A Practitioner’s Guide to Style Investing (1)

Style investing is a topic of considerable depth and breadth with important implications for fund investors and for the process of efficient portfolio asset allocation. We plan to approach different aspects of style investing over a forthcoming series of blog postings. In this first post we briefly introduce the concept and then examine what are arguably the two most prevalent styles employed in the equities category: value/growth and market capitalization (there are distinct styles employed in fixed income investing as well, which we will cover in a later posting).

A “style” relates to a class of assets within a larger investment category (e.g. stocks or bonds) that demonstrates a measurable set of common attributes and performance characteristics. “All companies with a market capitalization greater than $10 billion” is a style, as would be “all stocks with a Price-to-Sales (P/S) ratio greater than 3.0x”. The style investor believes that he or she has a particular insight into characteristics and tendencies, supported by empirical evidence, which will deliver returns in excess of the broader market, either in certain market or economic cycles or over the long term.

Style investing in this sense clashes with the principles of the Efficient Markets Hypothesis (EMH), which would hold that because stock prices always reflect the full value of all materially relevant information known, there can be no sustainable advantage gained from investing in any particular style. The EMH is the stock market’s version of the fabled economist who doesn’t pick up a $20 bill lying on the ground because according to rational expectations theory that $20 could not exist…but the empirical evidence in favor of certain style approaches like the renowned “value effect” would support the idea that one can indeed from time to time stumble upon the odd bit of legal tender lying around the byways of the global financial markets. We will come back to EMH and the value effect in more detail in one of our subsequent postings.

A very useful visual aid to understanding equities style investing is the three-by-three “style matrix” popularized by investment research firms like Morningstar, Inc. and Russell Investments. A version of this style matrix appears below:

The style box runs along a horizontal axis of valuation metrics (value to growth) and a vertical axis of capitalization (large to small).

So what do the different boxes actually mean, and why does it matter? To address the first question, let’s consider how Russell Investments, which operates the various Russell stock market indexes, divides up the boxes. Russell starts with a list of 4,000 companies that make up about 99% of the total publicly traded U.S. equities market. It ranks the companies by market capitalization (shares outstanding x price per share). The top 1,000 companies by market cap become the Russell 1000 Large Cap Index. The next 2,000 companies constitute the Russell 2000 Small Cap Index. The Mid Cap Index is comprised of stocks ranking from #201 to 1,000, i.e. excluding the largest 200 companies contained in the Large Cap Index. The Russell 3000 Index consists of all the companies contained in the large, mid and small cap indexes and is commonly used as a gauge for the broader market (note: the bottom 1,000 out of the total 4,000 companies mentioned above, along with the smallest 1,000 of the Russell 2000 Index, form the Russell Microcap Index of very small companies).

That gets us the vertical segmentation. For the horizontal categories from value to growth Russell employs two metrics: the Price-to-Book (P/B) ratio and the I/B/E/S forecast long term growth mean (a composite of professional analysts’ estimated growth forecasts for individual stocks they cover). Price-to-Book is typically thought of as a value metric: stocks that have relatively low P/B ratios are considered to be potentially worth more than their current price would indicate. I/B/E/S is a growth measure as it identifies stocks with the potential for faster-than-average earnings growth and thus to possibly increase in share price as well. Using a proprietary algorithm Russell ranks 70% of all companies as either pure value or pure growth, with the remainder allocated to the middle “blend” category.

So that’s one way to take several thousand companies and carve them up into style boxes. Why does this matter? To address this question let us consider some actual returns data as indicated below:

Now, remember that what we are looking at here are simply the same 3,000 companies sliced and diced in different ways according to style parameters. Notice how for any given time period from one year to 16 years (average annual) the returns are strikingly different for different value-growth and capitalization styles. Therein lies the importance. Different styles perform in different ways both over shorter term market cycles and over longer macro periods (consider how many dramatic market cycles we have been through since 1994, the start date for the 16 year returns data). For asset allocation purposes those differences are crucial: the cardinal rule of prudent portfolio allocation is diversification, and diversification starts with the ability to identify distinct asset classes that do different things at different times in different orders of magnitude.

In our next post we are going to examine some of this data in further detail, and in particular focus on a very prominent aspect of style investing called the “value effect”. If you look closely at the returns data shown here for value indexes versus growth or blend results you probably will have a pretty good idea where that discussion is headed. Stay tuned!