llwk, there are some problems with referenced link that I think you need to understand. If you have a basket of stocks that are fairly valued with stable growth you can expect the the stock price appreciation rate to equal the growth in eps. That's no big deal, and has nothing to do with the details of the baseline valuation metric. But the problem with that approach is that it fails to take into account risk. Clearly, high growth stocks are riskier than their lower growth brethren. That means that there must be a "premium" attached to those stocks to take into account their riskiness. You will notice that the poster simply assumed that the probability of the three hypotheticals achieving expected results were the same. But by using a PEG approach the EV (expected value) of the investments are the same and this supposedly reflects the greater relative riskiness of the higher growth stocks.
I have been critical of the PEG approach for several reasons. One is that it doesn't reflect the risk of the position, but simply assumes that risk and expected growth are directly related. Second, it ignores interest rates as a factor in determining stock valuation. This is a much more serious flaw.
To get around the interest rate issue I have employed what I call CNPEG (Chuzzlewit's Normalized PEG). It works like this: you calculate the YPEG for the stock in question. The YPEG is the leading PE divided by the consensus long-term growth rate. Next, you calculate the YPEG for the S&P500. Finally, you divide the stock's YPEG y the market's YPEG to give you the CNPEG. I interpret this number as the relative cost of growth. However, I am unhappy with the final number because it doesn't capture the relative risk. At this point I have no good surrogate for risk, although I am toying with beta -- i.e., multiply the CNPEG by the beta of the stock to attempt to capture risk in the price.
TTFN,
CTC |
