Using active share in conjunction with tracking error provides a more complete picture of active management.
In 2009, Martijn Cremers and Antti Petajisto created a tool for assessing just how active an equity manager is relative to his benchmark. Active share tries to more precisely identify which managers are differentiating themselves from their benchmarks versus those who might be closet indexing. (Morningstar.com does not currently publish active share as a data point, but it may appear down the road.) Active share measures the percentage of a fund's holdings that differ from the benchmark's weightings.
Cremers and Petajisto's research showed that high active share is perhaps a necessary--but not necessarily sufficient--precondition for outperformance. Our own findings largely corroborated Cremers and Petajisto's results. My colleague Russ Kinnel pointed out, though, that active share is more meaningful when considered relative to a fund's category peers. That's because most small-cap funds, for instance, have high active share (say greater than 70%) in an absolute sense because the universe is so broad. It's easier to see which funds are truly differentiated by making comparisons within a peer group.
However, active share by itself doesn't tell the full story. It's a good starting point, but it doesn't capture all the ways in which a manager may be active. Specifically, active share doesn't include so-called factor bets, such as overweighting a given sector relative to the index or holding lots of cash. This is where tracking error comes in. Tracking error measures the volatility that is not explained by movements in the benchmark. As a sample, the 72 actively managed large-blend funds that we cover have a five-year tracking error of nearly 5% relative to the S&P 500 Index.
Using active share and tracking error in conjunction reveals more of a fund's character than either could alone, as Petajisto recently explained in a paper. That is, a fund might be more active than active share alone might indicate. For instance, a fund might have a fairly low active share, but high tracking error. Active share would suggest a closet indexer in this case, but high tracking error implies active positioning of some kind relative to the benchmark.
Take Fidelity Dividend Growth FDGFX. The fund's latest active share is just 50% versus the S&P 500 Index. That's well below the 73% average of the active large-blend funds that we cover. But the fund's 7.4% tracking error over the past five years is comfortably ahead of the 4.9% peer average. Its relatively high tracking error stems in part from the fund's foreign equity stake, which has been more than 15% of the portfolio for the past three years.
At the other end of the spectrum is Davis NY Venture NYVTX. Its 3.6% five-year tracking error is on the low side, but its active share has been greater than 75% during that stretch, growing to 83% recently. Davis NY Venture fits the profile for what Petajisto calls a diversified stock-picker, or a fund that competes based on stock selection. Bottom-up managers Chris Davis and Ken Feinberg certainly fit that description. The fund has become more concentrated in recent years, with the number of holdings falling to 75 or so from 100 five years ago, which has probably contributed to its increasing active share.
The renegades, such as Longleaf Partners LLPFX, are funds that score high on both active share and tracking error. These tend to be concentrated funds, and Longleaf itself has just 18 holdings with more than half its assets in its top 10. This has led to among the highest scores in our large-blend subset for both active share and five-year tracking error, 96% and 11.2%, respectively. Not surprisingly, concentrated funds such as Longleaf tend to be more volatile than most peers. But it's worth pointing out that this isn't always the case.
As I mentioned in a Fund Spy last year, some funds with high active share are actually quite defensive. Again, these funds are the exception among the high active share/high tracking error group. They tend to look at risk in absolute rather than relative terms. That is, they define risk more as capital loss than as lagging a benchmark. First Eagle US Value FEVAX, one of the examples from that article, scores high on both metrics with 88% active share and 6.3% tracking error. Yet it consistently has been less volatile than most peers. Its frequently big cash stakes (currently 15.7%) attest to its capital-preservation focus. Weitz Value WVALX, which has current active share of 90% and a tracking error of 5.5%, places even greater emphasis on capital preservation these days given its nearly 27% cash reserve.
For Context, Not Decision-Making
Active share and tracking error are useful tools for uncovering new insights about actively managed funds. We plan to revisit active share, in particular, in the coming months. But it's also important to understand the limitations of such tools. They can provide context, but they shouldn't be the basis for decisions by themselves. Active share, for example, could be treated as a necessary but not sufficient criterion when looking for an actively managed equity fund. Investors still need to do their homework and look into what might be driving a given fund's active share or tracking error. As always, understanding what's behind a data point remains critical in determining whether a given fund is the best fit for the job.