Investors wanting to consistently exploit market inefficiencies should leave style biases behind, argues Mikhail Zverev.
6 minute read
Stock markets are inherently inefficient; their composition and the drivers of risk and return within them change over time. But while some fund managers focus on exploiting one type of inefficiency, be it growth, value or some other factor, markets are inefficient in all these areas. As such, fund managers who align themselves to one style are denying their clients a wide range of opportunities to generate strong and consistent returns over time.
Instead, our equity investment philosophy is built on a style-agnostic approach. As well as providing access to a broader opportunity set for stock selection, we believe this approach allows us to better navigate the volatility that arises as different style factors come into or fall out of favour.
In this article, we will dig deeper into the limitations of style biases and provide some insight into the benefits of style-agnostic investing and the way it shapes our approach.
Multiple market inefficiencies
Styles, which often serve as the foundation of investment philosophies, describe a specific market inefficiency that investors seek to exploit. They are in effect a short-cut, a useful taxonomy bucket, to help identify inefficiencies resulting from investors’ behaviour patterns.
The investment industry’s journey in defining investment edge and market inefficiencies around style factors goes back to the original three factors identified by investment theorists Eugene Fama and Kenneth French in the early 1990s. Those were market beta, market cap size and value.
The Fama-French model was later expanded to five factors to include profitability and return on investment, which together define what is usually referred to as quality. Cliff Asness, co-founder of the US hedge fund AQR Capital Management (and Fama and French’s PHD student), added momentum into the model. Others suggested growth and low-volatility factors.
A library of factors now exists that far exceeds those conceived in Fama and French’s original research
Through a blend of marketing and sophisticated quantitative analysis, factor investing morphed into smart beta, offering factors as investable products. A library of factors now exists that far exceeds those conceived in Fama and French’s original research.
Single-style advocates always point to empirical evidence to prove their argument. But sticking to just one style factor limits your opportunity set to a single type of inefficiency.
We believe our clients are better served through capturing a wider array of inefficiently priced stocks. Our goal is to find companies whose future fundamentals are mis-forecast by the consensus, whichever category they belong to.
Cyclically challenged: Falling in and out of style
To look at things another way, quality and growth have done well in recent years and value has done badly. There are many drivers of this cyclicality. Some relate to the macro environment; others relate to the fact business models have changed across an array of industries, meaning traditional fundamental metrics are outdated and no longer capture specific factors as they should.
The traditional book-to-price metric has lost some of its relevance
Intangible assets are a case in point. In many industries, leading companies now rely on intellectual property, employee and customer loyalty, brand or on the power of network effects to succeed. The resulting increase in intangible investment over recent years has led to accounting deficiencies that are effectively mis-specifying book values and earnings.
Further compounding things is the fact that many non-R&D intangible investments, such as in brands, IT, business processes and human resources, are included as sales, general and administrative expenses in companies’ income statements and not reflected in some traditional value metrics. This makes style definitions based on filtering fundamental metrics through a methodology unchanged over a long period of time somewhat suspect.
Figure 1 illustrates this phenomenon. As a bigger proportion of companies’ asset values is no longer captured by tangible book value, the traditional book-to-price metric used in the original Fama-French model loses some of its relevance.
Figure 1: The declining importance of tangible assets
Another problematic aspect of style-driven investing is the cyclicality and volatility of factor performance, as illustrated by Figure 2.
Figure 2: Style factor relative performance
Somewhat predictably, the bar chart showing the relative performance of different styles versus the market from year to year lacks persistency in the outperformance of any one factor. Factor investing is cyclical in the sense that it works for a while and then it doesn’t, even if this mean-reversion can sometimes take a while to show through, as has been the case with value investing recently. However, when the reversion to the mean does finally come, those on the wrong side of the trade will find it painful. In effect, by making that call you are trying to time the market, just in a factor sense. That, like any attempt to time the market, is very difficult to do.
Factor investing is cyclical in the sense that it works for a while and then it doesn’t
The correlation between the quality growth factor and interest rates is another good example. Many investors would agree the market is to an extent a discounted cash-flow (DCF) mechanism and that the discount rate is linked to an interest rate. Therefore, when interest rates are zero, or close to zero, assumed growth in year 20 becomes a meaningful contributor to your net present value. This is because growth companies have more of their value priced into the outer years; the present value and the share price largely depend on the discount rate you apply to those cash flows in the distant future. However, when interest rates normalise, suddenly year 20 is a lot less relevant and hyper-growth stocks appear to be yet another duration trade.
The point is that some factor investing can be contaminated by macro factors – a duration trade in disguise. Figure 3 shows the US 10-year Treasury yield (the risk-free rate component of the discount rate) and S&P 500 Growth index performance relative to the broader S&P 500 index over the past 15 years.
