What Worked for Fund Investors? Pinching Pennies and Letting Winners Run

I recently wrote about how funds' pre-fee returns did a remarkably good job predicting their subsequent performance. On average, the best past performers went on to earn higher pre-fee returns than their average peer while the lesser performers continued to lag.

That article elicited more feedback than I’ve ever gotten. Among the questions I received was: What about fees? Don’t they predict differences in funds' future performance? Excellent question! In 2016, my colleague Russel Kinnel published a landmark study that found, of all the variables he tested, expense ratios did the best job of predicting funds' subsequent returns. The lower the fees were, the better funds tended to do versus peers over subsequent periods and vice versa.

A Fresh Look

Given that, I took another pass at the data, this time sorting stock and bond funds based on the expense ratios they charged at the relevant times over the past 20 years ended Jan. 31, 2025. I grouped them as follows: “Cheapest” (the funds that were in the lowest decile by expense ratio in their category), “2nd” (the next 22.5%), “3rd” (the middle 35%), “4th” (the next 22.5%), and “Priciest” (most expensive 10%). Then I tracked each group’s subsequent net-of-fee returns.

What I found buttresses Russ’ original findings and subsequent research he’s done on the topic: Expenses excelled at predicting funds' performance. To illustrate, here are funds' forward average excess net returns (versus their average peer) over all rolling five-year periods between Jan. 1, 2005, and Dec. 31, 2024, sorted by fee grouping.

Not only did cheaper funds outperform, on average, the sort from low-cost to high-cost followed an almost perfect stair-step pattern. That pattern held when I measured stock and bond funds separately. (I excluded allocation funds because I don’t have access to a historical expense ratio series that’s inclusive of acquired fund fees.) Moreover, the cheapest funds subsequently outperformed the priciest funds, on average, in every single year that spanned the study.

What About Past Performance?

So, in summary, past pre-fee returns seemed to forecast future pre-fee returns, and expense ratios were a strong predictor of subsequent net returns. What if you were to combine the two by grouping funds based both on their past pre-fee performance and their cost? Here’s what it looked like:

The whole exceeded the sum of its parts: If you focused on the cheapest funds and favored the strongest past performers (before fees), you’d have earned a higher subsequent excess return, on average, than otherwise. Likewise, if you zeroed in on the strongest past performers (before fees) and tilted toward the lowest-cost funds, you’d have done better after fees than if you relied on past performance alone.

Takeaways

Investors choosing funds should start with fees, favoring those that levy lower expense ratios than peers. Russ’ 2016 research made that clear, and these more recent findings seem only to reinforce the case he originally made to pinch pennies.

That said, it didn’t hurt to factor in past performance, too. Even when we controlled for funds’ expense ratio rankings, it was evident that funds that had generated the best past returns before fees tended to deliver the best future returns, on average, both before and after fees.

Switched On

Here are other things I’m reading, listening to, and watching:

• Amy Arnott on how tactical-allocation funds failed yet again; Ben Carlson observes “you have to be right twice” to successfully time the market
• Christine Benz on what to do when your required minimum distribution exceeds your safe withdrawal rate in retirement
• Dan Sotiroff on Vanguard’s recent round of fee cuts
• “More than 60% of single-family houses in the United States were built more than 35 years ago”: Lawrence Hamtil on the implications of an aged housing stock
• Vanguard on the optimal way to rebalance target-date funds to adhere to the target asset mix while reducing transaction costs
• “Change, compression, maps, and mess”—Tom Brakke’s fortnightly roundup of writings in the worlds of investing and finance
• Tadas Viskanta’s indispensable “Abnormal Returns” blog
• The Jesus Lizard “Hide and Seek”; Dave Nadig’s Spotify mix
• The transcendent “Bill and Frank” episode (three) of The Last of Us

Feed Me!

I love hearing from you. Have some feedback? An angle for an article? Email me at jeffrey.ptak@morningstar.com. If you’re so inclined, you can also follow me on Twitter/X at @syouth1, and I do some odds-and-ends writing on a Substack called Basis Pointing.