication is the SS can somehow be taxed at a marginal rate of 27% - it's just not possible using two sources of unearned income such as a retiree would have (IRA & SS or Pension.)

I know the study you referenced is talking about what happens to the retirees taxes when $100 of RMD income is added to someone already on SS income - it can look like the marginal rate is now 27%, but the assumptions are way off base IMHO.

To avoid this problem, (as I do when filling out tax returns,) all I do when the tax payer is looking over my shoulder at the "amount due" box is either CLOSE the amount due box, OR fill in the SS income last, which can only ever raise their taxes by 85% of the next highest tax bracket - which for our example is the 25% bracket, which would be 21.25%.

I hope that makes sense, because that is much as I want to get into tax rates on this particular thread, which was basically news about a poorly written Money article.

fd

[House Blend]

^I suppose you can try to argue that if one writes "Many retirees in the 15% bracket actually have a marginal rate of X% or Y% when SS taxes are added in," then the one and only mistake here is that X% and Y% should represent possible marginal rates on SS income, so something like 50% or 85% of 15% or 25%.

Maybe, but I think that if there is any mistake here, it is only that it is not worded in a way that makes it sufficiently clear that the original author is writing about the marginal rates on ordinary income.

In any case, the marginal rate on ordinary income has actual tax-planning value for marginal decisions. This is the rate that matters when deciding whether to convert another $100 to Roth.

Most people don't have any control, at the margins, over their SS income.

[kenyan]

I read the print article, since I'm currently receiving the magazine in an effort not to lose my airline miles on a particular airline. My impression was that it was all predicated upon someone who could max out all retirement accounts, and then invest further for retirement in taxable. This is a pretty small segment of the population, and does not include me. I'll take my ~39% fed+state+tax credit phaseout benefit now, thank you. We do have Roth IRAs and can hopefully control our taxes in retirement to be lower than that rate.

Perhaps if our income unexpectedly jumps by a large amount, we would exercise this option, but I have other investments that are higher priority in my list (mortgage, college).

[FinancialDave]

^I suppose you can try to argue that if one writes "Many retirees in the 15% bracket actually have a marginal rate of X% or Y% when SS taxes are added in," then the one and only mistake here is that X% and Y% should represent possible marginal rates on SS income, so something like 50% or 85% of 15% or 25%.

Maybe, but I think that if there is any mistake here, it is only that it is not worded in a way that makes it sufficiently clear that the original author is writing about the marginal rates on ordinary income.

In any case, the marginal rate on ordinary income has actual tax-planning value for marginal decisions. This is the rate that matters when deciding whether to convert another $100 to Roth.

Most people don't have any control, at the margins, over their SS income.

I do understand there is "some" value in understanding what your current marginal rate is on various income streams, but my personal opinion is people spend way too much time on this aspect of tax planning. After al,l maybe you say that $100 came from a Roth conversion -- maybe I say that the $100 that moved your SS up the variable scale came from $10,000 from a stock sale you did in December, or the $5000 of qualified dividends from a mutual fund that paid a special dividend in December you had not counted on.

What really should have been stressed in "tax diversification" is that you need sufficient amounts of both Roth and non-Roth funds if you want any hope of being able to adjust your tax rates in retirement to optimize your income. There is no point in spending time analyzing todays tax rates when trying to make this Roth / non-Roth decision because it is just not possible to predict your tax rates 20-30 years in the future.

fd

[FinancialDave]

I read the print article, since I'm currently receiving the magazine in an effort not to lose my airline miles on a particular airline. My impression was that it was all predicated upon someone who could max out all retirement accounts, and then invest further for retirement in taxable. This is a pretty small segment of the population, and does not include me. I'll take my ~39% fed+state+tax credit phaseout benefit now, thank you. We do have Roth IRAs and can hopefully control our taxes in retirement to be lower than that rate.

Perhaps if our income unexpectedly jumps by a large amount, we would exercise this option, but I have other investments that are higher priority in my list (mortgage, college).

