Monday’s WSJ gives lengthy treatment to the scramble among analysts to work out whether stocks are cheap considering the uncertainty about the path of US corporate earnings.
There’s plenty of disagreement among the sell-siders and cheapness is itself a hazy concept. Using the Shiller-based P/E ratio of comparing prices against historical earnings going back ten years, stocks would appear to be, if not too expensive, then nothing like a bargain either. But shift to projected earnings and ratios still look mighty appetising — though of course, much will depend on whether the “projected” bit turns out to be accurate.
We’ll stick to the “E” side for the rest of this post. The pessimistic view on earnings, which we noted in June, has been that profit margins must eventually compress: the productivity gains from the last couple of years have been driven mainly by cutting costs and are unsustainable. We’re already seeing productivity growth slowing down and, as Doug Cliggott of Credit Suisse told us on Monday, unit labour costs have recently begun to rise.
The big difference between what happened in the past two years and what will happen going forward, says Cliggot, is the absence of deficit-fueled government support:
When we try to put all of this in a macro context, there were extraordinary cost savings by companies in 2009 and 2010. Those had a tremendous impact. But what happened also was that the federal government was simultaneously stepping in and providing some demand. Revenue growth would have been much weaker if we’d had both unit labour costs declining and no support from the government. And now you’ve got state and local governments cutting back, which we think will continue a bit longer. The second stage of tightening is at the federal level, and for a couple of years, from a profitability standpoint we had this sweet spot where corporations could be as aggressive as they could on the cost front, but with the government stepping in to support demand. Now those government levers are reversing, perhaps violently, and instead of unit labour costs declining, they’re now going up.
The current profits expansion is already nearing the average historical duration of its predecessors. Add to this renewed fears of a double-dip recession, the absence of healthy employment growth, and stagnant private sector credit creation, and you have a nice recipe for further earnings disappointments.
For what it’s worth, Credit Suisse is now pegging S&P 500 aggregate earnings to close the year at $95 a share, and to stay roughly flat next year. That’s considerably less than the $113 a share for 2012 that other analysts are expecting, according to Capital IQ (hat tip WSJ).
So is there a countervailing optimistic view? Well, there’s always the obvious, which is that fears of recession (or even of a continued decline in the pace of economic growth) could always be overblown and predicting the future is invariably a precarious endeavor.
But there also seems to be a decent case that even if we’re headed towards a recession, earnings might not suffer too badly. This argument, from BarCap, is also based on lessons from the past — beginning with the very recent past:
In addition, an overlooked fact of the recent GDP revisions was that nominal GDP during the recovery was actually revised higher; this includes the negatively revised 1Q11 (which seemed to be the most important data point in a global rerating of US trend growth). This is important because as flawed as these data appear, it is nominal GDP that correlates reasonably well with corporate profits and revenues, not real GDP. A little bit of extra inflation is a good thing for corporate profits (which are of course nominal results).
Going a little further back, BarCap looked at all post-war recessions other than the most recent two (which it describes as “asset bubble recessions”) and finds the average earnings decline to be a tame 9.4 per cent:
BarCap analysts, which expect S&P 500 earnings per share would be close to $90 next year if there’s a recession, add:
[An] interesting element of the history of recession-related earnings declines is that the average peak in annualized trailing earnings came five months prior to the beginning of the recession. In other words, for all the talk about the risk of slipping into another contraction, earnings would typically already be falling. In only two cases were they still rising when the recession began: the 1974-75 oil shock and the 1981-82 monetary policy-induced recession. Earnings are, in fact, still rising. For the year-on-year growth rate to fall, 3Q11 earnings would need to miss the bottom-up consensus forecast by 14% (3Q11 would need to come in below the 3Q10 result of $21.75). Keep in mind that earnings this quarter (2Q11) are beating consensus by 5.7% (89% of companies have reported results), so this would be quite a turn in fortunes for an unleveraged corporate sector. In addition, in the 1953-54 recession, earnings actually increased; in the sharp contraction of 1949, earnings fell only a meager 2.5%; the deep energy price shock recession of 1974-75 resulted only in a 7.2% drop (energy is another potential catalyst for a contraction); and in the deep recession of 1981-82 (in which unemployment went even higher than the Great Contraction), earnings dropped 13.4%. So while the ‘E’ in the PE is hardly cast in stone, a stroll down memory lane supports our valuation argument.