Cyclicality of Profit Margins and Why Most Common Valuation Metrics Are Unreliable

Often in the past I’ve discussed current market valuations and implications for market returns over the next decade. In those statements I’m careful to refer only to the most reliable valuation metrics with reliability defined as having the greatest correlation to actual subsequent returns throughout history.

This distinction must be made because it’s the least reliable valuation metrics that often get tossed around by retail investors, the financial media and even professional advisors.

One of the most common valuation metrics to which people refer is price-to-earnings ratio (P/E).

The P/E ratio is calculated by dividing the price of a company’s stock, for example, by the earnings per share over the trailing twelve months (i.e. price to earnings). The purpose is to determine if a stock might be expensive or cheap. It’s basically short-hand for a proper discounted cash flow analysis.

The problem is the correlation of the P/E ratio to actual subsequent returns throughout history is relatively weak compared to other far more reliable metrics.

One of the primary differences between P/Es and other more reliable valuation metrics is that the P/E ratio doesn’t account for the level of current profit margins.

This is important because corporate profit margins are cyclical meaning elevated profit margins don’t persist indefinitely just as very low margins don’t persist indefinitely. Margins are mean-reverting.

See the chart below showing U.S. corporate profits (after-tax) as a percentage of U.S. GDP going all the way back to 1947 to visualize this fact:

We should immediately notice the cyclical nature of corporate profit margins.

Therefore, any reliable valuation metric must account for the level of current profit margins along with the expected trajectory of margins over the period in question (typically 10-12 years) in order to identify likely returns.

Simply taking a valuation at a point in time without accounting for current or expected profit margins will produce incomplete, and likely inaccurate, projections. In other words, using a P/E ratio when margins are very high and then extrapolating those unusually high margins into the future is a fool’s errand and completely ignores history. It’s simply not how margins behave over time so it’s not a realistic analysis.

Over the last several years we’ve seen margins contract considerably from an all-time high of almost 12% to where they’re now approaching the long-term average of around 7%.

​Profit margins are now lower than they’ve been since 2009. Once all the 2nd quarter earnings results are in I’m guessing we’ll observe margins have dropped considerably more than is reflected here.