The true cost of beta: Part One

There are two components of an active portfolio’s return. Beta is passive and stems from simply being invested in the market. The other, alpha, stems from a manager’s skill in selecting investments that will add return above what the market gives on its own.

Beta drivers

The beta described in the example below is referred to as the classical beta. This is the standard beta that measures the exposure of a security or portfolio relative to a major index and serves as a measure of the security’s or portfolio’s exposure to systematic risk in the equity market. But beta is actually a continuum that starts with the pure classical beta and continues on to include other types of beta.

Classical beta has zero active return and a corresponding zero active risk exposure. As we move up along the continuum, active risk exposure increases, as does the possibility of active return. Here are some of the betas we find along that continuum.

Bespoke beta is a measure of exposure to local risks such as sector, country and style. (Style is beta, not alpha.) Sector-specific ETFs, such as iShares XEG, that cover the Canadian energy industry are good examples of local beta products, as are country or region ETFs such as Vanguard’s Emerging Markets ETF, as well as ETFs that cover a specific capitalization or style subset of the market.

Alternative beta is a measure of exposure to unusual systematic risks. Examples of alternative beta would be exposure to foreign currencies, commodities and real estate. Alternative betas expand systematic risk exposure beyond typical stock and bond portfolios.

Fundamental beta is meant as an alternative to the usual capitalization-weighted index. These indexes weight stocks within the index based on factors such as the price/book or dividend ratio.

Claymore’s suite of RAFI Fundamental Indexes is a good example of fundamental beta, as are ETFs invested exclusively in high dividend-paying stocks.

Active beta, also called enhanced indexing, tries to earn excess return in the markets using quantitative strategies resulting in portfolios whose return is mainly beta-driven, but use long/short strategies to a varying extent. 130/30 products belong to this type of beta. The beta of 130/30 (or any combination ranging from 0/100 to 200/100) is always 1, as with the market, but may allow for more alpha generation through their reduced short-selling restrictions. These are sometimes called beta 1 products.

Bulk beta includes the typical actively managed mutual fund. Although we would expect an “active” mutual fund to have a healthy dose of alpha in its performance, the majority of the typical mutual fund’s return is generated by factor exposure (beta) rather than active return factors (alpha). Bulk beta investments are highly correlated to their benchmarks. Also, since these products are basically packaged systematic risk exposure, they have access to the whole of the underlying market’s liquidity and therefore have a large capacity for AUM.

Beta first, then alpha

Alpha is the excess return that one is left with once all the beta-driven return has been accounted for. Let’s look at our example of a fund manager who beat the S&P/TSX beta-driven return by 2.39%. What if the manager was mostly invested in small cap stocks? Is the S&P/TSX index still an appropriate benchmark?

A better benchmark would be a small cap index. But given that alpha derives its value from what has been determined as beta factors, changing the reference beta means that we will wind up with a different result for alpha. The return of the S&P/TSX Canadian Small Cap Index for the year was 38.53% in 2010, which is the more relevant beta-driven return we should be looking at in evaluating the manager. A small cap manager with a 20% return would have greatly underperformed the appropriate benchmark.

The average return of small cap managers covered by Globefund in 2010 was 26.22%, or 12.41% under benchmark. Comparing firm size is a very important return factor that should be considered in defining a portfolio’s beta and therefore the manager’s skill in generating alpha. In this particular case, the manager could be seen as skilfully generating 2.39% of added return, or as a not-so-skilful manager trailing his benchmark by 18.53%.

Perhaps our portfolio manager does stick with large cap stocks, but has a value-oriented management style. He looks for securities he believes have temporarily fallen out of favour, or whose true value has not yet been discovered by the markets and whose price is consequently below what he considers to be fair value. For the year 2010, the S&P/TSX Canada Select value and growth indexes respectively had a return of 16.67% and 19.30%. The appropriate benchmark return for our manager in this case would be 16.67%, rather than the S&P/TSX’s 17.61%.

In this instance, choosing the wrong benchmark actually underestimated the manager’s skill at generating alpha-driven return.

Alpha is a tricky thing to measure properly. Its correct measurement depends on being able to identify all the beta-driven performance factors beforehand. Unidentified beta drivers can be mistakenly interpreted as manager skill. In addition, alpha is not static. Some investment strategies that had once been considered as alpha have been found to be previously unidentified beta, such as firm size and style. For example, a value manager may have beaten the S&P/TSX Composite over a number of years, but how did he fair against the Dow Jones Canadian Select Value Index?

To read part two of this story, click here.

Guy Lalonde is an investment advisor and portfolio manager at National Bank Financial and is based in Pointe Claire, Quebec.