And this is just the thinking propagated by these such articles and CFP's who don't seem to understand the complete picture.

Wasting money for no apparent reason other than you want to avoid taxes later, is a poor habit to get into, or to teach to others (IMHO.)

Now maxing out the Roth because you want to equalize the amount of Roth funds you have with other tax deferred money makes perfectly good sense.

The big problem that irritates me every time I see something like this is that many young people read or are taught this "stuff" and then go out and think that 100% Roth is the ONLY way to go because it means I will have to pay no taxes in retirement. Sure maybe they justify this as a future risk reduction strategy on tax raises, but at what price (nobody really knows.)

The title "Stop Passing Up This Great Deal" is actually a great topic, if the point had been to encourage people to at least have some Roth investments when they reach retirement and to understand that both Roth and tax-deferred savings can give you the same result for each dollar you earn, if your cumulative "in/out" tax rates happen to be the same - which is about all you can predict at the current time.

fd

[dodecahedron]

In any case, the marginal rate on ordinary income has actual tax-planning value for marginal decisions. This is the rate that matters when deciding whether to convert another $100 to Roth.

Precisely! The focus of the article is about the decision whether to recognize ordinary income now (Roth route) or later (traditional route), so it is the effective marginal rate with respect to ordinary income that matters in the context of this article.

As HouseBlend noted, computing the effective marginal tax rate with respect to a change in SS income itself has very limited value for decisionmaking because the amount of SS is not easy to manipulate at the margin.

It is the effective marginal rate with respect to other types of income (ordinary, dividends, tax-exempt) which are easier to manipulate at the margin that is worthwhile computing for tax decisionmaking purposes. The NBER TaxSim website I linked above will compute a variety of effective marginal tax rates for you with respect to many different types of marginal, incremental decisions a person might be expected to make. They don't bother to compute an effective marginal tax rate for SS income precisely for the reason HouseBlend noted.

The marginal tax rates they will compute for you (if you check the appropriate box) are with respect to:

Other Income (meaning most types of ordinary income that is not labor source)
Wage Income
Taxpayer Earnings
Spouse Earnings
Long Term Gains
Other Deductions (not mortgage interest)
Dividends
Mortgage Interest

These are the kinds of deductions most people can easily manipulate at the margin (at least the timing of).

[dodecahedron]

Financial Dave, I totally agree with you that if the article didn't urge tax diversification, it really missed the boat! Impossible to calculate these marginal tax rates exactly in any case, because there are so many potentially interacting factors. It is definitely worth keeping in mind the qualitative general principle that one's effective marginal tax rate in retirement is likely to be a good bit higher than the statutory bracket rate, but it is hardly an exact science!

[FinancialDave]

Financial Dave, I totally agree with you that if the article didn't urge tax diversification, it really missed the boat! Impossible to calculate these marginal tax rates exactly in any case, because there are so many potentially interacting factors. It is definitely worth keeping in mind the qualitative general principle that one's effective marginal tax rate in retirement is likely to be a good bit higher than the statutory bracket rate, but it is hardly an exact science!

And to be clearer on what I have already said -- A person in the accumulation / working phase of their life, which is what this article is about, has basically no reason to be trying to predict their marginal tax rate in retirement, which even if it is only 5 years from now, it is likely to last 20-30 years, when who knows even if there will be marginal tax rates!

So, yes, as you state it is very hard to predict potentially interacting factors for the current year you are working or retired, but the future (which I assume is when SS comes into the tax payers "marginal thinking," is next to impossible.

fd

[ObliviousInvestor]

The article is now available online, by the way:
http://time.com/money/3080353/roth-401k ... not-using/

[ObliviousInvestor]

I think that if there is any mistake here, it is only that it is not worded in a way that makes it sufficiently clear that the original author is writing about the marginal rates on ordinary income.

Yes, this is a good point. When I said that people often have marginal tax rates of "22% or 27% while in the 15% tax bracket" that was in regard to the marginal tax rate on income coming out of tax-deferred accounts. I should have been more clear about that.

