Non-traditional Investment: Quant versus traditional
Quantitative fund management will become more important in global small-cap and global emerging markets, writes Joseph Mariathasan, as detailed analysis of thousands of stocks for traditional management strategies would require masssive resources
The boundaries between a pure active quantitative management, smart beta approaches and traditional active fund management are blurring. By its nature, pure quantitative stock selection approaches have to rely on considerably less information than qualitative approaches. Although, even here, some quant managers would argue that unless it is insider information, there are systematic ways of incorporating most aspects of stock analysis.
David Purdy, a portfolio manager at Acadian Asset management, argues that it is even possible to systemise the information gleaned from transcripts of company meetings. Acadian already has a quantitative approach to incorporating corporate governance factors. Purdy sees the two key attributes of a quantitative management approach as, first, complete objectivity when it comes to analysis and, second, the breadth of the universe that can be analysed in a consistent manner.
Thomas Kieselstein, the CIO of Quoniam Asset Management, agrees. “The traditional active manager with 20 to 50 stocks will know each one extremely well and will probably have more information than we can hope to have. But we can analyse 5,000 stocks on a consistent basis, whereas a traditional manager using a screen to narrow down his universe will inevitably lose a lot of information.”
Another way of looking at this difference is what has been called ‘the fundamental law of active management’. This can be expressed simply in terms of the information ratio, which measures a fund manager’s ability to generate excess returns. The information ratio is merely the product of how skilled a manager is at any individual stock bet (the information coefficient) and the number of stocks that a bet is placed on.
A fundamental active stock picker such as Warren Buffett is able to have a high information ratio because he is highly skilled at selecting individual outperforming stocks and is therefore able to take concentrated bets. Quantitative approaches cannot claim to have anywhere near the skill of Buffett at selecting any specific stock. Their information coefficients are low. The only way, therefore, for quantitative approaches to have high information ratios is by taking bets on large numbers of stocks. That also has the advantage as Purdy points out, of diversifying away more of the stock-specific idiosyncratic risks for which investors are not rewarded.
On that basis, it can be easily seen why quantitative stock selection approaches are arguably more appropriate for mid and small-cap mandates and all-cap mandates where the universe of possibilities is large. “A top 100 stock would not be our playing ground but a good starting point would be a universe of at least 500 stocks,” says Kieselstein. “Our process works best if we have the full universe which in the US would be 3,000-4,000 stocks and in Europe would be over 2,000.” This would cover the full capitalisation range all the way down to small-cap. “We would typically have 15% small-cap, which would add to performance,” he says. Acadian’s funds typically have at least 250-500 stocks and can be upwards of 700 in some strategies, says Purdy.
Global small-cap and global emerging markets are two areas where quantitative managers will become increasingly important purely because of the economics of the business. Undertaking detailed fundamental analysis on thousands of small-cap stocks, or thousands of emerging market stocks spread across the globe requires huge amounts of resources if done using the traditional approach.
The ultimate expression of this would be gaining exposure to a global portfolio of emerging market smaller companies. There are probably only a handful of traditional fund managers that would even attempt to launch a global emerging market small-cap fund. But capacity is a challenge for an individual manager. With typically only between $1bn-2bn (€950m-1.9bn) seen as the maximum fund managers expect to be able to manage, the total capacity in the market place for dedicated emerging market small-cap mandates is limited, which makes it an unattractive business proposition using traditional approaches.
Accounting standards in emerging markets have generally been improving for a number of years to the extent that many successful quantitative fund managers have been able to adapt their existing models to produce emerging market strategies. In a universe of 20 or more highly disparate countries, quantitative analysis as a basis for fund management approaches allow the adoption of a consistent analysis across a whole universe.
This is particularly valuable in relation to smaller companies, where the number of firms is so large – at about 4,600, according to Purdy. Their relative illiquidity and the paucity of research on them creates difficulties for both active and passive approaches to investment. Having high numbers of stocks in a portfolio alleviates some of the illiquidity problems and has the added benefit of diversifying away more stock-specific risk. Acadian’s small-cap fund has almost 900 stocks optimised from a universe of 2,500.
Boundaries are also blurring between active quantitative management and smart beta indexation. The latter is usually based on the expression of a factor, whether value, low volatility or others, which has historically been shown to provide outperformance. “All quantitative fund managers in general are doing some sort of factor analysis because they are analysing a universe of stocks in a systematic way. Twenty years ago, they would have called it valuation ratios or styles; today we call them factors,” says Kieselstein.
Quoniam’s active quant approach based on multiple factors does, in a sense, compete with the smart beta products being launched onto the marketplace by the major index providers. “We are more flexible than using smart beta index products. Low volatility, for example, is very popular at the moment. While 10 years ago, portfolio would have been cheaper than the index, today it is more expensive and runs the risk of seeing a crash. Our approach is to combine factors and change allocations depending on the prevailing market conditions,” says Kieselstein.
Global Systematic Investors has just launched a global equity fund that combines features of both smart beta and active quant with an extremely high level of diversification: “Our target portfolio has over 1,500 stocks in it, covering global developed markets all cap excluding micro-caps. The benchmark S&P broad market index has nearly 9,000 stocks in the universe,” says Garrett Quigley, co-chief investment officer.
Its use of factor approaches to quant stock selection means the philosophy is one of starting with the complete universe of stocks, and then tilting the portfolio according to, in this case, value factors. It then applies minimum position and trade criteria to obtain the target number of stocks. As such, the information coefficient associated with any individual stock position is very low and it is only by high diversification combined with tilting that the factor bets contributing to outperformance can be isolated.
“We believe in very high levels of diversification for two reasons,” says Quigley. “First, it gets away from single stock and sector risk; secondly, rebalancing to a more diversified portfolio enhances return. We also want a small-cap bias in the portfolio because the smaller caps exhibit stronger factor effects than the mega-caps. How many stocks we have is determined by how large the actual portfolio is, since there are minimum economic transaction sizes, and so on. Clearly, investors wanting concentrated portfolios should not be looking at quant funds.”