I often have conversations with clients about valuations, and more specifically about P/E or price/earnings ratios. For some reason, perhaps because these ratios are so easy to calculate and so widely available, investors tend to view them as concrete valuations rather than what they really are, simply estimates. While it is true that historical P/E ratios – those based on historical earnings – are based on actual earnings, stocks don’t trade on historical performance. Instead, stocks trade on expectations for future financial performance. Therefore the P/E ratios that matter are those based on future estimated earnings, and therein lies the rub – anytime we are using estimates, there is room for error.
Earnings estimates come from analysts. Analysts are human beings, and are subject, therefore, to all of the same bias, faults, faulty reasoning, mistakes, etc., as everyone else. Beyond the normal human challenges, however, analysts are affected by other outside influences that tend to skew their estimates further. For example, analysts work for firms. These firms have relationships with most, if not all, of the companies their analysts follow. These relationships put a lot more money in the pockets of the firm than anything the analysts do to generate revenues. As a result, there is incredible pressure on analysts to say positive things about the companies they follow, and more concerning, to not say anything bad about them, no matter how bad things may be. The result is an overall, overly optimistic bias to analyst estimates and recommendations.
In addition to the problems outlined above, what is most worrisome about analyst estimates (and therefore the validity of P/E ratios) is that analysts look at stocks in isolation. In other words they are only concerned with the internal fundamental information about a stock, with no regard for the economic environment, the stock market environment, the sector of the economy the stock operates within, or its industry. Further, analysts have no direct connection with the markets, since they do not trade or manage money.
Adding to these concerns, the quality of the analyst community has fallen dramatically since the tech bubble burst in 2000. Prior to that time, analysts were very highly paid, and had much more autonomy. Many analysts of that era earned seven-figure incomes, and were highly regarded on the street. After the tech bubble burst, and as a result of several high-profile bankruptcies, regulators changed the way analysts could be compensated, and place onerous restrictions on the activities of analysts. The result of all of this was that the top analysts left the major Wall Street Firms for greener pastures – namely hedge funds and private equity funds – where they could get paid. Those that were left, and those that have come after, lack both the skills and experience of their predecessors.
Returning to my central theme, investors in many cases look at P/E ratios as if they are concrete valuation tools free from any error or bias. Nothing could be further from the truth. The S&P 500 has a historical P/E (trailing 12 months) of 17.25X; a mean P/E (trailing 12 months) of 15.5X; and a future P/E of ~13.4X based on $112 earnings and 1,500 S&P 500 level). The problem is the $112 earnings estimate. If this is accurate, one might be able to argue somewhat effectively that the market is relatively cheap. A 15X P/E is typical, and there have been plenty of times when the market traded well above this level in the past. However, this $112 estimate is based on a lot of assumptions, some or all of which may not come to pass as expected. To the extent that this number could come in below $112, the P/E could be a lot higher than the 13.4X. One leading economist has estimated that 2013 earnings for the S&P 500 could be in the $80s. If S&P 500 earnings were $80 instead of $112, the P/E would be 18.75X rather than 13.4X.
As you can see, differences in the earnings estimate can make dramatic differences in the P/E ratio. Investors who think 13.4X looks cheap and therefore that the market represents a good buying opportunity, might not think it is that attractive at 18.75X.
My point is that investors need to understand the way in which valuation measures such as the P/E ratio are calculated, and most importantly that these measures are based on estimates – the opinion of other people, many of which are 25 year-old kids fresh out of college with no practical experience. While it is easy to brush over things like P/E ratios because they have become everyday talking points in the media, it would be a mistake to risk capital relying on the validity of these measures. A greater understanding of what these things actually mean, how they are calculated, and their weaknesses and shortcomings will greatly benefit investors.