ETFs vs Theoretical Indexes

The factor returns published in the Kenneth R. French Data Library, including every series shown on this dashboard, describe theoretical portfolios. They are gross of costs, they involve short positions assumed to be free to establish and hold, and they rebalance on a fixed schedule at closing prices. No investor has ever received these returns. This article accounts for the gap between the published premium and what actually reaches an account, step by step, because that gap is where most disappointment with factor investing originates.

Step one: long-only implementation

Academic factors are long-short. HML is long value and short growth. Nearly every retail factor product is long-only, which means it captures the long leg and forgoes the short leg entirely. Since a meaningful portion of the historical spread often came from the short side underperforming rather than the long side outperforming, a long-only fund structurally captures only part of the factor.

A long-only value fund also retains full market exposure. Its return is dominated by what the market did, with the value tilt as a comparatively small overlay. Investors frequently attribute a bad year to the factor when the market was the dominant driver.

Step two: dilution of the tilt

Academic sorts often compare extreme groups, such as the cheapest third against the most expensive third, and weight within them without regard to tradability. A commercial fund cannot do this. It must stay diversified, respect capacity limits, avoid holding so much of any small company that it cannot exit, and usually track a published index with its own construction rules.

The result is a portfolio whose average characteristics are meaningfully milder than the theoretical sort. Two funds both labelled "value" can differ substantially in how much value exposure they actually carry, which is measurable and is the subject of the article on analysing a real portfolio.

Step three: explicit and implicit costs

• Expense ratio. The visible cost, charged continuously. Factor products typically cost more than broad market index funds.
• Trading costs inside the fund. Spreads and market impact from rebalancing. These do not appear in the expense ratio; they show up as a drag on returns. High-turnover strategies such as momentum are affected most.
• Bid-ask spread on your own trades. Paid on the way in and the way out, and wider for less heavily traded funds.
• Premium or discount to net asset value. An ETF's price can diverge from the value of its holdings, most notably during market stress, which is when investors are most likely to trade.
• Taxes. In a taxable account, distributions and realised gains are frequently the largest single cost. ETFs are generally more tax-efficient than mutual funds, but a high-turnover factor strategy is still less efficient than a buy-and-hold index fund.

Step four: tracking difference and index changes

Even against its own benchmark a fund will not match exactly. Sampling rather than full replication, cash held for redemptions, securities-lending revenue that offsets costs, and timing differences around index reconstitution all contribute. Tracking difference is worth checking directly, since it captures the total realised gap rather than the advertised expense ratio alone.

Index reconstitution deserves particular attention. When rules are published in advance and many funds must trade the same names on the same date, other participants can anticipate those trades. The cost is borne by the fund and therefore by its holders, and it is invisible in any published expense figure.

Step five: the behaviour gap

The final and often largest gap is between the fund's return and the return its investors actually experience. Money tends to arrive after good performance and leave after bad, so the average dollar earns less than the average year. Factor strategies are especially exposed to this because their underperformance can run for years, which is far longer than most investors expect when they buy.

This is not a cost a fund can fix, and it is the one most reliably underestimated at the point of purchase.

How to use the data on this site sensibly

Treat the series here as a research baseline describing the underlying phenomenon, not as an achievable return. When evaluating a product, compare it against a realistic long-only implementation of the same idea rather than against the academic factor, check the fund's actual factor loadings instead of trusting its name, and include the tax consequences in a taxable account. A strategy that looks compelling gross of all of this can be unremarkable net of it.

This article is educational and is not investment advice or a recommendation of any particular fund.