The case for active management weakens

Can active equity managers log better performance by being more active? An influential study published six years ago suggested that they could. That sparked a mini-revival for the beleaguered active management industry which, as regular readers will know, has been losing out to passive managers for a while. The key, it appeared, was to have the courage to divert from benchmark indices and take high-conviction, concentrated bets.

The logic was attractive. But now the debate has taken a new turn, as that research has come under serious attack.

The original idea, from the academics Martijn Cremers and Antti Petajisto, was to measure a fund's "active share". This is the proportion of a fund's portfolio that differs from the index to which it was benchmarked. A well-run index fund would have an active share of zero. An eclectic fund that held no stocks at all in the index to which it was benchmarked would have an active share of 100.

This research had two implications. First, it provided a simple way to spot "closet indexers", who were nominally active but in reality shadowed the index. There is no point in paying fees for active management to such people, when an index fund can be bought far more cheaply.

But secondly, it suggested a way forward for active managers, because it turned out that there was a marked correlation between active share and performance. The higher the active share, the better the performance, and the greater the chance of beating the benchmark.

This implied that there was every point in paying money to someone prepared to make concentrated bets.

To those who complain that ultimately all fund managers must sum up to the market, bear in mind that not all stock market money is held by institutions, and in most years the average active manager beats the benchmark slightly before fees - and loses to it slightly after fees. Active is close to a "zero-sum" game, but there is an opportunity for more to win than to lose.

So this research offered a lifeline justification for active managers, who are, after all, vital to the job of price discovery. Some pension funds, for example, insist that managers bidding for a mandate should reveal their active share, and offer an explanation if it is lower than 75. Others voluntarily publish their active share. And, full disclosure, I argued in a book in 2010 that managers should be required to publish active share. This, I thought, would be a defence against closet indexing, which was helping to create asset bubbles.

But now comes the riposte from researchers at AQR, a New York-based fund manager, who went to the Cremers/Petajisto data-set and crunched it again. They found that the apparent outperformance of active managers could be attributed to the small-size effect.

They divided funds according to the benchmark they tracked, and found that those with the highest active share were overwhelmingly those that tracked small-cap indices. This makes sense. There are more stocks in small-cap indices; managers tracing large-cap indices, with well-researched stocks, will inevitably be closer to the index.

Sorting funds on their active share turned out to be another way of sorting them according to their benchmarks. Once the AQR team produced results for each of the 17 benchmark indices used in the Cremers and Petajisto study, they found that high active share funds outperformed closet indexers in eight, but lost to them in nine.

As small-cap stocks fared well over the period of the study, from 1990 to 2009, high active share funds overall fared best. But it did not follow from this that managers with a high active share had a stronger chance than anyone else of outperforming their own benchmark.

What is left of the active share? Its use as a measure to work out whether an "active" manager should be charging high fees is undimmed. As well as the composition of their portfolios, funds might also be required to publish their tracking error - the extent to which their returns differ from index returns.

But the seductive finding that small concentrated bets had a stronger chance to outperform is badly damaged. Instead this research would point more towards "smart beta" strategies - which cynics will point out is what AQR offers. These take a benchmark index but weight it differently, screening for cheap stocks, or high dividends, or whatever. Such strategies have a good chance to outperform, albeit not by much. They will have low active share.

Provided they charge low fees, such low active-share management can make sense.

High-conviction concentrated bets have a better chance to outperform by a big margin. But high active-share on its own brings just as great a risk of underperformance. Managers who use such a strategy have to demonstrate that they have skill - and that is difficult. The case for conventional active management looks that much weaker.

[email protected]

© The Financial Times Limited 2015. All rights reserved.
FT and Financial Times are trademarks of the Financial Times Ltd.
Not to be redistributed, copied or modified in any way.
Euro2day.gr is solely responsible for providing this translation and the Financial Times Limited does not accept any liability for the accuracy or quality of the translation