In Focus: Style Investing

As you may remember, when we rebalanced our recommended portfolios at the end of last year, we used a fairly broad definition of an asset class. Specifically, because the benefit from diversification comes from risk reduction, we required that the "asset classes" we used could have no more than a .60 correlation of returns with each other. That definition eliminated from use a number of groupings of stocks and bonds that other commentators call "asset classes." Examples of these include small cap stocks or large cap growth stocks, and short-term bonds. In our view, all of these represent various "tilts" that one can make in order to enhance the risk/return trade-off within an asset class. At the time of our rebalancing, we promised that we would be taking a closer look at these "tilts" to see which, if any of them, made sense. We’re beginning that series of explorations this month, with a look at style investing. Next month we’ll look at sector investing. In July we’ll look at country investing, in August we’ll look at investing in different bond maturities, and in September we’ll look at momentum investing. It promises to be an interesting journey.

Our starting point is style investing, which is probably the best known and popular of the various tilts that investors employ in their search for better returns and/or lower risk. Depending on who you talk to, the term "style investing" can mean investing in groups of companies with similar market capitalizations (that is, large, mid, and small cap stocks), investing in groups of companies with similar book to market rations (that is, value and growth stocks), or a combination of both approaches (for example, small cap value stocks, or large cap growth stocks). Regardless of the approach taken, the fundamental question that must be asked remains the same: can you improve on the risk/return trade-off for the asset class as a whole by making a style tilt in your portfolio?

Let’s start with market cap tilts. We’ll use the S&P 500 as our large cap index, the S&P 400 for our mid cap index, and the Russell 2000 for our small cap index. This allows the longest possible time series of returns data, dating back to the start of the S&P 400 in February, 1981. Between then and the end of 2000, the average annual return on the large cap index was 17.30%, with a standard deviation of annual returns of 17.38%. In other words, by investing in a large cap index you received .995% of return for every 1.00% of risk (as measured by standard deviation) you took on. During the same period the average annual return on the mid cap index was 19.28%, with a standard deviation of 19.52%, or .988% of return per unit of risk. Finally, between February, 1981 and December, 2000, the small cap index delivered average annual returns of 14.09%, with a standard deviation of 21.59%, or only .653% of return per unit of risk. By way of comparison, during this same period the market as a whole, as represented by the Russell 3000 index, had an average annual return of 16.69%, with a standard deviation of 17.48%, or .955% of return per unit of risk.

In order to compare apples to apples, we will ask the same question in all of our analyses: by making a tilt, could I have earned higher returns than the market as a whole while taking on the same amount of risk? In this case, the answer is yes. A mix of 80% large caps and 20% mid caps would have had about the same standard deviation, but would have delivered average annual returns of 17.69%, or .988% of return per unit of risk.

Could I have done better if I had used a tilt based on growth versus value instead of one based on market cap? Let’s see. As proxies for the growth and value tilts, we’ll use the Russell 3000 growth and value indexes. The stocks in the Russell 3000 universe are placed into these categories based on two factors: their market/book ratio, and their earnings growth rate. Between February, 1981 and December, 2000, the Russell 3000 growth index had average annual returns of 16.13%, with a standard deviation of 20.19%, or .799% of return per unit of risk. During the same period, the Russell 3000 value index had average annual returns of 17.09%, with a standard deviation of 16.36%, or 1.045% of return per unit of risk. In this case, the best approach would have been to invest 100% of your portfolio in the value index.

Finally, we need to look at using a combination of market cap and growth versus value tilts. Because of the short time that growth and value indexes have been available for the S&P 400, we will use five indexes in this analysis: large cap growth and value (in this case, the S&P/BARRA 500 Growth and Value), mid cap (the S&P 400), and small cap growth and value (in this case, the Russell 2000 growth and value). During the February, 1981 to December, 2000 period, large cap growth had average annual returns of 17.09%, with a standard deviation of 19.21%, or .890% of return per unit of risk. Large cap value had average annual returns of 17.32% with a standard deviation of only 16.82%, or 1.03% of return per unit of risk. During this same period, small cap growth had average annual returns of 11.99% with a standard deviation of 25.97%, or only .462% of return per unit of risk. By comparison, small cap value delivered average annual returns of 16.20% with a standard deviation of 18.17%, or .892% of return per unit of risk.