[Iorek]

Looked at the first page of the T Rowe Price study the article cites and it appears they used a very lazy (and Roth-favorable) way of estimating taxes on the taxable account (where the savings from a tIRA were invested):

" a 25% tax is subtracted annually from the taxable account during the years leading up to retirement and then was taxed at the same rate as their income during retirement."

So the result only seem to really hold if you are investing in bonds your taxable account and you are in the 25% tax bracket. Leaving aside the tax bracket issue, I think many people will have 50-70% of their annual returns in form of capital appreciation-- which is not taxed at all during accumulation and is currently taxed at a significantly lower rate than ordinary income when realized.

I am surprised that TRP would bake such unrealistic assumptions into the study (ISTR some bogleheads have developed spreadsheets which try to take a more sophisticated, and realistic, stab at trying to estimate the differences between investing in a taxable account vs a tax advantaged account).

[Don Christy]

2) Some items in this article seem incorrect or misstated such as,

"Roth 401ks leave just about all workers, regardless of age or tax bracket, with more money to spend in retirement than pretax plans do"

"The Roth 401k should be considered the default investment"

"Every dollar you save in a Roth 401k is worth more than a dollar you put in a pretax account. That's because you'll eventually pay income taxes on those pretax dollars, while you get to keep every Penny in a Roth"

These are true but misleading. If you save $10,000 in a Roth 401(k), you will be better off than if you save $10,000 in a traditional 401(k), because the dollars are more valuable. However, if you are in a 25% tax bracket, you don't have this choice; if you can save $10,000 in a Roth 401(k), you have the option of saving $13,333 in a traditional 401(k) because you will reduce your taxes by $3333. (And you don't have to wait until next April to get your tax savings, as your employer will reduce your withheld taxes when you increase your 401(k) contribution.)

I've run into this bad argument before when people discussed the traditional or roth ira's. Grabiner gets it right as usual. I really can't believe a finance magazine would publish this.Where was the editor?

If you can't believe a finance magazine, how to you feel about an audiocast by the College of Consulting Actuaries (CCA)? I attended this audiocast on July 9, described below. Note one of the presenters was Maria Bruno.

Description: Actuaries are often sought out for a range of advice on the individual aspects of employee benefit plans. At this session presenters discuss a range of topics relevant to our consulting business covering individual retirement plan investing, retirement plan taxation and Medicare and Social Security taxation. The presenters share their perspectives from the accounting and financial planning professions.
Speakers: Brian M. Septon, The Terry Group; Maria A. Bruno, The Vanguard Group; Laura S. Goodman, FGMK, LLC

The presentation made two glaring errors in my opinion:

- in an example of tax efficient asset location, the asset allocations were not "tax adjusted". What I mean is that they compared before- and after-tax returns with the assets held in different locations. They always kept 50% in taxable account and 50% in tax-deferred account. The scenarios were all supposed to represent a 60/40 equity/bond allocation. So one scenario, 50% equity in taxable, 10% equity and 40% taxable bonds in tax-deferred (recommended as most tax efficient location). Another scenario was the opposite location, or 50% equity in tax-deferred, 10% equity and 40% taxable bonds in taxable account. These look like the same asset allocation, but they are not the same on a tax-adjusted basis. I'm not suggesting that tax efficient location isn't important, but that their scenarios didn't compare portfolios with the same AA.

- the second error is the same one made in this article regarding comparing $17.5 in 401(k) vs in Roth 401(k) and didn't take into account that they really don't represent the same level of savings.

EDIT: for those that are members, or are interested in joining, the presentation can be found here:
CCA Audio/Webcast
Is There an Actuary in the House? Financial Planning and Taxation for Actuaries

Non-members may also be able to buy it here, but I'm not certain whether this is just for a certificate proving attendance or provides access to the materials.
July 9 Audio/Webcast - Certificate Processing Fee for Nonmember Attendee