GMO's James Montier has just released his latest investment outlook, and given his well-known behavioural science background, makes for an interesting read. Below is an excerpt of the report, followed by the entire document in the slidedeck, which you can fullscreen for the better read, or download.
What Goes Up Must Come Down!
by James Montier, GMO
March 2012
A Little Detour into My Murky Past
Nearly a quarter of a century ago, I was a young, naïve, and foolish believer in an economic concept known as rational expectations – an elegant, mathematically beautiful theory with no practical use. In Star Wars parlance, I had effectively been seduced by the dark side. Thankfully, several of my university lecturers were determined to save me from this terrible fate. They insisted on teaching me a very wide variety of approaches to economics including the Marxist perspective and the something known as post-Keynesian macro. I owe them a huge debt of gratitude.
I thought at the time that these were at best esoteric distractions. Little did I know that they were going to provide some of the most profound insights into financial markets, illuminating many of the flaws that conventional thinking ignores. Early on in my career I was fortunate enough to interact with a number of colleagues who used some of these tools to uncover observations that the mainstream had completely missed. This made an indelible impression upon me, and these tools are still the ones I reach for when faced with trying to understand the world.1 Profit Margins as a Case in Point
Today I find myself once again digging through this toolkit, searching for a way to understand the development of profit margins. Currently, U.S. profit margins are at record highs according to the NIPA data (see Exhibit 1). More freakish still is that these record high profit margins are coming during the weakest economic recovery in post-war history.
At GMO, we are firm believers in mean reversion, and as such record elevation in profit margins causes us much consternation. Of course, we are constantly on the lookout for sound arguments as to why we might be wrong in our assumption of margin reversion. After all, believers in mean reversion are always short a structural break, and such a break clearly matters. For instance, Exhibit 2 shows that in simple trailing P/E terms the U.S. market isn’t actually expensive. However, the P/E is only one part of a valuation – it also depends upon the state of earnings. It is the margin component that is dragging our return forecast down. If we are incorrect on our assumption of mean reversion in profit margins, then our forecast radically alters. For instance, if instead of falling to 6% over the next 7 years margins stayed at today’s levels, our forecast would be closer to 4.5% p.a.
Clearly the first two elements of Exhibit 2 are all about cyclical adjustment: we are assuming that the market goes to a “normal” P/E based on “normal” E. Therefore, it is no surprise that we see the same point from a different perspective when we look at a comparison of the simple trailing P/E using the Graham and Dodd P/E (Exhibit 3). The latter tries to smooth out the business cycle’s impact upon earnings by using a 10-year moving average of earnings. Hence, differences between the two measures are a statement of how far earnings are from their “trend.” The simple trailing P/E is around 15x and the Graham and Dodd P/E is around 24x, again highlighting the divergence of profits from their long-run normal levels.
Whilst we at GMO fret over evidence of the strained nature of profit margins, the ever bullish Wall Street analysts expect profi t margins to continue to rise! Witness Exhibit 4. In our search for evidence of a structural break, this simple-minded extrapolation gives us some comfort because the Wall Street consensus has a pretty good record of being completely and utterly wrong.
