I love David Rosenberg. I read and quote him regularly. For one, he’s an excellent economist. Importantly, he is also practical, a good market observer and is not shy of his opinions. Needless to say, he has been an equity bear throughout this recovery, primarily on the basis of his views on valuations in a highly fragile economic environment. He was spot on regarding the economy but dead wrong on equities. Unfortunately, it is a lot more important and rewarding to be right on the latter than on the former.
All along, I thought that David was erring on both earnings and valuation. He failed to recognized the spectacular turn in earnings and failed to appreciate the impact of low interest rates on valuation. Furthermore, his bearish view on the economy colored his market comments even though he claimed that we should always use trailing earnings in valuing equities.
This morning, David had to acknowledge the recent 3-weeks rally in equities:
Almost one-third of the S&P 500 universe has reported, and the year-over-year earnings growth rate is now running at +28% from +24% last week. Fully 83% of the companies have beaten their bottom-line estimate, which is far above the historical norm of 62%; though barely over 60% are bettering their revenue estimates, which is below average.
(…) it is tough to square the circle of downgraded economic forecasts with what is still quite a spectacular earnings performance.
Unfortunately for economists, equities are not about the economy; they are first and foremost about earnings and, secondly, about valuation.
David is right in generally using reported earnings to avoid forecasting pitfalls. However, using truly depressed trailing EPS in 2009 was simplistic “bearwash” (see S&P 500 P/E Ratio at Troughs: A Detailed Analysis of the Past 80 Years). Then not using truly spectacular trailing EPS in 2010 because of the poor economic outlook was intellectually very weak and self-serving.
Valuation is more tricky because it can involve subjectivity. I have argued that The Rule of 20 is the best valuation tool to use for US equity markets because it incorporates inflation (therefore interest rates as a discount factor) into the equation. The often used historical average PE of 15 is alright on average and over a long enough period but it tends to overvalue equities when inflation is very low. The Cyclical PE approach is also interesting and useful but its long smoothing and inability to account for extremely low interest rates are limiting its usefulness in exceptional periods like this one.
The current reality, that David just acknowledged, is that earnings are continuing to surprise on the upside.
As a result, trailing PEs are 15.2x using current Q3 estimates of $21 (giving trailing 12 months eps of $78.44) and 14.3x using expected TTM eps by the end of Q4 2010 ($83 assuming Q4 = $22 vs $17.16 in Q4 2009).
The Rule of 20, taking into account that inflation is but 1%, says that fair PE should be 19, some 30% better than the current 14-15x.
There remain many hurdles to the economy and future earnings but this is a very cheap market awash in liquidity and about to get an overdose of same liquidity by a truly confused Fed that really needs to keep trying to push the string forward. For a list of caveats see US EQUITIES STILL VERY CHEAP BUT…