Though the video triggered me because of some inaccurate/misleading statements from Krohn, I started to wonder whether he had a point. And no, not a point about whether you should contribute at all to your 401(k). You should always contribute to your 401(k) up to the point where you get the entire match from your employer. This is free money and not something that I consider up for debate.
However, Krohn may have had a point about contributing beyond the employer match. So I decided to look into the costs/benefits of doing so. To start, let’s analyze a Roth 401(k), because the tax assumptions are simpler, before examining the traditional 401(k).
Testing the Long-Term Benefits of a Roth 401(k)
To find the long-term tax benefits of a Roth 401(k), I compared a $10,000 annual investment into a Roth versus a $10,000 annual investment into a taxable account (i.e. brokerage) over 30 years. As a reminder, contributions into a Roth 401(k) grow tax-free and can be withdrawn after age 59 and 1/2 without incurring any taxes/penalties.
For this comparison, I assumed that both accounts grew at 5% a year and that the taxable account had to pay long-term capital gains taxes (15%) on its 2% annual dividend before reinvesting it. I also assumed that no sales were made in the taxable account until retirement (when all long term capital gains taxes are paid). It’s buy and hold for three decades. The purpose of this exercise was to quantify the financial benefit of avoiding taxes on annual dividends and long-term capital gains in a Roth 401(k).
After running this simulation for 30 years, I found that the Roth 401(k) ended up with $114,000 more than the taxable account (after all capital gains taxes had been paid):
That $114,000 means that the Roth 401(k) ends up with 14% more than the taxable account after 40 years. How much of an annualized benefit is this when we consider all contributions? 0.73%. That’s it. You get 73 basis points a year to lock up your capital until you are 59 and 1/2. Yes, under certain circumstances you can pull money out of your Roth 401(k) penalty-free before 59 and 1/2, but I don’t think you should use your retirement accounts in this way.
More importantly, my analysis assumes that you have the same investment options (and pay the same fees) in a 401(k) as you would in a taxable account. But we know that this isn’t always true. In fact, if the investment options in your employer’s 401(k) plan are just 0.73% more expensive than what you would pay in a taxable account, then the annual benefit of a Roth 401(k) is completely eliminated!
And this isn’t a high bar to hit. For example, if we assume that you would have to pay 0.1% per year in fees to get a diversified portfolio in a taxable account, then paying anything more than 0.83% (0.34% + 0.1%) in your Roth 401(k) would eliminate its long-term tax benefit entirely.
The average American typically pays about 0.45% in 401(k) fees each year. This means that the typical American’s 401(k) plan provides about 0.38% per year (beyond the employer match) relative to a well-managed taxable account.
Of course, if the fund fees in your employer’s 401(k) plan are low (0.2% or less), then there is still some monetary benefit to using it. But, you have to ask yourself: is 60-70 basis points a year worth locking up a decent part of your wealth until old age? I’m not necessarily sure.
Personally I feel like I made a financial mistake by contributing too much to my 401(k) when I was younger. While my retirement projections look great now, I also placed some limits on what I can do with my money.
For example, because I maxed out my 401(k) throughout most of my 20s, I can’t currently afford the sizable down payment required to buy a place in Manhattan. I’m not even sure if I want to buy, but if I did, it would take me a few extra years to get there because of my excessive 401(k) contributions. This is partially my fault for not planning ahead, but it’s also because I was seduced by the “max out your 401(k)” advice when I was younger.
The good news is that I now know the truth about the benefits of a Roth 401(k). The bad news is that these benefits are much harder to quantify for a traditional 401(k).
What About a Traditional 401(k)?
Now that we have looked at an account where the taxes are paid upfront [Roth 401(k)], we have to also consider what happens for an account where the taxes are paid in retirement [traditional 401(k)]. To make this comparison fair, I increased the contributions into the traditional 401(k) to match the after-tax contributions into the taxable account.
So if you contributed $10,000 to a taxable account and your tax rate was 24%, the equivalent sized contribution into a traditional 401(k) would be $13,158 [$10,000/(1 – 0.24)]. If your income tax rate in retirement is 24%, then this $13,158 contribution would be $10,000 after taxes. By making these contributions tax-equivalent we can compare them over time.