In this case, a portfolio weighted 54% in large cap value and 46% in mid caps would have delivered average annual returns of 18.22% between February, 1981 and December, 2000, with a standard deviation of 17.48%. That’s 1.042% of return per unit of risk, versus .955% for the market as a whole, as proxied by the Russell 3000 index. Let’s look at this another way: over 20 years, on a $1,000 investment, the difference between earning an average annual return of 18.22% and earning 16.69% compounds to $1,355.

So, what can we conclude from this analysis? At first blush, it would seem that if history is an accurate guide to the future (and, to be sure, sometimes it isn’t), a mix of large cap value and mid cap stocks offers a superior risk/return trade-off than simply investing in the market index as a whole. What are the chances that this will be as true in the future as it was in the past? The first response to this question is you can never know for certain. However, what you can do is try to identify the underlying causes of the superior returns you have identified. If these seem likely to persist, then it is more likely that the future will indeed resemble, if not exactly mimic, the past.

So, what then might be driving the superior returns we have identified? Let’s look at mid cap first. One theory is that mid cap typically outperforms small cap because in effect, the former represents the cream of the crop of the latter. In other words, the best small cap companies – those that are growing quickly with a profitable business model that isn’t easy for others to copy – make it into the mid cap index. When this happens, the small cap managers have to sell them. Given their track records, new entrants to the mid-cap universe have a relatively high probability of continuing their success. Inevitably, some of them will later be added to the large cap indexes, which triggers another wave of buying by large cap managers, and a last surge in their stock price as they depart the mid-cap universe. In other words, mid-cap managers may enjoy a structural advantage that is denied to large and small cap managers.

But what about the difference in value returns versus growth returns? What might be going on there? Our view is that the underlying cause of the value index’s strong relative performance are two fundamental flaws in the design of the indexes themselves. The first flaw is mechanical. Consider what happened at the end of last year, when the indexes were rebalanced. Following sharp falls in their stock prices, companies like American Power Conversion, JDS Uniphase, Best Buy and Qwest were moved from the growth to the value universe. At the same time, companies like Bank of New York, Kimberly-Clark, Kroger, and Hershey Foods were moved from the value to the growth category. If, as we do, you believe that at the margin, financial markets behave irrationally (due to imperfect communication of information and flaws in human reasoning), then it follows that rebalancing is likely to put into the value universe stocks whose price has overreacted on the downside, while the growth universe receives stocks whose price has overreacted on the upside. As these misvaluations are recognized by investors, the value index should logically outperform the growth index, which is exactly what we see in the data.

This points to the second, and more important, fundamental flaw in the design of growth and value indexes. As we have written previously, there are basically only two types of investors in the market. Fundamental investors buy a stock because they think that its price is lower than its "true" value, and that eventually others will recognize this and the price of that stock will rise. Momentum investors buy a stock because its price has gone up, and they think that because others will be buying the stock, its price will continue to go up. The first thing to notice here is that the first definition says nothing about market/book ratios or rates of earnings growth. A fundamental investor could just as easily judge as overpriced a stock from the "value index" universe as she could judge as underpriced a stock from the "growth index" universe. The second thing to notice is that momentum investors could be just as interested in a stock in the value index universe as they could be in one from the growth index universe – for them, price history is what counts, not market to book or earnings growth. In other words, to put it bluntly, the current value and growth indexes don’t seem to measure phenomenon that are logically connected with the sources of superior investment returns. At best, they may be correlated with these; however, they are not based on them. From this point of view, the most favorable thing that one can say is that value indexes are probably better correlated with the logic of superior investing performance than are growth indexes, and for that reason the former are more likely to outperform the latter.

| In Focus: Style Investing | Real Estate, Liquidity Traps, and Announcements | Performance Update |