There is a clear link between the lower-for-longer interest rate regime and outperformance of growth companies. This link appears to be one of causation, not just coincidence.
Figure 3: US 10-year Treasury yield vs. S&P Growth outperformance
Some industry commentators go further in challenging the validity of style factors. In his article Alice’s Adventures in Factorland: Three Blunders That Plague Factor Investing1, Rob Arnott, founder of Scientific Beta and one of the key protagonists of factor investing, argues that many factors form the basis of investment strategies, even though the vast majority do not have scientific proof. Furthermore, he argues many have only performed well because of the weight of money chasing them; in other words, they are mere artefacts rather than real.
As he puts it, “after discovery, many investors try to exploit the anomaly so that the returns weaken and trading costs soar as the trade becomes crowded”.
Style-driven investing condemns managers to be permanently exposed to one factor’s performance volatility
This serves as a reminder that factors, and the momentum behind them, can take on a life of their own – way out of proportion to their underlying merits. Again, this validates our view that style-driven investing condemns managers (and more importantly their clients) to be permanently exposed to one factor’s performance volatility. Advocates of diversification should bear this in mind.
Connected thinking: Why style-agnostic investing improves fundamental research
You need to research a company holistically to understand it fully. Take the example of video streaming versus pay TV, a prominent change in the media industry that has been made more relevant by the COVID-induced stay-at-home trend.
To understand this fully, you need to analyse Netflix2 (which was valued at over 50x P/E multiple, growing revenues at nearly 30 per cent and just turned profitable – a classic “growth” stock); Disney or Comcast (mid-20s or mid-teens P/E multiple respectively, expected to grow revenues at mid-teens, with approximately 20 per cent operating margins – “quality” stocks); and legacy media content providers such as Discovery Inc. (c.8x P/E ratio, barely growing and viewed fundamentally challenged by this transition – a classic “value” stock).
These businesses are all inextricably linked in the supplier, customer and competitor relationship web that is central to the transition from cable to satellite to over-the-top video streaming. To gain complete insight into this this dynamic, you need to understand all of them. But to do that well you need to research a growth stock, a value stock, a quality stock, and so on. Importantly, you need to be able to do this without prejudice.
Being style agnostic gives you the scope to cover broader ground and understand the changes taking place
This is where being style agnostic brings a research advantage. It gives you the scope to cover that broader ground and understand the changes taking place, and from this identify the right non-consensus idea. Maybe you find out that Netflix is even more of a winner than people were expecting and so you buy a growth stock. Or maybe you realise Discovery has been punished too much for what its diminished, but still non-zero future, might hold and that becomes your idea. By approaching things this way, you not only understand the company itself but also its value chain, peers, customers, suppliers and so on.
The automotive sector offers another example. The industry is undergoing significant change as electric vehicles (EVs) become more affordable and consumers get more comfortable with the idea. It seems likely we are on a verge of meaningful acceleration of EV adoption.
To understand this dynamic, investors need to analyse Tesla, which leads the way on EV mass market success (valued at 70x forward P/E multiple, clearly a growth company), Texas Instruments (a semiconductor company that supplies many of the components that enable electrification, perceived as a ‘quality’ technology stock), and original automakers like Volkswagen (VW has one of the most ambitious EV programmes in terms of new model launches, R&D and capital investment, but – along with most of the auto sector – is unloved by the market, trading around 6x P/E multiple of pre-COVID earnings).
Best of the best
Again, to fully understand the change affecting these companies, and find the one where this change is most mis-priced and will lead to the most compelling upside, you need to research the whole value chain, not limit yourself to one particular sub-segment of it because of pre-existing style constraints.
Investors limiting their approach to one style factor may be less equipped
If you do this across sectors and regions and then put together a portfolio, it will not be linked to any one style. Whichever way factors move you should not be too affected by them. In contrast, an investor limiting their approach to one style factor is unlikely to have the resources to do in-depth research on companies that fall outside of their style bucket, making them less equipped to understand the implications of constantly-evolving industry dynamics.
The reality of investment management research functions, with their limited resources, is that investors analyse mostly (or only) what they can invest in. By being prepared to work across regional and sectoral silos – and even asset classes where appropriate – a research and idea generation advantage can be gleaned.
This explains our belief in the style-agnostic investment approach, where teams have sufficient analytical resources and scale and professionals connected through a common investment language and effective communication structure.
The objective for any fund manager should be to deliver repeatable and sustainable outperformance for their clients. We believe an approach that offers the flexibility to explore a broader opportunity set, avoid risks endemic in single-style investing and gain a research and informational advantage by understanding companies and their changing fundamentals across regions and sectors, gives us the best opportunity to deliver on this objective.