World Global Strategy

4 November 2009

Popular Delusions
Too much love will kill you: sector overcoverage and underperformance
Dylan Grice (44) 20 7762 5872 dylan.grice@sgcib.com

A trader I once knew used to say that when it came to taking risk, if it felt right it was probably wrong. Herding is a good example. It feels right because the act of conforming activates the brains reward system, releasing dopamine. But its nearly always wrong. The sell-side are as hardwired to herding as anyone else. The sectors most loved by them underperform those which are most neglected by them.

In 1895, Gustav le Bon wrote that the substitution of the unconscious action of crowds for the conscious activity of individuals is one of the principal characteristics of the present age. Shouldnt we have learned the dangers of groupthink by now? Probably not. Earlier this year, and quite disconcertingly, neuroscientists in the Netherlands led by Vasily Klucharev demonstrated that the neural mechanism which allows us to learn is the very same neural mechanism which causes us to herd! 1 We learn by making predictions and comparing outcomes to those predictions. When the predictions are correct the reward system of the brain is activated and dopamine is released by the nucleus accumbens (the brains pleasure zone). But when the predictions are incorrect, a prediction error signal deactivates the pleasure zone. This process of reward and punishment guides the continuous evaluation of our environment and steers us towards the gradual accumulation of knowledge. But what Klucharev showed was that, when we dont conform to the group, our brains flash the same prediction error signal we receive when weve failed to judge an outcome correctly. When it comes to developing our ability to think outside the crowd, were effectively hardwired with learning difficulties.
The BAAH index and sell-side herding: number of sector analysts per $bn sector market cap
4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 More herding = less opportunity Less herding = more opportunity

IMPORTANT: PLEASE READ DISCLOSURES AND DISCLAIMERS BEGINNING ON PAGE 4

0.0 Construction Clothing Textiles Printing Paper prods Transport Electronics Recreation M achinery Steel Retailers Chem icals Aerospace Food M fgs Utilities Autos Software Financials M iners Drugs & Health Consum er Drinks Energy Tobacco

www.sgresearch.socgen.com

Source: Factset, SG Global Strategy

See Reinforcement Learning Signal predicts Social Conformity, Vasily Klucharev et al (2009)

Popular Delusions

Lets draw a distinction between the herd and the crowd. To understand why the herding is almost certain to be wrong it helps to understand why the crowd is often right. James Surowiecki expressed it simply and elegantly in his 2004 book 2 which starts with the premise that any opinion contains only two fundamental parts: correct information and error. To get a groups expectation we aggregate across its individual members opinions. In doing so we add together all of those individual pieces of correct information with all of those individual errors. But if the errors are genuinely random and unconnected with one another, adding them up will simply cause them all to cancel out (statistically, random errors are expected to be zero). In this case the crowds best guess then consists completely of correct information. Surowiecki used this framework to explain Victorian polymath Francis Galtons surprise when he observed that the average guess at the weight of an ox made by a crowd at a country fair was better than any of individual guesses. Indeed, it even beat the best guesses of independent cattle experts. In case youre wondering, the weight of the ox was 1,198 pounds and the average of the guesses which beat all other individual guesses was 1,197. So crowds can be very right. But to see why herding is nearly always very wrong lets suppose someone in the crowd believes something silly, like all oxen weigh 2,000 pounds. With 787 people in the crowd Galton observed, this deviant view would clearly make little difference to the groups estimate. But what would happen if everyone held the deviant view? Suppose someone had a very convincing explanation for his theory that oxen could only weigh 2,000 pounds and managed to convince everyone of it. Coming from the same source, the errors are now all correlated with one another and so with the crowd herding around a mistaken belief the estimate will be very wrong. This is the herding around the popular delusions which so fascinated Charles Mackay, the mob madness which lies behind famous occurrences such as the Crusades, the Witch Hunts or the Tulipmania to name but a few. And its why contrarian investing works. But how do we know what the herd thinks? Social indicators, such as what everyone is saying are popular, but as George Soros said now that the contrarian viewpoint has become the prevailing bias, I have become a confirmed anti-contrarian. It was with all of this in mind that I read Charles De Boissezon, one of our Financial Engineering teams excellent analysis of sell-side coverage of stocks in Europe. He calculated the extent to which a stock was overbroked by dividing the number of analysts covering a stock to its market cap, and uncovered such gems as Nokia being the most covered stock with 63 analysts and still it can come up with shocking numbers and Q-Cells (1bn free float) being covered by 44 analysts. Hours of fun But it got me wondering whether or not this might be a good proxy for measuring herding within the sell-side research management community. The obvious motivation for hiring a sector team is that there is plenty of commission to be won. Indeed, the following chart shows analyst coverage of a sector rises proportionately with its market capitalisation. But what you can also see from the chart is that, although the scatter shows a good fit, its not perfect. Some sectors are above the trend line, suggesting they have fewer analysts than might be expected given their market caps, and some below it. Could this difference be a proxy for herding?

