This post is copied (and slightly
modified) from the
One-Fund Portfolio thread. Enjoy!
Three investors each have $100,000 split between a $50,000 traditional IRA and a $50,000 Roth IRA. They share the same 50/50 stocks/bonds allocation and, like most investors, they
don't tax-adjust their allocation. In other words, they all initially hold $50,000 in stocks and $50,000 in bonds.
They differ in their
asset location strategy:
- Investor A prioritizes the placement of stocks into the Roth IRA first.
- Investor B prioritizes the placement of stocks into the traditional IRA first.
- Investor C maintains a 50/50 stocks/bonds allocation in each account.
The three investors intend to deplete their accounts over a fixed 30 year period, taking withdrawals at the beginning of the year. Traditional IRA withdrawals are taxed 20%. The portfolio is rebalanced when taking the annual withdrawal.
A simple approach, often used to analyze outcomes, is to use a constant
real growth rate for stocks and for bonds. Let's pick simple numbers: 5% for stocks and 2% for bonds. In other words, in real terms, the portfolio annually grows by a constant
((5% X 50%) + (2% X 50%)) = 3.5% rate.
NOTE: Calculation tables are provided in a separate post of the One-Fund Portfolio thread to keep this post readable.
If everything goes according to plan, the three investors withdraw an identical total of $157,598 from their accounts over 30 years, before tax, because the asset location strategy has no impact on portfolio growth in tax-sheltered accounts.
Investor A withdraws a total of $64,276 from the traditional IRA, pays a total of
-$12,855 in taxes, and withdraws a total of $93,322 from the Roth IRA. Total net income:
($64,276 + -$12,855 + $93,322) = $144,743.
Investor B withdraws a total of $93,322 from the traditional IRA, pays a total of
-$18,664 in taxes, and withdraws a total of $64,276 from the Roth IRA. Total net income:
($93,322 + -$18,664 + $64,276) = $138,934.
Investor C withdraws a total of $78,799 from the traditional IRA, pays a total of
-$15,760 in taxes, and withdraws a total of $78,799 from the Roth IRA. Total net income:
($78,799 + -$15,760 + $78,799) = $141,838.
The difference in total net income of the three investors is only due the amount of taxes paid.
It's surprising how
small the impact of
choosing the worst asset location strategy is when everything goes according to plan. While investor B pays a total of
(($18,664 / $12,855) - 1) = 45% more in taxes than investor A (which might look impressive),
the impact on net income is only (($138,934 / $144,743) - 1) = -4% less.
The impact of investor C choosing a mirror allocation in all accounts on net income is only
(($141,838 / $144,743) - 1) = -2% less than the best
asset location strategy, when everything goes according to plan.
Obviously, when everything goes according to plan, investor A wins because, by prioritizing the placement of the slower growing asset into the traditional IRA, less money is withdrawn from it resulting into a smaller total tax bill.
First lesson: It's best focus on the
big numbers that matter, like total net income available to spend after taxes, instead of focusing on ratios between small numbers with little impact on the retiree's wellness, like ratios of between tax amounts.
Unfortunately, everything
doesn't always go according to plan in life. A simple and very effective approach to estimate the impact of unfavorable outcomes is to first apply an immediate
-50% loss to the stock allocation of the portfolio, then to conduct the same simple analysis with constant growth rates.
If stocks lose -50% of their value just before taking the first withdrawal, the three investors withdraw an identical total of $118,199 from their accounts over 30 years, before tax, because the asset location strategy has no impact on portfolio growth in tax-sheltered accounts. That's
(($118,199 / $157,598) - 1) = -25% less than when everything goes according to plan, corresponding to the impact of losing
(-50% X 50% of portfolio) = -25% of portfolio.
Investor A withdraws a total of $69,281 from the traditional IRA, pays a total of
-$13,856 in taxes, and withdraws a total of $48,917 from the Roth IRA. Total net income:
($69,281 + -$13,856 + $48,917) = $104,342. That's
(($104,342 / $144,743) - 1) = -28% less than when everything goes according to plan.
Investor B withdraws a total of $48,917 from the traditional IRA, pays a total of
-$9,783 in taxes, and withdraws a total of $69,281 from the Roth IRA. Total net income:
($48,917 + -$9,783 + $69,281) = $108,415. That's
(($108,415 / $138,934) - 1) = -22% less than when everything goes according to plan.
Investor C withdraws a total of $59,099 from the traditional IRA, pays a total of
-$11,820 in taxes, and withdraws a total of $59,099 from the Roth IRA. Total net income:
($59,099 + -$11,820 + $59,099) = $106,379. That's
(($106,379 / $141,838) - 1) = -25% less than when everything goes according to plan.
The difference in total net income of the three investors is only due the amount of taxes paid.
Investor A is now the loser, trailing investor B (winner) and investor C (average). The overall impact of choosing an
asset location strategy remains small. Investor A gets
(($104,342 / $108,415) - 1) = -4% less than the winner, and investor C gets
(($106,379 / $108,415) - 1) = -2% less than the winner.
The
after-tax impact of stocks losing
-50% of their value varies according to
asset location strategy. As a result of stocks losing
-50% of their value, investor A loses
-28% in total net income because the loss happens in the Roth IRA. In contrast, investor B only loses
-22% in total net income because the loss happens in the Traditional IRA. For investor C, the after-tax loss impact is identical to the before-tax loss impact.
What we see is a typical risk/reward outcome. It's as if, after tax, investor A is taking
more risk than a 50/50 stocks/bonds allocation by prioritizing the placement of stocks in the Roth IRA, winning when things go well and losing when they don't. It's also as if, after tax, investor B is taking
less risk than a 50/50 stocks/bonds allocation by prioritizing the placement of stocks in the Traditional IRA, losing when things go well and winning when they don't.
Investor C, in contrast, seems to be taking, after tax, the same amount of risk as a 50/50 stocks/bond allocation. In other words, investor C's
asset location strategy doesn't seem to change the amount of risk taken before and after taxes.
Another way to view this is to consider that an outcome similar to investor A can probably be achieved, in the above scenario, by increasing the stock allocation to
(-2 X -28%) = 56% of the portfolio and adopting an identical asset allocation in both the traditional IRA account and the Roth IRA account. Similarly, investor B could probably achieve a similar outcome by reducing the stock allocation to 44% and adopting a mirrored allocation.
Actually, this has been verified in this post of the One-Fund Portfolio thread.
Second lesson: It's a mistake to only consider good scenarios and ignore risk, when evaluating the after-tax impact of asset location strategies. A complex strategy delivering
better outcomes when things go well and worse outcomes when they don't, when compared to a simpler strategy, unnecessarily complicates the life of its investor. A slightly-higher stock allocation with the simpler strategy is likely to deliver similar outcomes.
Summary: An identical asset allocation in all portfolio accounts (a
mirrored asset allocation) doesn't affect the effective riskiness of the portfolio, even after taxes. In contrast, so called "tax-efficient"
asset location strategies often promise better "expected" outcomes without disclosing that they do so by increasing the effective
after-tax risk of the portfolio. In other words, a mirrored asset allocation is not only
good enough, it also delivers more consistent outcomes.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)