As a reminder, if your income tax rate while working is identical to your income tax rate in retirement, then it makes no difference whether you choose a Roth 401(k) or a traditional 401(k) as they will produce the same balance in retirement. If you don’t believe me, then read this, and read it again until you do.
From this axiom we can conclude that the benefit of a traditional 401(k) is equal to a Roth 401(k) when income tax rates are constant over time! And if we know that a Roth 401(k) has a 0.73% annual benefit (with the simulation parameters above), then so does a traditional 401(k) under the same parameters.
This means that the only thing that matters when calculating the benefit of a traditional 401(k) is the difference between your tax rate while working and your tax rate in retirement. Therefore, if you tax rate in retirement is much higher than your tax rate while working, the annual benefit of maxing out your traditional 401(k) could be completely eliminated.
And if you take this idea to its logical extreme, it can get much worse. For example, if income tax rates rise enough in the future, the benefit of of your prior 401(k) contributions could be negative relative to a taxable account. This simple observation illustrates the real risk of maxing out a traditional 401(k)—you could end up worse off if there are major tax hikes. I understand that you can make the same argument about taxable accounts and a rising capital gains rate. However, I would bet that capital gains rates will remain below ordinary income rates for the foreseeable future.
Does this mean that you shouldn’t contribute beyond the employer match in your traditional 401(k)? Not necessarily. But it does suggest that there is a significant risk in doing so. A risk that is based on future tax rates. While I will admit that I can’t predict future tax policy, it’s hard to imagine how the U.S. pays down its debts without increasing taxes. This chart from The Washington Post of U.S. deficit spending by year further illustrates this point:
I don’t show this chart to cause alarm, but to rationally consider the most likely path forward. Does this suggest that taxes will go down over the next few decades? Then why would you put so much of your money into an investment vehicle whose tax rate will be determined by a future Congress that is likely to be in dire need of cash?
This is the real risk of maxing out your traditional 401(k). Contributing up to the employer match is a no brainer that every investor should participate in (if they can). However, beyond that and you have to start considering the tradeoffs much more seriously.
The Bottom Line
Regardless of whether you have a Roth 401(k) or a traditional 401(k), the benefits of contributing beyond the employer match are smaller than you might have initially imagined. The issue is that so many personal finance experts have sung the praises of “maxing out your 401(k)” without providing any data to back up their claims.
How do I know? Because I used to be one of them. I used to advocate maxing out your 401(k) because it seemed like the logical thing to do. And with so many other financial personalities championing the same thing, why would I even question it?
But since running the numbers, I now know better. This is why it’s hard for me to support maxing out your 401(k) for an extra 0.73% per year. That illiquidity premium is just too small to be worth it even if you don’t need the money for something like a down payment on a house. And for those with a traditional 401(k), the benefits are likely to be even smaller given the current trend of U.S. deficit spending and likely future tax increases.
However, I do see why maxing out a 401(k) could make sense for behavioral reasons. The analysis above assumes that you can buy and hold assets in your taxable account for decades without touching the money. That’s easy in theory, but difficult in practice.
If you are someone who finds it hard to manage their own money, then the automation and illiquidity provided by a 401(k) could be a financial lifesaver. You won’t find these benefits in a spreadsheet, but they definitely matter.
Either way, the choice is yours. You have to make your own financial decisions. I just want to provide a useful framework for you when you do.
Lastly, don’t be so quick to write off someone else’s ideas just because you don’t agree with them. If I had done that with Kris Krohn’s TikTok video above, then I never would have written this post. I never would have learned something in the process and neither would you. Yes, Krohn gets lots of specific assumptions wrong, but his belief that “the benefits of a 401(k) are overstated” seems directionally right. His intuition is accurate, just not precise.
And if I had written him off based on what was wrong instead of trying to understand what might be right, then I would have failed to expand my financial knowledge. Remember, mocking someone is easy, but learning from them is hard. Happy investing and thank you for reading!
[Author’s note: This post was updated on 11/21/2021 to correct the estimate of the after-tax benefit of maxing out a 401(k). The original estimates were lower than the revised ones (0.34% vs. 0.73%), but I believe that this does not materially change the conclusion.]
If you liked this post, consider signing up for my newsletter.
This is post 210. Any code I have related to this post can be found here with the same numbering: https://github.com/nmaggiulli/of-dollars-and-data