Pioneers: Better be smart
Lynn Strongin Dodds finds that as the strategy becomes more popular, pioneers in the alternative-indexation field are warning investors to avoid being just performance chasers
At a glance
• Factor investing has its origins in the Capital Asset Pricing Model (CAPM) delivered over 50 years ago by William Sharpe.
• The idea of alternative indexing had started to germinate by the 1980s.
• Enhanced strategies typically aim to outperform their underlying market by 1-2 percentage points a year before fees.
• Some investment groups pursue multi-factor approaches.
Although smart beta has become the catch phrase for the alternative indexation trend, many of the pioneers in the investment style baulk at the name. Some prefer the terms ‘factor-based’ or ‘style investing’, while others favour ‘character biased’, ‘strategic beta’ or ‘genius bias’. Whatever the label, the common thread has been a move away from the traditional market-cap benchmarks to produce better risk-adjusted returns and greater downside protection.
Despite the recent spate of products, these ground breakers credit economist William Sharpe with creating the original framework in the shape of the Capital Asset Pricing Model (CAPM) more than 50 years ago. By streamlining the tenets of Harry Markowitz’s earlier Modern Portfolio Theory into a compact framework, the CAPM introduced ‘the market’ as a factor. It divided a portfolio’s returns into alpha and beta where, over time, the former became synonymous with the capitalisation-weighted return of the market, while the latter was used as a measure of a manager’s skill in generating performance.
Fast forward to the 1980s and the idea of alternative indexation began to germinate with Sharpe as well as many others realising that a sizeable portion of the ‘alpha’ attributed to the skill of active managers by the CAPM could be reproduced by using simple rules-based approaches. The model was further extended by using Arbitrage Pricing Theory (APT), which expands beta from a single market measure to include any number of factors such as style (growth and value), capitalisation (large, mid, small), and momentum (persistence among ‘winners’).
These categories remain the building blocks of today’s crop of smart beta products, although others have been added to the mix. Some of the most fashionable include low volatility, whereby companies with stable share prices tend to do well, and quality, which has different definitions but typically means a corporate with a strong balance sheet and management team.
The term smart beta was coined by Towers Watson in the beginning of the century and it has proven extremely popular over the past five years as interest rates have continued to languish. Figures from Morningstar show that assets under management have soared from $103bn (€94bn) in 2008 to $616bn in 2015. However, as with many popular investment fads, the field is becoming overcrowded and returns are in danger of being squeezed. In addition, some strategies are becoming overcomplicated and lines are being blurred with traditional active management.
Many of the forerunners are already sounding warning bells. Rob Arnott, chairman and chief executive of Research Affiliates, the US company that launched some of the world’s first smart beta indices in 2005, says that, in some ways, the popularity could be its downfall.
“One of the biggest problems with today’s products is that many investors are performance chasers and make the mistake of forecasting the future by extrapolating the past,” he says. “If investors buy a stock that has already gone up just because it is rising, then they are buying for reasons that have nothing to do with valuations. Investors need to be careful, because past excess returns may not be sustainable in the future. Indeed, our evidence suggests that mean reversion could wreak havoc in the world of smart beta.”
To be meaningful, ‘smart beta’ strategies should studiously sever the link between the price or the market cap and a security’s weight in the portfolio, so that a company will not rise simply because the price jumped, Arnott says. This characteristic is at the core of the RAFI indices which weight companies based on their economic footprints of revenues, profits and dividends, rather than aggregate market values.
Arnott is also careful to point out that returns from the fundamental strategies have not always been in a straight upward line but they have turned in a cumulative annualised performance from 31 December 2005 to 31 December 2014 of 9.4% against 7.1% for the actively managed median Lipper Large-Cap Core mutual fund and 7.9% for the S&P 500. In addition, in the 10 years ended 31 December 2015, the FTSE RAFI US 1000 index delivered 8.22% a year, against 7.31% for the S&P 500 and 7.40% for the Russell 1000.
INTECH, a subsidiary of Janus Capital and also one of the factor-investing forerunners, says the pitfalls can be avoided by diversifying the holdings and rebalancing.
The idea was first espoused in a seminal paper in 1982 published by Dr E Robert Fernholz, the founder of the firm and a creator of the enhanced equity portfolio construction method. “In the beginning, these findings were very controversial, but we are able to prove mathematically that the factors themselves are not generating returns – rather, the regular rebalancing of the portfolio does,” says David Schofield, president of the firm’s London-based international division.
He explains that the goal is to isolate the alpha or excess returns of these strategies and target this alpha source while trying to minimise exposure to the various extraneous risk factors that may or may not be in favour at any moment in time. This means that instead of, for example, creating funds of just small-cap or value equities, it selects stocks based on their volatility and low correlation with each other. Those with higher volatility are preferred because they offer better opportunity for rebalancing – buying low and selling high.
Although performance varies depending on the strategy, its relative risk or ‘enhanced’ strategies, which typically aim to outperform their underlying market by 1-2 percentage points a year before fees (which are typically 20 to 40 basis points a year for institutional investors), have outperformed by up to 4.89 percentage points a year since inception.
