Knowing What We Know

Regardless of the claims made by the multitude of pundits and prognosticators on CNBC and elsewhere, no one knows what the stock market will do tomorrow, or next week, or next month (unless they have inside information, which seems to be disturbingly common judging by the headlines lately). The best an honest investor can do is to know that he doesn’t know.

What we do know, however, is that in the long term the market can not go crazy in one direction forever. If the economy grows an average of 4% every year, then the stock market can not forever grow at 8%, nor will it forever grow at 1%. It can do either of those things for years at a time, but as it does it gets more and more disconnected from economic reality. You might say that the market is tethered to the economy with a rubber band – it can stretch the rubber band in either direction for a while, but eventually it snaps back (and usually overshoots in the other direction).

Similarly, the economy itself has, for a very long time, trended to a fairly steady growth curve. It also gets stretched above or below this trend growth for years at a time, but (so far at least) it has consistently reverted to the trend line, again as if tethered by a rubber band.

Some very smart guys at an outfit called Crestmont Research have done a lot of work on this. They have calculated a trend line for S&P500 earnings that has proved to be extremely robust over time. In fact, the trend line in 1975 predicted that 2013 trend earnings would be $79.59, and now with 24 more years of data the trend line is basically unchanged, trending to 2013 earnings of $80.01. You can get all the details at http://www.crestmontresearch.com/pdfs/Financial%20Physics%20Presentation.pdf.

This research also looks at average P/E’s for the market, and finds that when the pricing environment is benign (i.e. there is neither severe inflation nor deflation) then market P/E’s have historically averaged approximately 16x. (Deflation or severe inflation both drive average P/E’s meaningfully lower.) A 16x P/E multiple implies a return of about 6% (1/16) for stocks, which seems reasonable when inflation is low and under control.

We can utilize this knowledge of these trends to craft a successful investment strategy. We can have some comfort level that when inflation is contained then over time the S&P500 will tend to revert back to a price level of around 16X trend earnings. When the S&P500 falls significantly below that, it is a long term buying opportunity, and when it rises significantly above that level it is likely a good time to sell. Of course this does not guarantee stocks will go up after you buy them, since the rubber band could stretch even further to the downside, but over time this approach should outperform the market’s trend line.

So where are we today? Trend S&P500 earnings for 2009 are about $65, and around $68 for 2010. A 16x multiple of these earnings implies an S&P 500 price range of 1040 to 1090. Friday’s closing level was a hair under 1080, so for the long run at least stocks appear to be fairly valued. My gut is that in the short-term the rally will extend upward for a while longer, but every time I think something like that I have to remind myself that in the short run, the best an honest investor can do is to know that he doesn’t know.