| |
 July 2008 Relative to What? Everyone Claims to Add Alpha, but Few Can Measure It
There are definitely some plusses in this. Peer groups are more clearly defined by aligning managers by both style and capitalization. It also makes it easier to assign specific style-based indexes for benchmark comparisons. In addition, it also facilitates asset allocation by dividing managers into the same categories that are frequently used in designing optimal mixes. But there are some downsides, too. Gone are the days when a manager was simply an "equity manager". Investors expect a specialization and failing that, may pass on the manager. It's equally difficult to be a "small cap" or "large cap" manager without some allegiance to a specific style. This is unfortunate because many managers -- arguably the best managers -- are able to deftly shift between styles and capitalizations as market conditions change. Those who strictly adhere to a specific stylebox give up this ability often lowering their odds of success. Those who support stylebox investing believe managers who specialize in one specific corner of the market know that area better than a multicap or multi-style manager would, and thus offer the potential for better returns. Those who argue against stylebox investing see no need to limit a manager -- and the investor's potential returns -- to a specific style, particularly when that style may be generally acknowledged to be out of favor. Why force the manager to pick from a universe of stocks that are expected to underperform? This is an ongoing debate that has experienced a number of twists and turns. Most recently, multicap managers have railed against the practice of classifying them within a particular stylebox based on their current holdings only to later be accused of "style drift" when they move into another style and/or capitalization. The "style drift" label is often a kiss of death given investors' stylebox fixation. It's frequently seen as a weakness to be avoided. Multicap managers probably have a legitimate complaint here. This, in fact, is one of the biggest arguments against over-reliance on the stylebox: Good managers have their hands tied if forced to remain within a specific yet artificially drawn category. There is no room for true multicap managers in a stylebox-driven world.
The Critical Question This is a totally different issue. There's little reason ever presented to believe that a good manager -- especially those in the less efficient corners of the market such as small cap value -- can't consistently produce alpha while remaining true to the category. Arguably, if the category is inefficient, why would a manager ever want to leave? When would it make sense to seek alpha in a more highly efficient sector? Even more fundamental is the concept of multicap alpha. Any time you hear someone talk about "creating" or "generating" alpha, the first thing that should pop into your head is the question, "Relative to what?" Alpha isn't something that just exists out there in the world. You can't point to it. It has to be measured against a specific benchmark and that's where multicap alpha gets interesting. While investors often speak of alpha, you rarely see it defined outside of a statistics class. It's a statistic that gained popularity with the general acceptance of the Capital Asset Pricing Theory. In regard to investment return, alpha is simply a measure of the difference between an investment's actual returns and its expected performance, given its level of risk as measured by beta. As you probably already know, the Capital Asset Pricing Theory postulates that an asset's return is the sum of return associated with the market and excess return not explained by market movements.
This equation breaks the overall asset return (Ra) into two components: (1) The return resulting from that of the market (rm) and the asset's level of market risk (βa), and (2) The asset's excess return originating from other sources (αa). (The error term (є) is generally negligible so is often ignored.) Because market return is the same, differing asset returns are determined by their respective α a and βa values. In essence the, alpha is that part of an asset's return that's not explained by the return of the market and it's sensitivity to that return. That's why it's dependent on "the market" and why alpha is always measured against something. In the original Capital Asset Pricing Model, the "market return" was postulated to be just that: The return of all investable assets. In theory that's OK, but it's not really practical in everyday investing. Instead, specific market indexes are usually used for this purpose. The S&P 500 is often used as a broad equity benchmark while specific capitalization and style indexes such as the Russell 2000 Value or the S&P 400 Growth are used as proxies for specific "stylebox" categories. In general, these are fairly sufficient representatives of the given "markets". Now recall the original question,: Does the stylebox inherently prevent managers from creating alpha? Arguably a good manager in any specific stylebox should still be able to produce value over and above that predicted by the market return and market risk. Through skilled stock picking and trading, he or she should still be able to stay within the bounds of the category and outperform the risk-adjusted index. At most, the stylebox has limited the universe of possible holdings, but at the same time has limited the index (the "market") as well. There's nothing prohibiting a skilled manager from adding alpha.
