ETFs have yet to prove that they can beat active funds

The performance of actively managed mutual funds has been intensely scrutinized in academic literature. The seminal Carhart study found that active mutual funds, on average, underperform the market after management fees and transaction costs.¹ This has resulted in a broad consensus among academics that investors should reallocate from actively managed funds to passive solutions of the broad market portfolio.² To this end, passive investing has been facilitated and popularized by the introduction of ETFs.

The vast majority of ETFs track indices that represent active strategies

However, we believe it is premature to render conventional mutual funds obsolete. For one, not all ETFs have low costs. While the cheapest ETFs have annual expense ratios below 0.05%, others have expense ratios above 1%, making them more expensive than many mutual funds. Also, if the purpose of ETFs really was to facilitate passive investing, then just a few ETFs on the broad market portfolio would be needed, certainly not thousands. Actually, the vast majority of ETFs track indices that represent active strategies.

More importantly, not much is known yet about the realized performance of ETFs. Our research aims to fill this gap by analyzing the performance of a comprehensive, survivorship-bias-free sample of US equity ETFs. We start by examining the aggregate performance of the ETFs in our sample, similar to how the collective performance of conventional mutual funds has been evaluated in literature.

Looking at the combined performance of all ETFs allows us to assess how much the entire investment community has been better or worse off as a result of investing in ETFs. We show that the performance of ETFs is not as impressive as one might expect it to be. Investors in these ETFs have collectively realized a performance that does not appear to be much different from the performance that can be expected from conventional actively managed mutual funds.

We also perform textual and statistical analyses to sort ETFs into common investment styles, such as size, value, momentum, quality and low risk. Our research highlights that none of these smart beta ETFs has managed to consistently add value relative to a capitalization-weighted market portfolio of all US stocks. But this can be partly attributed to the generally poor performance of equity factors over the most recent part of the sample period. Conversely, anti-factor ETFs – funds with negative exposures to these styles – have either lagged the market or delivered performance that was not very different from that of the market.

Overall, we conclude that the perceived superiority of ETFs finds little empirical support in the data and that ETFs have yet to prove that they can generate better performance than conventional actively managed funds.

1 Carhart, M. 1997. “On Persistence in Mutual Fund Performance.” The Journal of Finance 52 (1): 57–82.
2 French, K. 2008. “The Cost of Active Investing.” The Journal of Finance 63 (4): 1537–1573.