Bogleheads

[grok87]

http://online.wsj.com/article/...._nonsub_pj

In today's column J. Clements makes the case for only contributing to your 401k up to the match, then maxing out Roth, and then going with taxable investing rather than 401k.

I think this is bad advice myself. But he does have an interesting argument about diversifying tax risk. This is one of the reasons I have a substantial muni bond allocation and have some stocks in my 401k (mostly REITS and small caps). I'd rather take some of my tax breaks now (the muni tax break is pretty substantial now that munis yield same as treasuries) rather than make a big bet that long term capital gains tax rates on stocks will remain low when I retire.

cheers
grok

[nisiprius]

That's interesting.

Certainly 401(k)'s are not the no-brainers they used to be. One by one all of the various assumptions that one made them obviously favorable have been undercut to some extent.

a) The tax brackets are so close to flat, at least in the middle incomes, that the assumption that you'll be withdrawing at a lower tax rate than you did when you contributed is now a weak one.

b) A substantial capital gains tax break is now available to taxable investors.

c) The "tax torpedo" effect, not mentioned by Clements. If you are drawing Social Security at the same time as you are taking 401(k) or IRA distributions, then not only are the distributions taxed as ordinary income, they also have the effect of increasing the amount of Social Security earnings that are taxable, creating some ranges in which the marginal tax rate can be as high as 50%. As nearly as I can make out, those effective-50% tax brackets are fairly narrow, but still.

d) Now throw in the hidden fees in the 401(k)s. (I wish I'd had the gumption to buy equal amounts of the same Fidelity fund in my Fidelity-managed 401(k) and a brokerage account, just to compare results).
_________________
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

[jh]

...

[Wagnerjb]

IMO, the quality of your 401k should play a part in this decision. If you have crappy funds with high costs, I would go the taxable route. However, if you have a wide variety of low-cost passive choices in the 401k I would consider putting some (if not all) of the remainder in the 401k. I do believe in having some assets in taxable for flexibility, but it is hard to beat the tax deferral in a good 401k.

Best wishes.
_________________
Andy

[EmergDoc]

It has to be a pretty crappy 401K to make you want to pay taxes three times instead of once.

In a 401K you pay your regular tax rate upon withdrawal.

In a taxable account you pay your regular tax rate when you make the money AND you pay taxes annually on the distributions AND you pay capital gains taxes upon withdrawal.

Previous number crunching has shown that you need an ER>1.5% and no match to hold stocks preferentially outside the 401K.
_________________
1) Invest you must 2) Time is your friend 3) Impulse is your enemy
4) Basic arithmetic works 5) Stick to simplicity 6) Stay the course

[VictoriaF]

[SoonerSunDevil]

While the quality of the 401(k) should be considered, the length of time you plan on staying with your employer is perhaps more important. I knew that I would work for my current employer for 19 months and that I would contribute as much as possible because I would soon have the ability to move the funds to a Rollover IRA. While not everyone knows how long they will be with their current employer, certain circumstances might warrant a 401(k) participant to contribute up to the maximum despite a high fee structure.

John

[petrico]

[petrico]

[EmergDoc]

[Wagnerjb]

Hi EmergDoc: I don't disagree with the issues you have highlighted, namely the time until rollover or retirement, the ER, the current and future tax rates, etc.

However, there are several small tax benefits that should also be considered:

a) The ability to tax loss harvest
b) The loss of the foreign tax credit in the 401k (double taxation),
c) The ability to time sales in retirement to optimize taxes (selling high cost investments first, for example), and
d) The hassle of minimum required distributions in an IRA.

These are smaller issues than the ones you have highlighted, and probably won't change the overall conclusion. But they are factors that need to be weighed along with an investors age, expected time at the employer, ER of funds in 401k, current tax rate, etc.

Best wishes.
_________________
Andy

[tfb]

I disagree with what Clements suggested. Time to rollover is a big factor. Avoiding paying state income tax is another big factor for people in high tax states who will not be in high tax states when they retire. Taking deduction from the top while filling in the tax brackets from the bottom at retirement is another biggie. All in all, it's almost impossible to beat 401k with a taxable account.
_________________
Did you search the forums and the wiki?

