Performance Trickery, part 3

Success is hard to come by for active managers, as readers of our SPIVA reports know well.  Sometimes what appears to be stock selection skill is in fact simply a byproduct of style drift across the capitalization scale.

A majority of large-cap active managers outperformed the S&P 500 only 3 times in 19 years of SPIVA data.  It’s surprising, therefore, to see that 68% of midcap managers outperformed the S&P MidCap 400 Index last year (and 62% of small cap managers outperformed the S&P SmallCap 600).  In fact, a majority of midcap managers outperformed in each of the past three years.  In those same years, an average of 66% of large cap managers underperformed the S&P 500.

Why are the results so good for mid-cap managers?  It’s possible that smaller stocks are less efficiently priced than the largest companies, so that stock selection in the mid- and small-cap universe is easier; the higher dispersion of midcaps and small caps would support that view.  Or perhaps mid- and small-cap managers are simply more skillful than their large-cap counterparts.

These explanations would be more plausible were it not for the long-term record.  For the 10 years ended December 31, 2019, for instance, 84% of mid-cap and 89% of small-cap managers lagged their index benchmarks.  If smaller companies were an easier game with more astute players, the score would be better than this.

We argue instead that style drift – for example, the ability of a midcap manager to buy large cap stocks – can shed light on mid- and small-cap managers’ success.  In 2019, e.g., the S&P 500 rose by 31.5%, well ahead of the 26.2% gain of the S&P MidCap 400.  A mid-cap manager who tilted his portfolio slightly up the capitalization scale might have been rewarded for doing so.  If our conjecture is correct, we’d expect midcap manager performance to improve whenever the S&P 500 outperforms the S&P 400.  Exhibit 1 shows that to be the case.

Exhibit 1.  Style Drift Helps to Drive Midcap Performance

Source: S&P Dow Jones Indices. Data from Dec. 31, 2000 through Dec. 31, 2019. Table is provided for illustrative purposes. Past performance is no guarantee of future results.

A majority of midcap managers outperformed the S&P 400 six times, for an overall success rate of 32%.  When the S&P 500 beat the S&P 400, however, the managers’ success rate rose to 57% (4/7), vs. a success rate of only 17% (2/12) when the S&P 500 lagged.  Alternatively viewed, if the majority of midcap managers outperformed, the 500 usually beat the 400; if the majority underperformed, the 400 was typically in the lead.

Logically, if midcap underperformance creates an opportunity for midcap managers, midcap outperformance might create an opportunity for large cap managers.  Exhibit 2 shows that it does.

Exhibit 2.  Style Drift Helps to Drive Large Cap Performance

Source: S&P Dow Jones Indices. Data from Dec. 31, 2000 through Dec. 31, 2019. Table is provided for illustrative purposes. Past performance is no guarantee of future results.

Midcaps beat large caps in each of the three years when the majority of large cap managers outperformed, suggesting that large-cap managers might have augmented their performance by tilting down the cap scale.  There were no years when the S&P 500 beat the MidCap 400 and the majority of large cap managers outperformed.

Whenever there are significant differences between the performance of capitalization-specific indices, there are opportunities for managers to add value by moving up or down the cap scale.  So far in 2020, of course, the S&P 500 is well ahead of its smaller counterparts; mid- or small-cap managers might well benefit in next year’s SPIVA data.  Such opportunistic moves may be commendable, but they are not evidence of skill at stock selection.

The posts on this blog are opinions, not advice. Please read our disclaimers.

Leave a Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>