While trying to time the market is a flawed and futile exercise, timing the taxation of investments is not. In fact, as an “advisor” (not “adviser”) I would argue this is mandatory.
Recently I wondered, could we use a hated, dreadful tool, like a 401(k) loan to financial benefit the consumer? Yep, I’m that cool that in my spare time my thoughts are perpetrated by fun ideas like that. As boring as it sounds, a 401(k) loan can be useful if the loan proceeds are deposited into a Roth IRA. Odd but true!
There are three main components that make a 401(k) loan/Roth IRA combination advantageous.
(Related: Want Better Sales? Ask Better Questions)
First, it is the manipulation of consumer behavior. In my experience, individuals who take out a 401(k) loan do not factor the slight additional federal tax liabilities into their decision.
Second, if we take the loan after a short-term negative fluctuation in value, then we can exploit the “buy and hold” strategy, or the “the-market-will-always-back” belief, by paying the tax on the depressed amount and allowing the growth to be tax-free (similar to the required minimum distribution example). Thus, allowing for increased withdrawal capacity.
Third, tilting the Roth to be completely growth creates a definitive mathematical advantage. Interestingly, except for the increased withdrawal capacity, the first two benefits are purely behaviorally superior while mathematically equal; whereas, the third and final benefit is mathematically superior while behaviorally inferior.
Timing the taxation of investments to take advantage of short-term market fluctuations isn’t as difficult as it sounds. In fact, I bet you’re already doing it.
I want you to imagine one of your clients, Bob and Mary, who own small business, have just shared that their income this year will be the lowest it’s been in decades. They expect their taxable income to drop from north of two-hundred large to just about fifty-grand. You all agree this is a short-term fluctuation (think industry shift, or unusually large depreciable deduction) and although it’s painful to pay additional taxes right now, you show them how a partial Roth IRA conversion would take advantage of their unseasonably and temporarily lower marginal rate. Very important, although often overlooked, Roth IRA distributions do not factor into whether future social security benefits are taxable, since all the growth is 100% tax-free.
There’s other examples of how we can utilize tax rules to increase net investment gains, but the 401(k) loan is one I haven’t seen mentioned before. Let’s get the icky part of this over.
Yes, I know a 401(k) loan typically has a reasonable interest rate, and yes, it’s true the interest paid goes back into the loan recipient’s 401(k), but the loan repayment is characterized as income, making the interest double taxed (only the interest since the “principal” already reduced the tax liability once and was not taxed when it came out as a loan). Most importantly though, too often I’ve seen a 401(k) used as a quasi-emergency fund, which it is not.
If you believe the market will always come back, then we can manipulate when taxes are paid to capitalize on negative short-term market fluctuates. Further, we can use negative consumer behaviors to increase long-term consumer successes.
To illustrate, let’s look at one of your clients again. She is 72-years-young and owns XYZ investment which has recently dipped 30%. She it’s a good investment and doesn’t want to sell it to cover her $7,000 annual RMD. With her beliefs, she should sell it now and reinvest the RMD into an after-tax account. The $7,000 of XYZ stock today, was previously worth $10,000. Selling now, she would owe ordinary income tax on the $7,000 but would pay long-term capital gains (LTCG) on the growth, if she holds the new shares for at least one year. The long-term capital gains rate is 0% for those in the 10% and 12% federal brackets. If XYZ does recover, she will save ordinary income tax on $3,000. Despite this, this is a behavioral advantage, not a mathematical one.
Technically, if she reinvests $7,000, rather than an amount after taxes, then she’s slightly increasing her total investment. If she invested the after tax, then it would be equal. For example if $100 pre-tax grows by a factor of 10, then the after-tax value is $800 ($1000 – 20%). If $80 after-tax grows by the same factor, then the value is also $800.
However, wouldn’t you agree, consumer sentiment is “I’ll take it now and pay for it later (taxes included)”. Thus, she’ll defer the tax liability to the follow year, or have it withheld and still re-invest an even $7,000. There’s also something to be said that this is a Required distribution.
What about wash rules, you ask? They don’t apply when going from a qualified account (IRA) to a non-qualified account, but important they do apply when going the opposite direction, non-qualified to qualified. But let’s move on
Okay we’ve established we can manipulate simple tax rules for your clients’ benefit during periods of short-term income fluctuations and possibly during short-term market fluctuations, but can this be done with a 401(k) loan?
Before we go further, it’s easier for me to list my assumptions. Below are the basic assumptions I used:
- $100,000 401(k) balance
- $415 per month either for contributions or loan repayments
- $21,991 401(k) loan at 5% Interest repaid equally over 5 years
- 60/40 Allocation (U.S. Stock Market and Total U.S. Bond Market)
- Same costs etc. between 401(k) and Roth IRA
- 20% marginal federal tax rate
- 30-year time horizon
Here are some answers to common questions I can imagine readers asking? If you don’t care skip down past this section.
Q: Why a 5-year loan term?
A: It’s, since this is the longest allowed by the IRS. A longer term creates more interest, which is double taxed, thus this should favor the no loan examples.
Q: Why $21,991 401(k) loan?
A: The payment equals $415 exactly which made the example cleaner. Further, I this would be a combined amount between husband and wife done earlier in the year when they could make both current and previous tax year’s contributions.
Q: Why assume they have $100,000 401(k) balance?