[petrico]

[petrico]

[yobria]

Poor ole John, he's devoted his career to writing about a topic (Diehard-ish investing) that's just too dern simple to merit a regular news column. So he discusses the same material over and over, (over)thinking it in new and often dubious (as in this case) ways. He does this with exotic, high fee asset classes now and then too - seems so desperate for something new he can't resist the urge to toss them in.

Nick

[Adrian Nenu]

401k vs taxable rule of thumb:

Fees/return > tax bracket, then probably not worth the tax deferral compared to low cost, tax efficient Vanguard account (assuming no employer's match).

But still have to do the math because EVERY investor's situation is different. Then there are other issues to consider like tax bracket, probable length of employment, discipline to save in a taxable account vs one which has the contribution automatically deducted from the paycheck, etc, etc.

Consider that the traditional advice is to hold the bond allocation in the 401k. Does it really make sense to earn maybe 5-6%, subtract 2% in fees with a net return of 3-4%, considering that inflation is killing any real return?! This is how expensive 401k plans fail the investor and should be avoided.

Adrian
anenu@tampabay.rr.com

[MWCA]

[fishndoc]

VictoriaF wrote

[EmergDoc]

[MP173]

This is an interesting conversation. The concern about rising taxes, is a valid one. How will it affect our retirement? That is the crystal ball everyone wishes they had.

Adrian, you mentioned 2% 401k plan fees. Was that a general statement about poor plans, or is there something I am missing with my plan?

I am at a crossroads with our plan. A Roth 401 is being offered this year and at age 52 I am at the point that it probably wouldnt make sense...under the current tax code.

However, I am inclined to participate in order to diversify my retirement holdings a bit more. Comments?

ed

[EmergDoc]

As you can see, I think it would be dumb of you not to participate in your 401K. The dollars you're saving now will probably sit in your 401K for 10 years, then be rolled into a low-cost Vanguard IRA and not be spent for another 10-30 years. Hard to beat tax-deferred growth AND a tax-break up front when you're at the peak of your lifetime earnings cycle.

Edit: NM. I missed the key word Roth. You are choosing between a Roth and a regular 401K. Totally different question.
_________________
1) Invest you must 2) Time is your friend 3) Impulse is your enemy
4) Basic arithmetic works 5) Stick to simplicity 6) Stay the course

[Plainsman]

From Clements:

[EmergDoc]

[Tramper Al]

[philip]

EmergDoc wrote:

[tfb]

[Plainsman]

[financialguy]

I think you may be misinterpreting Clements' article. The whole middle section (subtitled "Still winning") says that you should in fact invest in your 401k, and the taxable account only wins in rare cases (such as a very low time horizon).

[plake15]

The surprising answer: Even if your income-tax rate jumps sharply, the 401(k) will likely be the better bet, calculates Allan Roth, a financial planner in Colorado Springs, Colo. Indeed, if you have 15 years to invest, the 401(k) will leave you with more money, as long as your tax bracket doesn't climb above 33%.

thats the problem..as long the tax bracket doesn't climb above 33%.

15 years from now 33% might be a low tax bracket rate.Between AMT,social security,medicare,the huge goverment budget defiect...etc..

Historically rates right now are very low..I think we can almost all agree they will be higher in the future,the big question is,how much higher?

[Adrian Nenu]

[grok87]

[whitemiata]

[petrico]

[petrico]

[stan1]

Estimated value of 401(K) (actually TSP) at retirement:
$1M

Estimated tax to convert into a Roth (Stretch) IRA (spread over 10 years)
$343K (Yikes! Maybe I'll take RMD's instead)

Assumption: 25% federal, 9.3% CA

[EmergDoc]

[EmergDoc]

[grok87]

[petrico]

[petrico]

[EmergDoc]

[Easy Rhino]

[petrico]

The basic premise of the Clements article is tax risk, and how it might affect asset location decisions. To keep things simple, it seems as though some simplifying assumptions were made in his analysis. First, expenses were neglected in both the 401k and taxable location. Second, tax-generating dividends seem to have been neglected in the taxable account. The analysis only considered varying future tax rates and time horizons.

