It's very hard to outperform the S&P 500 on an ongoing basis. Given that the S&P 500 has no taxes, commissions or slippage, an investor has to take larger risks to beat it. As we discussed above, sector concentration increases the risk of a portfolio but also the chance of outperforming the benchmark. In measuring a portfolio's risk, we track several parameters as compared to the S&P 500: dividend yield, price-to-earnings ratio, projected growth rate, price-earnings/growth ratio, or PEG, and Beta. You can get these values for the most part from the TSC site. You can also get analysts' five-year forward growth estimates and PEG ratios from Zack's Web site or from Yahoo! Finance The PEG ratio can also be calculated from price-to-earnings and growth rates.

The yield of a portfolio is practically an anachronism in an era where the yield on the S&P 500 is below 1%. It is still true that a portfolio with a lower yield than the S&P 500 has higher risk, however, so we still track this information.

In our example, we used the P/E ratios obtained from trailing 12-month data, which is the way this information is most commonly presented. With a little extra effort you can substitute forward P/E ratios (from Zacks, for example).

Important point: Any company with a P/E ratio greater than 99 or showing a loss (so P/E is N/A) is converted to a P/E of 99. A P/E of 99 implies you have to wait 99 years to earn back every dollar invested (unless earnings increase as rapidly as investors hope). From a discounted cash flow point of view, a P/E of 99 is the same as a P/E of 999 or 9999. (The latter cash flows have a negligible net present value.) If we don't do this conversion, the weighted average P/E of the portfolio won't make sense.

Zacks and First Call compile analysts' projections of how quickly companies will grow over the next five years (and summarize the estimates at the industry level and for the entire S&P 500). These estimates are wild guesswork for the most part. (Who predicted five years ago the impact of the Internet on corporate growth?) However, the relative comparisons of these growth rates are useful. (It makes sense that Cisco (CSCO:Nasdaq) has a forward growth-rate estimate of 29.6%, while Consolidated Edison (ED:NYSE) has an estimate of 3.1%.)

The PEG ratio is a parameter which justifies a stock with a high P/E ratio as long as the forward growth rate is high enough. It is most commonly calculated as the one-year forward P/E divided by the one-year forward growth estimate. As with the five-year growth estimate, precision is less important than the relative comparison, so we're just as comfortable calculating this ratio with the trailing 12-month P/E and the five-year forward estimate in the spreadsheet.

Lastly, we obtain Beta where available. Beta is an estimate of correlation between a stock's price and the S&P 500. A stock with a Beta of 1 will on average gain 1% for every 1% move in the S&P 500. High-performance stocks, such as Internet stocks, have Betas of 2 or even 3, which is good when the market is moving up -- not so good when the market is moving down.

Beta is generally calculated with 60 months of data; stocks with less than two years of data often have N/A for Beta. In that case, substitute a Beta of 1.5 or 2 for initial public offerings, 1.25 in all other cases. Generally, Betas of individual stocks are not reliable predictors of future gains relative to the S&P 500. However, Betas of groups of companies (e.g., industry groups, portfolios) do have predictive value because company-specific "noise" cancels out.

Conclusion

In comparing our example portfolio to the S&P 500, we see:

Sample Portfolio vs. S&P 500
	YTD return	Dividend yield	P/E Ratio	5-Year growth rate	PEG ratio	Beta
Example portfolio	69.4%	0.4%	68	21%	3.4	1.43
S&P 500	19%	1%	32	13%	2.6	1.00
Relative value	265%	-58%	110%	66%	31%	43%
Source: Heron Capital Management

Beta would have predicted returns 43% higher than the S&P 500 (about 30%), but actual returns were much higher. It's not surprising that the dividend yield was substantially lower. The portfolio's P/E ratio at 68 is substantially riskier than the overall S&P 500 and, unfortunately, the forward expectations of growth do not justify the high P/E ratio. Only if the portfolio had a lower PEG ratio than the S&P 500 would we feel comfortable leaving the portfolio alone. Instead, we will take profits on about 50% of our technology exposure and will invest the proceeds in companies with valuations more in line with future growth expectations.