Skip to Content
Rekenthaler Report

Can $155,000 in 401(k) Fees Be Correct?

Yes--but that doesn't tell us much.

The Great Delicatessen Scandal
You're probably seen the articles. A typical American household will end up paying $155,000 in lifetime 401(k) fees. The figure comes from a study called "The Retirement Savings Drain: Hidden & Excessive Costs of 401(k)s," released in May 2012 by Demos, a New York-based think tank, and it's been cited in dozens of mainstream reports, among them NBC News, Forbes, PBS, and U.S. News & World Report. Never with the intent to compliment. One headline: "401(k) Fees Are Robbing You Blind." Another: "401k Expenses Still a Pernicious Rip-Off."

In other words, the study was a communications success. It was not, however, convincing as financial analysis. 

First, a lifetime rollup number is created to impress, not to instruct. None of us has the faintest idea if $155,000 is twice as high as it should be, or three times too high. It could even be too low. It just sounds like a lot, you know.

I can play that game, too.

My wife and I buy a cup of coffee each workday from a local deli. We're disciplined: it's only one cup, only on workdays, and always just the $2 basic cup, none of the frilly drinks. We're 25 years old (for the purposes of this exercise), we'll live for another 60 years, and inflation will grow at a modest 2% annually. The lifetime cost for those daily coffees? $114,000.

In addition, during the 40 years of our working lives, from ages to 25 to 65, we each pay $7 for lunch at that same deli, getting a salad, bread, and a drink. We could instead brown-bag it, which would cost us $3 per meal, but we prefer the freshness of the salad. Our lifetime cost for not packing lunches? An additional $121,000 (again, this figure includes the effect of 2% annual inflation). Between that decision and the morning coffee, we are out $235,000.

From the looks of it, that deli owner has cost us a big chunk of our retirement. She also must be a multimillionaire, as she's collected $235,000 just from us--and we're one of hundreds of her daily customers.

Was that exercise useful? Have you learned more about the true cost of coffee and of eating lunch? Should I circulate this study so that others across the country know the damage that their local delis are doing to them?

Come on. 

Say What? 
Second, even if the study's figure permitted easy interpretation, the derivation of that number would not. Calculating 401(k) balances, and then the fees charged upon them, requires a raft of assumptions. My coffee/lunch example had the virtue of being easy to understand. The final total of $235,000 might be head-scratching, but the extra $4 each workday for the coffees and the extra $8 for the lunches is clear. Not so with 401(k) fees. A mainstream journalist, writing for a mainstream audience, will have great difficulty highlighting and explaining the study's critical choices.  

This shows in the descriptions of the study's findings. One article states that an American household "could spend as much as $155,000" in 401(k) fees; another that the "average American couple pay [sic] nearly $155,000 in 401(k) fees;" a third that an "ordinary, median-income two-earner household will pay nearly $155,000 over the course of their lifetime in 401(k) fees."

Three articles, three different explanations. The first article states the number as a maximum, that an American household "could spend as much as $155,000" in 401(k) fees. The next is written in the current tense, implying that these are today's dollars for those who retire now. The third is written in the future tense, suggesting that these are future dollars. They can't all be correct. In fact, only one can be correct. 

For the record, these were the study's assumptions:

  1. Two wage earners
  2. 40-year working career (ages 25 to 65)
  3. Starting date of 1965
  4. Initial savings rate of 5%, rising to 8%
  5. The median U.S. income
  6. No employer contributions
  7. No plan fees
  8. Half of assets invested in stock funds and half in bond funds
  9. Average current asset-weighted expense ratio for stock funds (0.95%)
  10. Average current asset-weighted expense ratio for bond funds (0.72%)
  11. Fund trading costs that equal the funds' official expense ratios
  12. Actual stock- and bond-market returns (before costs)

The second article wins the prize. The study is not intended as a maximum, and it uses something approaching current dollars, rather dollars from a projected retirement 40 years from today. 

In some ways, the study is quite reasonable. The author certainly could have made decisions to increase the $155,000 total. He could have used future dollars. He could have assumed a higher savings rate, or a company match. He could have modeled plan fees. By no means was the study rigged to come up with the highest possible defensible number (which is what I had suspected before reading the paper).

Hidden and Excessive?
But ... those key claims of "hidden and excessive," the strong words that appear in the article's title, are unsupported. The fees are hidden, says the report, because: 1) Employees don't much understand them; 2) Fund fees are taken quietly from asset management charges rather than explicitly billed; and 3) Besides their official expenses, funds carry high trading costs that go unreported.

The answers: 1) That doesn't make the fees hidden, plus recent Department of Labor legislation mandates additional disclosure; 2) That ain't news; and 3) You'd better prove that. 

On item No. 3, fund trading costs are a subject worthy of a full column (or two), but suffice it to say that the average fund's trading cost is either much lower than this assumption, or the average fund manager has the skill to overcome this hurdle. Either way, the point is that active funds tend to lag the market averages by roughly the size of their official expense ratios. There is no extra gap that needs to be explained--nor modeled in a fee exercise.

That decision alone is worth half the $155,000; without it, the headline instead would read $78,000. 

Is $78,000 excessive? Or for that matter $155,000? The study doesn't make that case. It can't. As I wrote, that number resists interpretation.

In summary, the report was very effective at its job of sparking criticism of 401(k) costs, and its calculations were mostly reasonable. However, it demonstrated neither hidden nor excessive fees. Nor did it yield a conclusion that could be put to use. 

John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.

Sponsor Center