“One of the biggest problems with today’s products is that many investors are performance chasers and make the mistake of forecasting the future by extrapolating the past. If investors buy a stock that has already gone up just because it is rising, then they are buying for reasons that have nothing to do with valuations. Investors need to be careful, because past excess returns may not be sustainable in the future”
Last year, its global low-volatility strategies outperformed the MSCI World index by 4.4 percentage points. The US strategy outstripped the Russell 1000 by 3.6 percentage points and emerging markets were down by 6.5 compared with a fall of 14.6% in the corresponding MSCI index, according to Schofield.
Diversification is also a key ingredient for success, particularly for Yves Choueifaty, president and founder of TOBAM which he launched in 2005, “Everyone talks about diversification but it is usually not a well-defined concept. Asset managers know how to measure volatility, tracking error, alpha, beta, gamma, delta.”
Choueifaty rectified this by developing so-called anti-benchmark strategies using a systematic quantitative methodology that avoids the concentration risk associated with market cap-weighted indices. The aim is to create the most neutral portfolio possible, taking no bets on style or valuation in the hope of delivering stable performance across market cycles while capturing the full risk premium available for a given asset class.
Unlike many of its rivals, TOBAM, which has $9bn (€8bn) in assets, does not have a salesforce but, instead, places a high priority on research, which occupies about half of its 36-strong team. The main focus areas include elaborating the mathematical properties of the diversification measurement, determining how best to implement it and finally, exploring how well it will work in other markets and asset classes.
This approach probably explains why Choueifaty eschews the term smart beta. “I believe there is no smart beta, but there is one dumb beta – the cap-weighted benchmark. What we are trying to access is the true beta [the source of the equity-risk premium]. I am the only portfolio manager who will promise to deliver zero alpha.”
AQR Capital Management, which opened its doors in 1998, is also an advocate of diversification but places greater emphasis on the integration of what it calls different styles. These include the classic value, whereby relatively cheap assets tend to outperform relatively expensive ones, and momentum, where recent relative performance is expected to continue in the near future. Also in the group are carry – higher-yielding assets that provide higher returns than lower-yielding assets; and defensive – lower risk and higher-quality assets that typically generate higher risk-adjusted returns.
Founders Clifford Asness and John Liew provided credence for their views in a paper entitled Smart Beta: Not New, Not Beta, Still Awesome in 2014. They noted that “the excess returns of value, momentum, and profitability factors tend to be lowly correlated with one another, with performance often coming at different times. “As a result, diversifying across various smart betas can and has provided a more consistent way of beating a traditional benchmark.”
The firm has adopted a multi-strategy approach which combines the styles together in one portfolio instead of blending four different portfolios together, according to Ronen Israel, a principal with AQR. “We are not trend followers but believe by looking at the relative merits of the styles we can leverage the benefits and synergies and create the best combination net of fees and one that reduces transaction costs.”
According to its own analysis of market data from January 1990 to June 2013, diversified style premia theoretical portfolios delivered both positive risk-adjusted returns – Sharpe ratios ranging from 0.9 to 1.3 – as well as diversification from equity-directional risk – correlations to global equities ranged from approximately -0.1 to +0.2. Its own set of results reveals that the large-cap multi-strategy style generated 17.3% a year, on average, since inception in April 2013, against 11.3% for the Russell 1000 US large cap index.
“When it comes to multi-factor investing, you really have to know how to blend the portfolios and which sectors, factors and countries to use”
The approach does not only apply to long-only equity portfolios but also across the asset classes. Two years ago, the firm launched its AQR Style Premia Alternative fund which employs a long/short strategy in an attempt to isolate specific factors in stocks, bonds, commodities and currencies. “We take the other side of the bet and hold unwanted stocks. There is some unwanted risk but you also get a premium and natural balances,” adds Israel.
Francois Millet, head of ETF and index product development at the French asset manager Lyxor Asset Management, is also a proponent of the multi-factor approach, although he sees merits in combining single indices together. He says that while single-factor exposures have been shown to provide significant long-term outperformance, they can also go through periods of short and medium-term underperformance. As the correlations between each factor are often quite weak, combining them should provide diversification benefits which can smooth performance.
Last year, Lyxor, which has been in this area since 2008, joined forces with JP Morgan to launch a new range of risk-factor ETFs based on five factors that Lyxor research shows has a solid theoretical grounding. They include value, which it defines as a composite of book-to-price, earnings yield and free cash flow yield, low size and momentum, which is total return (with dividends reinvested) over the last 12 months. It also covers low beta or the beta of the stock relative to its local market, as well as quality – an aggregate of total equity/net debt, return on equity and net income/sales.
The indices are not global but regional and local because, Millet says, this is the best way to capture outperformance. “If we implement the factor-scoring process on a global basis we tend to arrive at portfolios that are heavily concentrated in particular geographical markets,” he notes. “Also, the performance of individual risk-factor strategies has diverged across different markets.”
For example, the size factor has done well in the US, Europe and Japan since 2000, but has underperformed in the Asia Pacific region. These differences also often reflect differences in market structure.
Finding the right combinations is also not always easy. “When it comes to multi-factor investing, you really have to know how to blend the portfolios and which sectors, factors and countries to use,” says Ryan Taliaferro, senior vice-president and portfolio manager at Acadian Asset Management which has offered active strategies based on low-volatility investing since the early 2000s. “You also need to have strong risk modelling and attention to transaction costs and price impact because you don’t want to bleed away the returns.”