See the Wisdom of Crowds: Why the Many Are Smarter Than the Few, by James Surowiecki (2004)

4 November 2009

Popular Delusions

Analyst sector coverage and market capitalisation

M arket cap ($ bn) 5,000 4,500 4,000 3,500 3,000 2,500 2,000 1 ,500 1 ,000 500 0 0
Source: SG Global Strategy

# analysts 1 ,000 2,000 3,000 4,000 5,000 6,000

So I took Charles analysis a little further. Lets call his ratio of coverage to market cap the Banks Aggregate Analyst Herding Index, or the BAAH Index for short (as in the noise bleating sheep make). I used Factset sector classifications with the FTSE World Index and got the results given on the front page. That chart shows herding is most pronounced in the clothing and construction sectors and least pronounced in energy and tobacco. Then, I backtested the strategy on Factset by ranking each sector according to its BAAH index (beginning in 1987) and calculating the subsequent years returns. I rebalanced the portfolio annually. It turns out that a long-short strategy of buying the sectors where herding is less dominant and shorting those where coverage is deepest would have generated around 7% p.a. since 1987. The cumulative results of the strategy are shown in the chart below. For a given market cap, it seems, the more analysts you see swarming around a sector the more wary you should be of being able to extract alpha from it. Leave the herd to its dopamine fix and have another think about the less-crowded sectors.
Cumulative performance of under-covered sectors versus over-covered sectors
6000

5000

4000 Low herding sectors 3000 High herding sectors

2000

1000

0 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008

Source: SG Global Strategy

4 November 2009

Popular Delusions

IMPORTANT DISCLAIMER: The information herein is not intended to be an offer to buy or sell, or a solicitation of an offer to buy or sell, any securities and including any expression of opinion, has been obtained from or is based upon sources believed to be reliable but is not guaranteed as to accuracy or completeness although Socit Gnrale (SG) believe it to be clear, fair and not misleading. SG, and their affiliated companies in the SG Group, may from time to time deal in, profit from the trading of, hold or act as market-makers or act as advisers, brokers or bankers in relation to the securities, or derivatives thereof, of persons, firms or entities mentioned in this document or be represented on the board of such persons, firms or entities. Employees of SG, and their affiliated companies in the SG Group, or individuals connected to then, other than the authors of this report, may from time to time have a position in or be holding any of the investments or related investments mentioned in this document. Each author of this report is not permitted to trade in or hold any of the investments or related investments which are the subject of this document. SG and their affiliated companies in the SG Group are under no obligation to disclose or take account of this document when advising or dealing with or for their customers. The views of SG reflected in this document may change without notice. To the maximum extent possible at law, SG does not accept any liability whatsoever arising from the use of the material or information contained herein. This research document is not intended for use by or targeted at retail customers. Should a retail customer obtain a copy of this report they should not base their investment decisions solely on the basis of this document but must seek independent financial advice. Important notice: The circumstances in which materials provided by SG Fixed & Forex Research, SG Commodity Research, SG Convertible Research, SG Technical Research and SG Equity Derivatives Research have been produced are such (for example because of reporting or remuneration structures or the physical location of the author of the material) that it is not appropriate to characterise it as independent investment research as referred to in European MIF directive and that it should be treated as a marketing material even if it contains a research recommendation ( recommandation dinvestissement caractre promotionnel ). However, it must be made clear that all publications issued by SG will be clear, fair, and not misleading. Analyst Certification: Each author of this research report hereby certifies that (i) the views expressed in the research report accurately reflect his or her personal views about any and all of the subject securities or issuers and (ii) no part of his or her compensation was, is, or will be related, directly or indirectly, to the specific recommendations or views expressed in this report. Notice to French Investors: This publication is issued in France by or through Socit Gnrale ("SG") which is authorised by the CECEI and regulated by the AMF (Autorit des Marchs Financiers). Notice to UK investors: This publication is issued in the United Kingdom by or through Socit Gnrale ("SG") London Branch which is regulated by the Financial Services Authority ("FSA") for the conduct of its UK business. Notice To US Investors: This report is intended only for major US institutional investors pursuant to SEC Rule 15a-6. Any US person wishing to discuss this report or effect transactions in any security discussed herein should do so with or through SG Americas Securities, LLC (SGAS) 1221 Avenue of the Americas, New York, NY 10020. (212)-278-6000. THIS RESEARCH REPORT IS PRODUCED BY SOCIETE GENERALE AND NOT SGAS. Notice to Japanese Investors: This report is distributed in Japan by Socit Gnrale Securities (North Pacific) Ltd., Tokyo Branch, which is regulated by the Financial Services Agency of Japan. The products mentioned in this report may not be eligible for sale in Japan and they may not be suitable for all types of investors. Notice to Australian Investors: Socit Gnrale Australia Branch (ABN 71 092 516 286) (SG) takes responsibility for publishing this document. SG holds an AFSL no. 236651 issued under the Corporations Act 2001 (Cth) ("Act"). The information contained in this newsletter is only directed to recipients who are wholesale clients as defined under the Act. IMPORTANT DISCLOSURES: Please refer to our website: http:\\www.sgresearch.socgen.com http://www.sgcib.com. Copyright: The Socit Gnrale Group 2009. All rights reserved.

4 November 2009