Smart beta investing and the seduction of labels

Smart beta is an impressive investment branding story. The name neatly encapsulates the idea of beating conventional indices consistently, but with little effort and much lower costs. Intellectually seductive, it conveys the concept of market exposure that avoids the costs and behavioural failures of active managers. And simultaneously it gives the impression of an edge over most of the passive conventional index tracking strategies. If only the actual performance matched the image. But, the popularity of smart beta tells us even more about investment narratives, and the psychology of investors.

There are two clear issues for smart beta. The difficulty is proving a strategy on subsequent out of sample experience after it has been optimised against historic data. No matter over how long a historic period involved in developing a strategy, the real test is what happens after. Another key issue is the seductive appeal of individual words and narratives. Terms such as emerging markets, new energy and dynamic all convey the concept of buying the future and have an emotional appeal. It is this emotion that investors need to be alert to.

Stories are embedded in much of the presentation of smart or alternative beta. Many of the exchange traded funds focusing on this come with strongly emotive labels. Examples range from names like “Buyback Achievers”, to terms like “progressive” and “dynamic”. What’s not to like? Representativeness is a strong psychological bias, making it easy for investors to convince themselves they are buying a rosy future. It is a more acute version of the emerging markets syndrome; that growth must be a good investment. It triggers emotional responses much as hedge funds once did.

Representativeness can make us think that a growth area represents a good investment, much as we think that gold shares should behave like gold itself. Unconscious impressions and beliefs compromise our analysis. We can readily understand the argument that gold mining businesses should in some way be leveraged plays on the gold price. Yet, gold mining stocks have spent five years mainly not correlating. They look like they should be a play on the gold price, but have proved not to be. Similarly, we can imagine a future in which there are lots more solar panels, yet fail to calculate whether anyone will make money out of them.

Research suggests there is little linkage between share performance and economic growth, not least because valuation can change dramatically, as we see with the 75% fall in China’s stockmarket p/e over 10 years, broadly cancelling out earnings growth. And, it is worthwhile looking at the truly long term to get a perspective on economic growth. China may actually have been the largest global economy for all but one century of the last 20.
There is a danger in an investment strategy becoming simply a story. Investors are too easily seduced by phrases like “the China story still has a long way to run”.

The danger of alternative beta is not just that many of the strategies are created by data mining, and so are optimised against an experience that may not repeat. Certainly, this apparent endorsement of history gives specious credibility. But, the more insidious risk is that we are too readily taken in by a marketing concept that conveys the idea of systematically better investment performance with less risk and lower costs. It has created an excitement that the old brand of ‘factor tilts’ never had.

Thematic approaches to smart beta require investors to hold beliefs about the future. Some fund labels are essentially narratives in themselves, and deserve deeper examination. There is a danger in an investment strategy, or ETF, in becoming simply a story. Fund managers use stories to rationalise underperformance, but with smart beta the strategy itself is often the story, summed up in the label. Typically investors have made a deep emotional and long term commitment to it.

Professors Richard Taffler and David Tuckett highlight the importance of stories in re-affirming conviction amidst a sea of data. These are stories that managers tell themselves, tell colleagues and use in client reporting. Tuckett & Taffler also highlight the importance of meta-narratives, or overarching distillations of an investment philosophy. Linking information together in the form of a narrative not only assists recall, but also helps us to make sense of something. We try to incorporate our own life events in a personal narrative also. This preference for stories over otherwise apparently unrelated facts and possibilities creates a bias. Think of analyst reports that cover poor results or a setback with the phrase “poised for growth”, a story that moves us on from disappointing facts to hopes. We see these phrases and narratives so frequently, we can easily miss their persuasiveness.

The data mining issue remains a challenge. Time after time, we have seen that optimising a strategy on historic data, modeling in a sample that is known, can subsequently turn out quite differently in the real out-of-sample world. An illustration of this is the record of stockmarket behaviour when Congress is in session or out of session. Academic studies present that 46 years of empirical data shows that over long periods of time the stockmarket performed dramatically better on days when Congress is out of session as compared to days when in session; with the split of annualised returns approximately 16.1% to 0.3%. However, a fund set up to apply this long term pattern has actually underperformed the S&P over the five years since its inception. It shows how dangerous it can be to turn long term anomalies into fund strategies.

The clearest area where behavioural economics supports some of the smart beta approaches is in the persistent underperformance of some of the very largest companies. Strategies that more evenly weight a portfolio may rightly be avoiding over-exposure to these underperforming mega-caps. Poor executive incentivisation appears to be a factor in the failure many of the largest companies to deliver shareholder value. In some cases, they may pay compensation based on beating or matching an ever narrowing group of global peers. Think, for example, of the five oil majors and incentives that mean three out of five make median performance or better and may pay bonuses based on that. These type of soft incentives are prevalent across mega-caps. For smart beta strategies and active managers, a long term bias against mega caps may be justified.

Undoubtedly there are flaws in traditional indices. And lowering portfolio management costs can assist long term returns, something that smart beta helps with. But smart beta looks mainly like a branding approach. Investors should question the labels, and recognise the emotion that may be involved in the narratives.

This article was published in CFA Institute Inside Investing on November 11 2013 and on ValueWalk.
Disclosure: The opinions expressed in this article are my own, and may not represent the views of any firm or entity with which I am affiliated. Data and content provided are for information, education, and non-commercial purposes only. The information presented in this article has been obtained from sources believed to be reliable, however, I make no representation as to their accuracy or completeness and accepts no liability for loss arising from the use of the material. Articles and information do not represent investment advice.

Photo: http://www.flickr.com/photos/guerito/. Jurgen Guerito, © Guerito 2005 source Wikimedia Commons