In MarketBeat’s Stocks Too Cheap? Or Earnings Estimates Too Rosy? (subscription required), Matt Phillips has an interesting article on forward price-to-earnings ratios:
The forward-earnings in question are really consensus forward-earnings estimates, churned out by Wall Street’s broker dealers, the so-called “sell side.” And the “sell-side” as a whole tend to be a bullish bunch.
Phillips takes Oppenheimer Asset Management analysts to task for their Monday report:
With that in mind, we turn to the question of whether stocks are cheap right now, or not. Some look at the numbers and argue, of course they are. On a price-to-forward-earnings basis, you can make that argument. Oppenheimer Asset Management analysts did so Monday. They write:
The S&P 500’s current forward multiple of 12.9x represents the lowest level for the index since early 1995. According to our analysis, market performance has been relatively strong in all periods following similar forward P/E levels. The S&P 500 averaged a roughly 12% 12-month return in all periods since 1960 when the forward P/E was between 10x and 15x.
In the same note, Oppenheimer says that some clients are starting to wonder whether analyst expectations being too high could be the reason stocks look so cheap. “Based on our client conversations, there appears to be a growing concern that 2010 earnings expectations have become too optimistic and that forward price multiples are thereby understated,” Oppenheimer writes. Oppenheimer analysts counter this concern saying the firm doesn’t “find earnings expectations out of context from a normalized growth perspective,” writing that after a recession as sharp as the one we had, it’s reasonable to expect the earnings growth rate to be a bit higher than usual as the economy recovers.
Phillips points to Robert Shiller’s calculation of Graham’s P/E10 ratio as evidence that stocks are not be as cheap as Oppenheimer’s one-year forward estimates might have us believe:
Prof. Shiller tracks P/E ratios back to the 19th century, smoothing out short-term ups and downs in profits by using a 10-year earnings average. Even after the May declines, his P/E ratio still is almost 20, well above the historical average of about 16.






