Earlier this month, we looked at the question of whether Stocks were Cheap or Expensive. That was a follow up to our glance at how much Earnings and Equities had rallied off the 2009 lows.
The problem is the many ways we can define earnings-per-share. Do we use analysts’ forward earnings estimates? Trailing 12 months reported earnings? GAAP? Operating earnings? Blame the accountant industry for lacking sufficient spine to create a formal, uniform consensus on exactly how we should be defining earnings-per-share.
Which earnings type we use is only the first issue we run into. The next concern is our data set. For example, the Merrill Lynch analysis look at 15 earnings measures, but used data that only went back as far as 1960. Yale University professor Bob Shiller’s database goes all the way back to 1871.
Choosing which data set to work with may create a different outcome. Merrill shows stock prices between moderately undervalued and fairly priced. Shiller shows equities as modestly over-valued. Note the choice we make when selecting data sets can have a very significant impact on the final numbers.
The reality is far more nuanced. We do not know if equities are either over or under valued, because we have no idea what future earnings are going to be. The debate about valuations often turns on which group’s forecast is going to be correct: Expansion or contraction. (We shall ignore the extremeists on both ends of the scale, as they are annoying little twerps who are almost never correct).
Consider how earnings might change in the near future: If the economy were to accelerate (Sequester ends? China improves? Europe recovers?) than earnings could increase significantly, thereby making stocks “suddenly” look cheap. The opposite of this is a US recession, where earnings fall 20-30%, making stocks appear pricey and due for a more significant correction than the 10-19% blips we have plowed through the past 5 years.
Perhaps a better answer to the question “Are stocks cheap or pricey?” is It depends upon what happens in the future.
It is a deeply unsatisfying answer to most people. That is mostly because of its characteristic of being a) true, and b) recognizing the inherent unknowable future as such, and therefor frustrating to those investors who cling to the illusion of clarity.
The Flawed Fed Valuation Model (February 5th, 2008)
Why Using P/E Ratios Can Be Misleading (March 21st, 2012)
Earnings and Equities had rallied off the 2009 lows. (July 22nd, 2013)
Stocks were Cheap or Expensive. (August 9th, 2013)
Price-to-Earnings Ratios Aren’t Always What They Seem
P/E Calculations Based on Differing Views of Earnings Paint Competing Pictures of the Market
WSJ,August 16, 2013