The examples that have been considered in this thread so far have maybe been more realistic, by further considering higher expenses in the 401k and taxes on dividends in the taxable account. Expenses are usually higher in 401k’s, which tends to favor taxable accounts, but on the other hand even tax-efficient equities generate some dividends that will be taxed, which tips things back a bit towards tax-deferred investing.

However, is it possible to make an even more realistic comparison? Two additional factors neglected so far are 1) the possibility of subdued equity returns over the next couple or so decades; and 2) the effects of inflation. Both factors tend to produce lower real returns, and that tends to favor taxable investing, given higher future taxes (the basic type of risk assessed in the article) and higher expenses in the tax-deferred location. I think that’s because the lower the returns, the larger the percentage of the returns in the 401k are lost to expenses and taxes compared to the taxable location. (This seems to be the same case Jack Bogle makes in his Little Book of Common Sense Investing on the effects of lower returns, investment costs, taxes and inflation on real returns, pp. 72-74.)

The case for lower long term returns, using the Gordon Equation, is made by Bernstein (in Chapter 2 of The Four Pillars of Investing), Malkiel (in Chapter 13 of the latest edition of A Random Walk Down Wall Street), and Jack Bogle (in Chapter 7 of The Little Book of Common Sense Investing). Bernstein and Bogle both estimate real returns of roughly 5% for equities. Malkiel estimates a 7.5% nominal return, which is reduced to a real return of about 5.1% using the 2.3% inflation estimate cited by Bogle. They’re all right around the same ballpark.

So if we look at a previous example, is it valid to reduce the returns to reflect (decent) estimates of long term real returns of equities?

If we again assume:
$10,000 investment
25% initial income tax rate
35% future income tax rate at withdrawal
20% future cap gains/dividends rate
1.0% ER in 401k
0.2% ER in taxable and Rollover IRA accounts
10 years to rollover
but now assume only a 5% real return

401k w/ rollover
@ 10 yrs to IRA: FV(.048,13,,-FV(.04,10,,10000)*.65 = $17,698.78 (real dollars)

Taxable account: (FV(.044,23,,-7500)-7500)*.8 + 7500 = $17,653.35 (real dollars)

The lower real return produces about a 23-year break-even period before it’s better to use the 401k. Alternatively, it “only” takes about a 31% future income tax rate to produce the 16-year break-even period found before with a 35% tax rate and higher nominal returns.

It seems that even this example is simplistic, because these are presumably marginal tax rates, not effective tax rates.

But otherwise, is this a valid way of analyzing the effects of taxes, expenses, and inflation on real returns? Or did I miss something basic?
I’d appreciate a reality check by someone more knowledgeable.

Thanks,
--Pete

[EmergDoc]

Pete-

Whoa nelly! That's serious stuff using real returns! Good point about the effects of lower future returns though. Naturally if real returns are higher than expected than that gives a big advantage to the 401K.
_________________
1) Invest you must 2) Time is your friend 3) Impulse is your enemy
4) Basic arithmetic works 5) Stick to simplicity 6) Stay the course

[petrico]

[snowman9000]

[EmergDoc]

[Buckeye]

[petrico]

[Jack]

Another consideration is that each year your choice to defer taxes or not is a lifetime decision. If you give up this year's opportunity to defer taxes, it is lost forever. Later in life if you decide that you have too much tax risk in deferred accounts you can always choose to put 100% of your income into taxable accounts. The reverse is not true. You don't have the option later in life of putting 100% of your income into a deferred account. Since future tax rates are unknown, you are better off delaying an irrevocable decision as long as possible.

Giving up your one time opportunity to defer each year should not be determined only by the expenses in your current 401k plan. Those circumstances can change. Many people only work 3 to 5 years for one employer, allowing you the opportunity to roll over into a low cost IRA. Or your employer could change your plan provider to a less expensive one. But those options are closed to you if you failed to contribute in the past.