The small-cap premium is one of the oldest return anomalies (or risk factors, if you're so inclined) identified in the research literature. But after decades of analysis and mixed results, the idea that small company shares will outperform large caps also draws plenty of skepticism.

Earlier this year, for instance, Morningstar asked: Does the Small-Cap Premium Exist? Even if it does, it's “unreliable,” advised an analyst in January. “Investors should not count on a small-cap tilt as a way to boost long-term performance.” Raising doubts about this corner of the market isn't new—a 1999 study in The Journal of Finance, as one example, finds that the size effect evaporates after correcting for a “delisting bias.”

Have we been hoodwinked? Perhaps, although the degree of skepticism seems to be a cyclical affair. Casting aspersions on the strategy resonates whenever smaller stocks stumble in relative and/or absolute terms, which describes recent history. How does the small-cap premium compare over the longer term? Minds will differ, for the simple reason that we can torture the data to force any confession we choose. Tell me the outcome you're looking for and I'll provide the appropriate historical window.

That said, the standard performance summaries through yesterday (Dec. 10) don't look encouraging, based on numbers published by Russell Investments. Across several trailing periods, the Russell 2000 Index (a popular US small-cap benchmark) trailed and/or matched large caps (Russell 1000).

But before we close the door on this risk factor, consider how a real-world example via ETFs compares. As a simple test, let's measure the spread for the daily return for the iShares Russell 2000 (IWM) less the iShares Russell 1000 (IWB) and use that as a basis for a buy-and-hold strategy with a start date as of the Dec. 29, 2000. If there's a cumulative advantage to small caps, this spread strategy will rise through time; if large caps prevail, the strategy will lose money. As you can see in the chart below, rumors of the small-cap premium's death appear to be premature. This spread strategy is up roughly 50% since the end of 2000, although it's been fading in recent months.