A Broader Market To be sure, multicap managers can't operate solely within one stylebox category -- if they could, they wouldn't be multicap managers. By definition, they operate in a much broader market.
But once you exit the stylebox, how do you define the "market" and more precisely, what index is the appropriate proxy when evaluating a multicap manager's alpha? Remember, alpha has be be measured relative to something. As suggested above, investors often look to the S&P 500 as a measure of the market. Not surprisingly, it's frequently used as a comparison for domestic multicap managers. In this case, "adding alpha" means providing returns in excess of what would be predicted by the portfolio's beta (calculated against the S&P 500) and the return of the index. Although that sounds reasonable, it's well worth a closer look. According to S&P, as of December 31, 2007, the market capitalization of companies in the index ranged from $511 billion to $710 million. The average was around $25.74 billion, but this was skewed upward because of the indexes cap weighting. The median capitalization was $12.03 billion. Investors generally put the midcap/large cap cutoff around $10-12 billion, so even the median is above this level. In other words, despite the fact that the S&P 500 is often viewed as a broad market index, it's really a large cap index. That's more than just a matter of semantics when it comes to multicap alpha. Managers won't often tell you this, but the easiest way to beat an index is by investing outside of it. In this case, if you were to limit yourself to the S&P 500, you would still have 500 stocks to choose from running the gamut from $511 billion all the way down to $700 million. However, most of the benchmark's return is going to come from the top -- the largest companies. Remember, the top ten are about 20% of the overall index. The stocks at the bottom aren't really going to have much of an impact. Whether you wanted to or not, you'd have to hold stocks in that top ten when the index was headed up, and you'd have to avoid them when it was going down. More importantly, if you wanted to add alpha, you'd have to carefully pick the stocks to hold and the ones to avoid. This decision as well as the decision to jump in and out of specific stocks would be the only way you could add value. It's generally accepted that most of an investor's long-term return comes from asset allocation rather that stock picking and timing. In fact, the latter two sources -- all you'd have at your disposal if you limited yourself to investing in the benchmark -- would account for about 7% of the long-term return. You'd better be an excellent stock-picker and timer if you plan to add value under those conditions. Of course multicap managers aren't limited to investing in the index. They can look for alpha anywhere, but with the S&P 500 serving as a good proxy for large caps, most look in other capitalizations. But here's where it gets tricky: If a manager's portfolio is made up completely of small and midcap stocks, is the S&P 500 truly the correct benchmark index? S&P apparently doesn't think so, that's why they also provide the S&P 400 index as a midcap benchmark and the S&P 600 for small caps. Remember alpha has to be calculated relative to a "market" or index proxy. If the S&P 500 really isn't a good benchmark for wandering multicap managers, what does it mean to say they "add alpha"? Are you really "adding alpha" if you're investing in asset classes not really measured by the benchmark? Multicap managers who acknowledge this problem occasionally measure their performance (and alpha) against broader benchmarks such as the S&P Super Composite 1500, the Russell 3000, or the Dow Jones Wilshire 5000. This may be better, but capitalization weighting and market float still impact the results. For example, although the S&P 1500 contains all the stocks in the S&P 400 and S&P 600 in addition to those of the S&P 500, it's cap weighting still leaves it highly correlated (.997) with the S&P 500. The point here isn't that all multicap managers are trying to put one over on the investing public. Those who seek other, broader measures of the market, are proof of that. It's also not an argument that multicap managers don't really add value. On the contrary, many probably do, you just can't simply measure it -- and that's the point. Too many managers from all styles claim to "add alpha" without really answering that all important question, "Relative to what?" The one thing that is certain in all of this is that stylebox pigeonholes are not keeping them from doing so. The stylebox may be a convenient excuse, but in this case that's all it is, an excuse. Search this site! Just enter you key word or words:
Get current quotes or follow your own custom portfolio,
courtesy of E-Line Financials:
|
||||||||||||||||||||||||||||||
HINK WHAT YOU WILL OF Morningstar, but the equity "stylebox" has had a lasting impact on investing. You're doubtless familiar with it, the matrix with three rows representing large, mid, and small capitalization, and three columns for value, blend, and growth styles. Morningstar developed it a s a means of classifying mutual funds, but now almost all equity managers are typically forced into it.
