Analysis of New High Beta and Low Volatility ETFs from PowerShares
Invesco PowerShares brought out two new and unique ETFs last Thursday (5/5/11). PowerShares S&P 500 High Beta Portfolio (SPHB) and PowerShares S&P 500 Low Volatility Portfolio (SPLV) are innovative products that employ quantitative beta-weighting and volatility-weighting as part of the underlying index construction. As with any new investment strategy, you need to understand how these new ETFs will function before putting them in your portfolio.
- PowerShares S&P 500 High Beta Portfolio (SPHB) (SPHB overview) tracks the new S&P 500 High Beta Index, which consists of the 100 stocks from the S&P 500 with the highest sensitivity to market movements, or beta, over the past 12 months. The 100 stocks are weighted proportional to their 12-month beta coefficient at each quarterly rebalancing.
- PowerShares S&P 500 Low Volatility Portfolio (SPLV) (SPLV overview) tracks the new S&P 500 Low Volatility Index, which consists of the 100 stocks from the S&P 500 with the lowest realized daily volatility over the past 12 months. The 100 stocks are weighted inversely proportional to their 12-month realized volatility at each quarterly rebalancing.
Beta & Volatility
Beta is one of the most misunderstood investment terms, and it is often incorrectly assumed to be a word that is interchangeable with volatility. However, that is not the case. To understand these new products, investors must understand the difference between volatility and beta. A thorough discussion is beyond the scope of this article, but I will attempt to summarize here: Volatility describes fluctuations in the price of a security or index independent of anything else. Beta describes the movement of a security relative to another security or index (such as the S&P 500) and is a function of the correlation between them.
A good example is gold mining stocks – they are typically much more volatile than the S&P 500 while also having a very low beta. This is because while mining stocks tend to have stronger price movements than the S&P index, they rarely move in the same direction and speed as the index at the same time, resulting in a low correlation. When beta is measured against a broad market index, such as the S&P 500, it can be interpreted as a measure of market risk (systemic risk).
Analysis of SPHB
The underlying S&P 500 High Beta Index is brand new, launched less than a month ago, but the High Beta Index fact sheet (pdf) provides “selective” hypothetical back-tested data covering 20 years. A graphical representation of the 60-day rolling beta appears to average about 1.75 while ranging from 0.95 to a high of around 2.70. However, for the 11 calendar years from 2000-2010, the annual return was -4.3% at a standard deviation of 42.1% versus +0.4% at 20.5% for the S&P 500.
The four largest sectors currently are Financials 30.5%, Consumer Discretionary 18.7%, Energy 18.2%, and Technology 12.9%. The largest individual holdings include Goodyear Tire & Rubber Co. (GT) 1.4%, Cliffs Natural Resources Inc. (CLF) 1.4%, Allegheny Technologies Inc. (ATI) 1.4%, Marshall & Ilsley Corp. (MI) 1.3%, and Abercrombie & Fitch Co.- Class A (ANF) 1.3%. The S&P back-test does not provide any data regarding portfolio turnover or the sector allocation changes over time. I expect turnover to be relatively high for this ETF.
With volatility more than twice that of the S&P 500, long-term performance trailing by 4.7% a year, and drawdowns of about 70%, SPHB is not a suitable buy & hold ETF. It will likely make a good trading vehicle for investors that can successfully catch the uptrends and manage the risk.
Analysis of SPLV
The underlying S&P 500 Low Volatility Index is also new, and the Low Volatility Index fact sheet (pdf) provides “selective” back-tested data. Unfortunately, it appears S&P deliberately used different periods (yearly data ending in November) in order to avoid direct comparisons with SPHB.
A graphical representation of the 1-year rolling volatility (standard deviation) appears to average about 5% less than the S&P 500, although there are periods when it is actually higher than the benchmark. Using the same 11 calendar years as above (2000-2010), I calculated the annual return to be +7.8% with a standard deviation of just 14.4%. As previously mentioned, the S&P 500’s return during this period was +0.4% at 20.5% standard deviation.
As you can probably surmise, the sector allocations for SPLV are nearly opposite those of SPHB. The four largest sectors presently are Utilities 29.1%, Consumer Staples 28.3%, Health Care 10.2%, and Financials 7.8%. The largest individual holdings include Johnson & Johnson (JNJ) 1.4%, Southern Co. (SO) 1.3%, Abbott Laboratories (ABT) 1.3%, Consolidated Edison Inc. (ED) 1.3%, and Clorox Co. (CLX) 1.3%.
The long-term performance chart looks similar to many “Value” funds, especially the drastic underperformance that occurred in 1998 and 1999 while the overall market was experiencing one of its best two-year periods ever.
If SPLV can live up to its back-test, then it could potentially become a core holding for many portfolios. The +7.4% annual outperformance of the past 11 years should not be expected going forward, as that period is guilty of some cherry picking by not including 1998, 1999, and prior years. Still, the ability to outperform the S&P 500 at lower volatility over extended periods is very attractive.
Despite the risk-related short-comings of SPHB, these two new funds are a welcome addition to the ETF marketplace. Using beta and volatility to filter the universe and weight individual holdings make these products unique. The characteristics of SPLV make it potentially suitable as a buy & hold fund while SPHB should be viewed strictly as a trading vehicle. Both funds carry an expense ratio of 0.25% and share a common prospectus (pdf).
Disclosure covering writer, editor, and publisher: Long SO. No positions in any of the companies or ETF sponsors mentioned. No income, revenue, or other compensation (either directly or indirectly) received from, or on behalf of, any of the companies or ETF sponsors mentioned.