Andrew Clare, Stephen Thomas and Nick Motson trace the roots of smart beta that began as a test of the Efficient Market Hypothesis in universities in the 1970s 

At a glance 

• The ideas behind smart beta emerged in the finance departments of universities in the 1970s.
• The previous assumption that returns were related to volatility was called into question.
• The discovery of such anomalies helped provide investors with new opportunities.
• It is possible to  use such anomalies to create indices for investors to follow.

When it comes to constructing an equity market index, the most widely accepted approach is to weight each constituent according to its market capitalisation. But there is an infinite number of ways in which one could specify the constituent weights of any equity index. The financial industry has given alternative-indexing approaches the moniker of smart beta, although others refer to the notion of alternative beta. So where did smart beta come from or, in other words, how did the industry choose alternatives to the market-cap approach from an infinite set of alternatives?

Much of what passes today as smart beta investing began life as a test of the Efficient Market Hypothesis (EMH) in the finance departments of universities in the 1970s, 1980s and 1990s. These experiments seemed to indicate the existence of sources of systematic risk to which investors could get exposure to earn attractive risk-adjusted returns. 

One of the main tenets of modern portfolio theory is that as long as an investor holds a well-diversified portfolio of risky securities then, over time, the higher the inherent expected risk in that portfolio, the higher should be the expected return. Mean-variance analysis, as the name suggests, characterises risk as volatility. This is the accepted practice and few question this idea today, although it was quite revolutionary back in the 1950s when Harry Markowitz first proposed it. If high risk should lead over time to higher return, then one could expect that stocks that produce returns with low volatility should generate lower returns over time than stocks that generate a higher return volatility. 

In 1972, two academics, Robert Haugen and James Heins, tested this hypothesis. Remarkably, they found that there was a strong negative relationship between return and volatility in both the stock and bond market. That is, low volatility stocks tended to outperform high-volatility stocks over long periods of time. This was, arguably, one of the first pieces of evidence that modern portfolio theory was missing something – something that investors might be able to benefit from.

Annualised returns by decile, from investing based on smart beta factors

In 1981, Rolf Banz published a paper looking at the relationship between stock size and performance. Not only did Banz find that small-cap stocks outperformed large-cap stocks, he found that they did so even though, on average, they embodied lower risk than large-cap stocks. Banz had uncovered what is now called the ‘size effect’. Using a similar methodology, in 1983 Sanjoy Basu published a paper that demonstrated the ‘P/E effect’ – that is, where stocks with low price to earnings (P/E) ratios tended to outperform those with high P/E ratios. Again this outperformance seemed to come with lower risk.

Urged on by these findings, in 1985 Barr Rosenberg, Kenneth Reid and Ronald Lanstein published a paper documenting a ‘book-to-market value’ effect, where high book-to-market value stocks tend to outperform those with a low book-to-market value. In the same year, Donald Keim identified the dividend-yield effect. He showed that investing in high-dividend-yielding stocks produced higher returns over time than an equivalent strategy focusing on investment in stocks with low dividend yields. And, once again, he found that the higher performing, high-dividend stocks produced this performance with lower measured risk.

In 1993, two researchers, Narasimhan Jegadeesh and Sheridan Titman published a paper that investigated the phenomenon of momentum investing. They found that by buying stocks that had performed well in the past and selling stocks that had performed poorly in the past, significant positive returns, with lower risk could be earned by investors. 

Indeed, by the early 1990s there appeared to be a whole range of phenomena that offered attractive investment opportunities to investors. High, risk-adjusted returns could be generated by simple rules-based investing in, for example: low-volatility stocks; small-cap stocks; stocks with low P/Es; stocks with high book-to-market values; stocks with high dividend yields; and stocks with high price momentum. These results were all at odds with the EMH. For example, if it were possible to earn high, risk-adjusted returns simply from investing in stocks with high dividend yields, why didn’t rational investors realise this and buy high-dividend-yielding stocks, thereby increasing their price and reducing their return advantage?

“In 1972 Robert Haugen and James Heins found that there was a strong negative relationship between return and volatility in both the stock and bond market. That is, low-volatility stocks tended to outperform high-volatility stocks over long periods of time. This was, arguably, one of the first pieces of evidence that modern portfolio theory was missing something –something that investors might be able to benefit from”

Each of these anomalies can be accessed using a set of rules very similar to those that are required to create a market capitalisation-weighted index, or portfolio. Indeed, the methods used by the academics to unveil these risk factors are very similar to a set of index rules. As an example, consider how one could gain exposure to the dividend yield factor:

• At the end of a quarter, consider all the stocks in the London Stock Exchange;

• Identify the 10% of stocks with the highest dividend yield;

• Invest in these stocks on either an equally-weighted or a market cap-weighted basis;

• Hold this portfolio for the following quarter;

• At the end of the quarter repeat the process, by once again identifying the 10% of stocks with the highest dividends and investing in these stocks on either an equally-weighted or a market cap-weighted basis;

•  And then simply repeat this process.

By following this rule, an index provider could create a ‘High Yield Equity index’. By following the rule, an investor could access the attractive returns available from high-dividend-yield investing. To see just how much such a strategy could yield, in the table we have presented the returns that might have been generated by investing in deciles of stocks according to each of the ‘smart beta’ criteria mentioned above. 

The values presented in the table represent the annualised return on decile portfolios formed on the basis of each one of the smart beta filters. For example, in row one we present the results of using a volatility rule. Over the period from 1963 to 2014, constantly investing in the decile of stocks with the lowest past volatility would have produced a return of 10.7% year, while investing in the 10% of stocks with the highest volatility would have produced an average annual return of 4.6%. The other annualised return figures in the row represent the returns that would have been achieved by investing in the intermediate, volatility deciles. 

In every case, investing according to the rules produces high returns relative to doing the opposite of what the rule says. Perhaps the most impressive performance is produced by the application of the momentum rule. Investing in the 10% of stocks with the highest price momentum produces an annualised return since 1927 of 19.9%, compared with a return of 4.0% that could have been achieved by investing in the 10% of stocks with the lowest price momentum. 

Investors are being bombarded with many new smart beta offerings today. So which one should they choose?

Of course, the first thing to note is that past performance is no guarantee of future performance. But if an investor is willing to abandon the market-cap approach to investing and to take the smart beta plunge they would be well advised to investigate the origins of the smart beta investment approach first. Was the approach dreamt up in the marketing department of an asset manager, or did it come to light many years ago as the result of academic research into the validity of the EMH? I know which one I would be more comfortable with.

This article is based upon a Cass Business School paper, that can be found at:

Andrew Clare is professor of asset management at Cass Business School in London. Stephen Thomas is professor of financial markets and Nick Motson is senior lecturer in finance at the same institution