A: It’s not a magical number, nor is the loan amount. The max loan amount is $50,000 up to 50% of the account value. These amounts are both fungible and changing them would be without consequent to the findings below.
Q: There’s no 20% marginal federal bracket, why did you use it?
A: Marginal rate means the bracket each additional dollar is being taxed at, until you go into a higher bracket, at which time that would become your marginal rate. For this discussion, we need a net effect, and effective rates are irrelevant. For example, if you have an effective tax rate of 9%, but each dollar of additional income is being taxed at 22%, then do you really care that your earlier dollars were taxed lower, or do you care that all additional dollars are at the 22% rate? The later of course. For more accurate comparison, I used a hypothetical 20% marginal tax rate.
A 401(k) with ‘xy’ balance is no different than the sum of a 401(k) with ‘x’ balance and a Roth IRA ‘y’ balance. They each total ‘xy’ balance. Or in this case, a $100,000 401(k) is equivalent to a $21,991 Roth IRA and an after-loan 401(k) with a balance of $78,009. Ceterius peribus, if the starting values are the same, and the contributions are the same, then the ending values will also be the same.
The only difference between these examples, is the non-deductibility of the 401(k) loan payments versus the 401(k) contributions. However the principal payments aren’t being double taxed, only the interest is, and so we must on consider what the future growth could have been if the interest payment’s taxable liability had been invested. For example, if the interest payments totaled $1,000 during the tax year, then the liability wouldn’t be $1,000 it would be, $200 in this example.
To estimate what the potential value of the loan interest tax liability would be if it were invested, I used the free MonteCarlo simulator from http://www.portfoliovisualizer.com (statistical returns, no inflation on contributions, and $1,000 starting balance which is the minimum). After 30 years, the 10th, 25th, 50th, 75th, and 90th percentile values were:
$9,166 / $12,300 / $17,342 / $24,328 / $32,448
If we do the same for the Roth IRA (and add a $1,000 starting balance to offset the min. used above) the same percentile values were:
$138,678 / $188,540 / $263,392 / $371,549 /$507,996
Therefore, all else equal, the 401(k) loan, Roth IRA combo is mathematically superior if the future marginal tax rate is approx. 6.5%% more than the current marginal rate. Further, the equality of this comparison relies on the consumer adjusting his or her contributions, or loan amount, to offset the federal tax liabilities created. While that would be mathematically 100% correct, it would 100% ignore consumer behaviors, which by-and-large are not tax proactive unless used as a rational justification for a desired durable good (think tax rebate on electric vehicles).
Unlike the RMD example, the previous 401(k), Roth IRA combination example doesn’t try to exploit negative short-term market fluctuations. What if this strategy was triggered every time the US stock market entered the Alaskan frontier, IE bear territory?
Go back to the Roth IRA balance above. The difference between the 10th and 90th percentiles is largely explained by what gains, losses, occur when and how often. Can we agree that those who invest near cycle bottoms versus at cycle tops are likely to experience different overall returns. Thus, I would suggest, it’d be more accurate to use the 50th or 75th percentile values when used this way.
Since I’ve already shown the after tax values aren’t affected by when the tax is paid, if it’s at the same rate, the biggest benefit here it the increased withdrawal liquidity. Having $260,000 to a $370,000 tax-free is superior to the equivalent adjusted pre-tax value given the inherent limitation to gross withdrawal capacity necessary to avoid entering into a higher marginal tax bracket.
Mathematically, where this strategy becomes superior, is when we get a little more creative. If we skew the Roth allocation to be completely allocated toward growth, while modifying the 401(k) allocation to achieve a combined 60/40 between the two accounts, then the net value should increase since the Roth IRA growth is tax-free. FYI, in general, this should be done with anything that has Roth in-front of the title.
After modifying the allocations of the others accounts to offset the change to the Roth, the final, after-tax net increase for each percentile are:
$27,022 / $44,434 / $78,739 / $136,288 / S232,578
The larger net benefits at the higher percentiles is partially due to the 60/40 aggregate allocation straying more towards growth. For example, the 75th and 90th percentiles have aggregate stock to bond allocations of 66/34 and 69/31. In general greater allocations toward growth lead to greater account balances. However, at the 10th percentile, the aggregate allocation is only slightly greater than 60/40 (about four 1/1000th of a percentage point), yet it still garners a $27,000 net advantage. This is significant considering this the loan was only about $22,000. Interestingly, the mathematical advantage ignores the behavioral disadvantage, which is a consumers unlikeliness to view aggregate change in value among all their accounts, rather than the singular downward movement of an individual account during periods of negative market performance. This could cause a greater propensity for irrational reactions during these periods within the Roth IRA in comparison to the 401(k) despite being the same, overall, allocation.
Why is this so important? Easy, for quite some time, we’ve recognized the importance of returns net of fees, partially thanks to Mr. Bogle, to whom I sincerely thank! Yet, why aren’t was as focused on returns net of fees and taxes? Every dollar we help our clients save on taxes, is one dollar their investments don’t have to generate. If the investments don’t have to generate as much income or growth, then they can be allocated with less risk. And isn’t it easier to more accurately forecast less volatile, more stable investments than more volatile, less stable investments?
— Read 8 More Dave Ramsey Myths Debunked on ThinkAdvisor.
Michael Jay Markey Jr. is a co-founder and owner of Legacy Financial Network and its associated companies. He has been a member of the Million Dollar Round Table member and a winner of Court of the Table and Top of the Table honors.