Styles debate is now over
Even as the academic debate mumbles forward and current research diligently examines how styles might best be theoretically defined and where, and how, they might be relevant, recent equity returns and vivid investor experiences have already spoken very clearly. From 1997 to 2000, value underperformed and growth companies soared. Also, from the beginning of 2000 until almost the end of 2001, value rebounded dramatically.
Among practitioners, questioning the relevance of styles is now viewed as idle, meaningless sophistication. In the light of the overwhelming evidence of painful (or joyful) experience, how could anyone doubt the importance of styles?
Over the past four years, style rewards have been particularly dramatic across the major developed markets. Value fell precipitously from early 1998 to early 2000, only to rebound spectacularly over the following 18 months. While this volatility is striking in the light of the past 10 years of data, it is not at all unprecedented over the longer term.
These dramatic patterns of rewards have secured style a prominent position in market psychology and investor awareness. At the same time, they have also served to highlight the profound risks inherent in style management. If you get the timing wrong and are tilted the wrong way, even for only a few months, it can cost dearly.
There is, it seems only one sensible way forwards for risk-averse managers: offer both value- and growth- oriented investment products. In that way, there will always be something to offer in each important category and, if you can distance yourself from the value versus growth timing decision, you are unlikely to have serious business risk problems. If you can avoid making the big and risky decisions, you are not going to be implicated if your favoured style suddenly underperforms, but you can still benefit from always having something in fashion to market.
So, it is really not at all surprising that in the US the mutual funds universe, as well as individual segregated investment product offerings are divided according to clearly defined style orientations. And it is very likely that, as in the US market, European funds and fund management will soon go the same way.
While this may benefit consumers of investment products it raises some serious issues among the providers of these investment services and for the consultants attempting to adjudicate between them. Just to start: How can a business manage both value and growth investments? Can the same process be used to manage both value and growth investment products? Can the same team manage both value and growth investment products? How do you assess skill, and what do you look for among value managers/among growth managers?
You need a structured methodology to look into these issues. And you have to be particularly careful how you put it together. For example, if you simply reveal that different investment criteria are at work within the value (simply high book to price stocks) category from those that are at work within the growth (low book to price stocks) category you may conclude that different selection criteria need to be applied within the investment process. But you still have no evidence that suggests that the investment process itself within the value universe ought to be different from that within growth. That is to say, you will have no insights about the qualities of the individual managers that destine them to be successful value managers or successful growth managers, if indeed these qualities might be different. And these are the important issues.
Our exploration into these questions focuses on the performance characteristics of markets themselves.
Within the UK equity market, we track the performance of the value universe of stocks and the growth universe of stocks. Since this is research and not simply index creation, we look at a rarefied sample of each style, the top 25% capitalisation (sorted by book to price) being value, and the bottom 25% (by book to price) being growth. Since we want to be sure that the starting point of the analysis or the frequency of the rebalancings does not affect the results, we construct the value and growth sample portfolios using one-month, six-month and 12- month re-balancing frequencies. To ensure that recent sector volatility did not contort the results, all analysis was conducted in sector-neutral terms, assessing all security factor scores against same sector stocks and conducting the sorts and selections sector by sector. Then we examine the relative returns of sub-portfolios (selected from within each sample, and measured relative to the performance of the sample), selected on the basis of a handful of value factors and growth factors.
The selection factors explored are: earnings yield, earnings growth, forecast earnings growth, forecast earnings revisions, return on equity. Some of these might popularly be considered as value measures and others as growth measures. But to broaden the scope of the enquiry, we looked at each factor applied within both style universes. That is, we looked at value factors within the value universe as well as within the growth universe; and the same for the growth characteristics.
To ensure that the results are not period/frequency specific the analysis was done using one-month, six- months and 12-month rebalancings. The results are displayed in figure 2.
A couple of observations are immediate. The return patterns appear not to depend systematically on the rebalancing frequency. However, as one would expect, there are performance differences among the portfolios selected according to earnings forecast revisions, where early responses to brokers’ forecast revisions seem to be better rewarded than delayed responses.
Most significantly, there are three strategies that appear to work systematically throughout the period within the value universe. By contrast, nothing offered any systematic pattern of rewards within the growth universe.
A number of conclusions and caveats also follow quickly: systematic investment in accordance with established value and growth selection criteria appears to offer the potential to outperform within a value universe. There is, however, no evidence that investors benefit from such systematic practices within the universe of growth stocks.
Consequently, value managers can succeed by being methodical, systematic, theme-based investors while growth managers must succeed through stock selection.
Senior investment professionals should assemble value and growth investment teams according to the methodological disciplines and stock selection skills of the individual portfolio managers in their department. On the basis of this analysis there may be wide differences in approach between good value managers and good growth managers and one should not substitute for the other.
Consultants should recognise that the characteristics of successful value managers can differ from those of successful growth managers and appraise each accordingly. Value managers should be able to articulate a consistent investment philosophy and demonstrate disciplined buy and sell practices. Growth managers ought to be able to assemble, collate and analyse specialist knowledge of individual securities and provide evidence of an underlying supporting process.
But, we must be careful not to over-interpret these initial results. While we have defined value and growth according to the ‘standard’ book to price measure, other factors or factor combinations are becoming popular redefinitions of value and growth.
All in all, the conclusions seem sensible against our intuitive understanding of what drives investment thinking among value and growth investors. Value managers generally hold themselves out to be systematic in their approach, while growth managers usually emphasise their ability to identify future winners on the basis of specialised company research.
In fact, the whole culture of being a value investor is based on a broad theoretical foundation, which promises value investors positive market relative returns as systematic compensation for accepting the higher risk and lower popularity of under priced securities. What is so reassuring about the research offered here, is that for value investors such systematic strategies actually work.
Growth managers, according to the theoreticians, need to be opportunistic and good stock selectors, since the cards are already stacked against them as they buy popular securities with perceived lower risk. Our results show that, at least using the criteria we have selected, there is unlikely to be a systematic way to be a good growth investor. So their stock selection skills must be good.
Robert Schwob is director of Style Research in London