Active fund managers are closet index huggers; Study finds ‘closet indexing’ is rife in US active funds

March 12, 2014 10:39 am

Active fund managers are closet index huggers

By John Authers

Study finds ‘closet indexing’ is rife in US active funds

How active is your fund manager, and what chance do they really have of beating the market?

These questions are linked. A crude measure of how much fund managers deviate from their benchmark can help predict which funds will outperform. That measure is “active share” – the percentage of a fund’s portfolio that differs from its benchmark. Thus a well-managed index fund will have an active share of 0 and an esoteric fund that holds no stocks in its index has an active share of 100.

The concept was popularised by academics Martijn Cremers and Antti Petajisto, who made two important discoveries. First, “closet indexing”, where funds limit the risk of underperformance by clinging to the index, is rife in the US. Second, the higher a fund’s active share, the stronger its chance of beating its benchmark.

This theory aligns perfectly with common sense. To paraphrase Mr Petajisto (who has since taken a job at BlackRock but completed his research while working at Yale and New York University), funds that charge a fee for active management but then barely vary from the index make it hard for themselves to justify that fee.

This research is profoundly unhelpful for much of the traditional regulated fund business in the US and Europe, and an effort by the industry is now on to discredit the concept.

Deviants outperform

This month, Fidelity Investments published a research paper saying that active share had been “misunderstood”. It made two particularly damning points. First, the tendency for high active share funds to outperform was driven in large part by a “size effect”. Over time, smaller cap stocks tend to outperform large-caps. Indices are based on market capitalisation. Virtually by definition therefore, funds with a high active share will have fewer large-caps and relatively more small-caps.

More video

Second, Fidelity points out that it should not be used in isolation. Even if active share is coupled with higher returns on average it is also – again in line with common sense – coupled with higher risks to the downside, and with greater total dispersion of returns. This is why managers hug the index in the first place – to reduce the risk of an embarrassing loss.

It makes other points. Some funds are compared to an inappropriate benchmark. And small-cap funds tend to have a higher active share, just because they have more funds to choose from. And high active share can show “style drift” – where managers invest in sectors outside their mandate. Fidelity’s conclusion: “Investors should be wary of trying to make precise distinctions about manager skill or return potential using active share alone.”

These points are valid as far as they go. But that is not very far. Yes, the size effect is at work for active managers. But a paper that Mr Petajisto published last year looked at results separately for large, mid- and small-cap managers. In each case, those who deviated most from their benchmarks were more likely to outperform.

And nobody is suggesting that active share can be used on its own to predict who will outperform. Identifying outperformers in advance with certainty remains impossible, but academics have accumulated a number of useful pointers. Smaller funds tend to outperform; bigger fund groups tend to outperform; lower fees always help; and funds do better when the managers are themselves invested in them. There is also a huge new literature in identifying skill independent of performance.

Finally, Fidelity fails to mention that its own funds are singled out for criticism, The modern actively managed mutual fund is in many ways a Fidelity invention, thanks to spectacular returns it generated in the 1970s and 1980s. But the Cremers/Petajisto research detailed how Magellan, the first mutual fund to exceed $100bn in assets, moved from an active share of more than 90 per cent in 1980 to only 33 per cent by 2000. This reflected a “conscious decision to become a closet indexer” according to the academics. The performance? From 1979 to 1989, Magellan beat the index by 150 per cent; from 1996 to 2005 it lagged behind the index by 1 per cent per year.

Other funds have managed to maintain outperformance even as their active share reduces over time, but Fidelity remains the home of a great cautionary tale against the dangers of closet indexing.

Fee payers fight back

How far could this concept go? Recent events in Scandinavia suggest that large fund managers in the US should try to keep the genie of “active share” in its bottle. Denmark’s securities regulator announced late last year that it was consulting on regulations to force funds to publish their active share figures. Those below a certain active share would be barred from marketing themselves as “active” or from charging accordingly.

Sweden’s Swedbank Robur, the fund management arm of the country’s biggest bank, may have discovered the limits to candour. It has used active share as part of its investment process since 2007, according to its English language website. In November last year, its chief executive candidly admitted that two of its biggest funds had hugged so close to the index that its chances of success were “extremely low”.

In response, Sweden’s shareholders’ association announced that it was launching a lawsuit, alleging that savers had been made to pay for something – active management and a chance to beat the index – that they did not receive.

This is an intriguing principle. US fund managers must be hoping that no class action lawyers try to make the same claim in US courts.

Against such a background, it is not surprising that large fund management groups are attempting to discredit “active share”. It certainly has its limitations and should not be used in isolation.

But it rams home a message all investors should understand. There is no eminent point in paying a small fee for a fund that merely matches an index (which need not be based on market capitalisations), or in spending good money to unleash an active manager to attempt to find performance, matched against an absolute return benchmark, and perhaps being allowed to use short selling or derivatives. There is little or no case for what lies in between – and that middle ground includes much of the traditional regulated fund management industry in the US.