I have been asked to repeat some of my past post on the stock market. In particular they wanted to see this long term chart of the S&P 500 price/earnings ratio. Until the 1990s bull market a PE of over 20 on trailing earnings was always a signal of an impending bear market as a PE of over 20 was never sustained. There is a long pattern of the market PE swinging from over 20 to under 10. Moreover, even though many Wall Street strategist talk about the long term average PE of about 15, there is no central tendency for the market to converge on the average of 15 and it really does not stay around 15 any longer than around any other value. Prior to 1926 the data is from Robert J. Shiller’s Irrational Exuberance . From 1926 to 1989 it is from S&P and after 1989 I calculate it from S&P’s estimated operating earnings data.
I use operating earnings because it is what most professional investors use and I believe that the estimate of on going operations is the most relevant measure to value future earnings. Of course, right now it does not make much difference as operating EPS and reported EPS are about the same. Last year’s big difference between the two measures stems largely from the massive write-offs by financial firms . Moreover, much of the recent big snap back in earnings is the write-offs rolling out of the four quarter trailing earnings data.
But as these write-offs rolled out of the data is generated a big drop in the market PE based on trailing earnings to about 19. In my valuation model that makes the market fairly valued to cheap.
In a highly cyclical environment as we are now experiencing many people like to use an estimate of “normalized earnings” to eliminate cyclical earnings distortions. So if you use the 7% trend growth in the above chart of operating vs reported earnings to estimate the market PE you get very similar results. Shiller, and some commentators at this blog use the 10 year trailing earnings. When you are dealing with an individual company where the long term earnings growth trend can change, that rule is a very good practice and builds in a measure of safety. But the long term earnings growth for the entire market is unlikely to change rapidly, so I feel comfortable with this approach.
If the economy is in a Japaneses deflation trap of course the trend earnings would be lower.
So maybe we should look at the prospects for earnings. The recent productivity report received much attention. But I did not see anyone point out that the spread between nonfarm corporate prices and unit labor cost was 5.25%, the widest spread on record.
This spread is the single most important variable driving corporate profit margins and implies that you should expect major positive earnings surprises.
In addition, corporate profits are very much a function of deviations from trend in real GDP growth. So with margins extremely wide and reasonable prospects for above trend growth, profits should continue to rebound strongly.
The other thing investors worry about that could be a major threat to the market is inflation. I’m in the school that believes sufficient excess capacity in the economy, whether measured by the output gap, the unemployment rate or capacity utilization, to keep inflation under control for the time being.
However, there may not be as much excess capacity as many believe. Much of the unemployment is structural . Moreover, the Fed estimates that industrial capacity is actually contracting and the growth of potential real GDP is at record lows.
For this year, I do not expect this to impact many prices outside of basic commodities, but in the out-years it could become a significant problem.
Contrary to many claims, the reason capacity creates a potential problem is weak business fixed investment. Despite low marginal tax rates and record profits real
business fixed investment was extremely weak over the last cycle. I will not go into a whole litany of reasons for weak investments, but I will just point out that in the 1960s, 1980s, 1990s, and the 2000s that capital spending was inversely proportional to marginal tax rates.