Adapting Beta for Assessing Tech Stocks

Tools are means to an end and we ought to adapt them to meet our ends. Beta is just such a tool. Below Jim Pyke writing in Seeking Alpha presents this interesting analysis......

For investors who ascribe to modern portfolio theory, beta represents a key measure of risk. However, does beta always provide the right insights into thinking about a given security? Conventional wisdom often suggests that high growth technology driven companies are often risky and hence would have pretty high betas. However, thinking a little more deeply about the situation, investors should realize that the success of technology companies is often based upon innovation and creativity - factors that have essentially no correlation to the market performance. These two factors - volatility and correlation - are what drives beta calculations. Most investors learn that beta is the covariance of the market's returns and the security's returns relative to the variance of the market's returns. It sounds nice; however, this does not really provide any intuition around beta. Rearranging the mathematical terms gives beta to be the product of two numbers:

1. The ratio of the security's volatility to that of the market's volatility
2. The correlation of the security's returns to the market's returns

Thinking about beta in these terms provides more intuition. The following list of stocks was used for the analysis:

Source: data provided by Zacks.com services. Industry classification is not exact but focused around ensuring a good mix of companies. IBM stretches across all classifications from services to hardward to software.

This list provides good cross section of companies often categorized as technology companies - it does exclude telecom service providers such as Verizon (VZ) and AT&T (T) as well as biotechnology companies. The expectation is that these companies will show a wide range of betas moving from companies like IBM and MSFT, which should probably have betas relativiely close to 1 given their size and broad scope. Their fortunes are more tied to the market than other companies'. However, if the notion of simply looking at beta for technology companies as a poor measure of risk, I'll also find companies that have similar betas but differing volatilities and correlations. Two companies might both have a beta of 1, but one might have just a 50% correlation and a volatility ratio of 200% while the other is perfectly correlated to the market and has the volatility as the market. Their impact on portfolio volatility would thus be different.

I used SPDR S&P 500 Trust ETF as my market portfolio benchmark. I also compared both the tech heavy PowerShares QQQ Trust (QQQ) and the Technology Select Sector SPDR ETF (XLK). I looked at beta using 36 month cycle with data from Yahoo!Finance. In a previous article, I found that 36 month calculations correspond quite closely to what Yahoo!Finance posts under its key statistics section for stocks. The following table shows the break down of betas for the securities selected:

Source: Derived from data from Yahoo!Finance using split and dividend adjusted monthly prices. Implied beta calculations are based on author methodology and should be close to the betas from Yahoo!Finance.

The first observation is that essentially every single technology company selected shows a volatility higher than SPY. However, almost half have a beta below 1 due to relatively low correlations. In fact, may of these individual securities have lower correlations to SPY than broad based ETFs from emerging markets. The next observation is what I hypothesized to find that technology companies might have similar betas, which according to the capital asset pricing model, would thus have the same hurdle return rate; however, those companies' betas could have substantially different volatilities and correlations.

A clear example is INTC, LXK and AAPL, which all have 1.1 betas. However LXK has about a 50% correlation to SPY, while INTC's is 77% and AAPL's is 71%. The most remarkable beta belongs to NFLX. Many investors see NFLX as a potentially heavily overvalued stock, however, it has a tiny beta of just .3x. The capital asset pricing model would suggest that the appropriate equity hurdle rate is 2.9% + .3 x 7% = 5%. Personally, I would want a lot more than a 5% return to be willing to invest in NFLX. In contrast, MSFT would have a 9.9% equity hurdle rate according to the capital asset pricing model.

Conclusions

When looking at these technology companies, I don't think beta is a really a great measure of risk without digging further into the numbers. While I believe there is significant value to the tools provided by modern portfolio theory, it is always important to dig deeper and truly understand the factors that are driving the high level metrics.

Based on this analysis alone, I think some interesting opportunities might be with GOOG - a relatively low beta (<1) is driven by a low correlation and reasonably low volatility ratio. GOOG continues to innovate and grow. From a risk perspective this could be an interesting opportunity if the valuation seems reasonable.

Also based on this, I would see DELL as a less interesting opportunity. A leading player in providing computer hardware, I would be concerned by a relatively high beta driven by a pretty high volatility ratio. NOK also looks unattractive from this analysis due to its very high beta of 1.7x. However, NOK has been struggling greatly in the market and is being battered. A quick look at its financials shows a net cash position (cash - debt) of around $10 billion with an equity cap of around $23 billion. Does NOK become a potential long at some point?

I would also probably take AMZN over EBAY given both are providing retail options on the internet. AAPL appears to be more compelling than SNE.

This analysis also shows that QQQ and XLK, which are more focused on technology than SPY, have very similar characteristics to SPY. An investor seeking diversification through them would probably be disappointed.

It is important to caution readers that my conclusions are essentially based upon the analysis in this article. It doesn't consider any fundamental analysis or strategic assessment of the company within its market space. I believe strongly that this additional work is critical prior to making any investment decision. I would want to find a stock that seems to carry the benefits of low correlation/low risk as well as a strong market positions, compelling products/services, and low valuation. This analysis also did not assess returns, which on a historical basis can often have no correlation to the stocks beta. This further emphasizes the need for fundamental analysis and research. For example, YHOO, DELL, CSCO, LXK and SNE all have 5 year returns around -45 to -50%. ORCL's 5 year return is around 60%, which shows great contrast of MSFT's return of -11%.

Disclosure: I am long SPY.

Disclaimer: This article is for informational and educational purposes only and shall not be construed to constitute investment advice. Nothing contained herein shall constitute a solicitation, recommendation or endorsement to buy or sell any security.