Style Investing

Melissa Hill

Melissa Hill, Sabre Fund Management Ltd
Originally published in the March 2005 issue

"Oh, never mind the fashion. When one has a style of one's own, it is always 20 times better." – Margaret Oliphant, British historian.

Style investing is not a new concept and has been employed in the long only asset management world since Graham and Dodd introduced the concept of value investing and T. Rowe Price disputed its merit and advocated Growth. The market has since segmented into Value and Growth management styles with many asset managers subscribing to the view that these investment philosophies are mutually exclusive. However, there are stocks that have both Value and Growth properties and there are some that have neither. Contrary to popular opinion, we have found that both styles have positive alpha in the long run – in other words, style investing is not a zero sum game.

Style Rotation

There is a fundamental link between the economic cycle and style cycles. Value stocks tend to do well when the economy is strong or during cyclical recovery. Growth stocks tend to do well when growth is scarce – i.e. when the economy is growing more slowly than usual. So, economic and style variables are intrinsically correlated. These cycles are further intensified by behavioural trends – the phenomenon of active managers "herding" due to fear and greed emotions. The combination of long term economic cycling effects and short term behavioural trends can be captured by style rotation strategies.

Style Arbitrage

The existence of style cycles and trends implies that style returns are reasonably predictable. Style Arbitrage extends the methodology of style investing and style rotation to a long/short market neutral framework. Stocks are ranked according to their style characteristics and spread portfolios constructed. These portfolios are akin to holding a series of uncorrelated synthetic financial instruments. A Value portfolio is long cheap (value) stocks and short expensive (non value) ones. Value portfolios have positive near term cashflows and are thus similar to holding a short duration bond. A Growth portfolio would be long high growth stocks and short low growth stocks. Growth portfolios have negative near term cashflows and are ultra long duration. By focussing on capturing cycles and trends on both the long and the short side, predictability of style returns is enhanced which, in turn, leads to more robust forecasting.

In addition to Value and Growth, two more investment styles have demonstrated a strong causal and empirical relationship with stock returns – namely, Momentum and Quality. Together, these four key styles explain more than 80% of stock movements.

Forecasting Style Regimes

In a static world we would like to be long of quality and good valuation companies with solid growth prospects and momentum and short of poor quality, poor growth, poor valuation and momentum companies.

However, the market place is largely sentiment driven and styles come in and out of fashion. This implies that there is a tendency to focus more on Value at times when the market sentiment favours cheap stocks and that we tend to focus on Growth when the market is rewarding stocks with good growth prospects.

The Style Arbitrage philosophy is to develop models that suit most market conditions and respond promptly to changing market sentiment without being too sensitive in choppy markets. Key to achieving this is the ability to accurately forecast the direction of style returns so that under performing styles can be 'downgraded' or 'switched off'. In this way, the Style Arbitrage investment process differs from static risk model based style strategies. The latter can perform strongly when the market is favouring their particular style but can lose out when the market reverses out of that style. For this reason, if one favours market neutral style investment, it would be prudent to allocate to both dynamic (high turnover) and static (low turnover) style managers.

Persistence and Opportunity (strong v. weak regimes)

As style portfolios are constructed using constant rules, their history can be backtested over many years. Going back to the early 80's, incorporating many examples of changing regimes, style portfolios display positive bias – i.e. Value, Growth, Quality and Momentum all make money. We call this Style Opportunity. The portfolios also display positive auto-correlation – i.e. style factors go up or down consistently for a period. We call this Style Persistence. Style returns can be exceptionally strong in defined regimes and particularly in falling markets, as persistence is stronger on the downside (the fear sentiment is more quickly assimilated by market participants than greed). For this reason, Style Arbitrage represents a good 'insurance-type' strategy to hold in a portfolio of hedge funds and has considerably less downside risk than say, a CTA strategy. In fact, Style Arbitrage is also an excellent alternative to short biased managers and sometimes unlike the latter, is able to perform well in rising markets.

Style returns are weaker when there is no common consensus in the market place (such as, at times of uncertainty as to the future direction of interest rates). It is very unusual for persistence and opportunity to be negative at the same time (only one previous period in 20 years' history) but this is what occurred during the first half of 2004. It is important to be diversified by market to counteract weak sentiment in a particular region. Sabre's Style Arbitrage team has hitherto traded UK and European portfolios but will be commencing US trading in April with a plan to incorporate Japanese stocks by the third quarter of 2005. Interestingly, US style portfolios have surprisingly little correlation with UK/European portfolios and should prove to be a major enhancement.

Risk Management

Style Arbitrage is a true market neutral methodology with no beta, market, or sector exposure. In line with most quantitative strategies, risk limits and targets are model calibrated and there is no discretionary input. Diversification by holding hundreds of names in small size mitigates stock specific risk. This is key to success as, in recent times, markets have experienced some significant individual stock moves on a day.

In fact, the multiple non-correlated style portfolios could be compared with the portfolio allocation of a fund of funds manager. Instead ofmanagers, substitute style portfolios (each portfolio containing in excess of 500 names) – some with Growth bias, some Momentum, some Value and some Quality.

Myths and Reality

One of the attractions of the Style Arbitrage methodology is that it is a fairly niche approach. It's neither statistical arbitrage nor fundamental quantitative equity market neutral – rather something between the two. As such it has little correlation to other market neutral strategies. Because not too many people are paddling in the Style Arbitrage pond, the strategy can be researched, implemented and enhanced by a small focussed team supported by a robust trading platform originally developed for high volume transactions. It has become fashionable to believe that quant cannot be 'done' without an army of PhD's but, in true hedge fund entrepreneurial spirit, it is possible to preserve the integrity of proprietary models whilst taking advantage of excellent 'outsourced' research to rival that produced from some of the largest R&D spends.

The other myth is that volatility equals lack of opportunity. It is true that it is difficult to generate super style returns in periods of low volatility but, with adjusted risk targets it is perfectly possible to achieve a double digit return.

One final thought

In the same way as the insurance industry is concerned with the distribution of losses (the frequency of losses multiplied by the size of losses), the hedge fund industry should be concerned with the distribution of gains (frequency of gains multiplied by the size of the gains). In other words, looking at the frequency or the average gains/size in isolation is a flawed approach. Frequency and size of gains should be considered together. Unsurprisingly, the Style Arbitrage methodology demonstrates positive skewness! Monthly gains have by far outstripped monthly losses providing an attractive overall distribution with little downside risk and good upside potential.

Melissa Hill is Chief Operating Officer at Sabre Fund Management Limited. Sabre is an independent boutique hedge fund business specialising in trading and developing innovative quantitative strategies. Sabre was founded in 1982and, as such, is one of the early pioneers of the European alternative investment business. The Style Arbitrage Fund launched in August of 2002and was followed by the Long/Short Style Fund in February of 2004. The Style strategies have been designed and developed by Dan Jelicic, CIO. Dan is supported by a team of portfolio managers.