The Market, Valuation and Future Earnings

I was playing with some numbers I got off Robert Shiller’s website. I was curious to see historically how long it’s taken the market to earn back its value. The P/E ratio is concerned with past earnings, but I wanted to see how good the market is at valuing future earnings.

It turns out, the market is generally worth about its earnings for the next 40 quarters, or 10 years. This makes sense since the historic P/E ratio is around 14-16, so going by normal growth, it ought to take roughly ten years to earn your money back. Let me add that the market has been known to be wrong about such things.

Here’s a look at the S&P 500 compared with its earnings ten years hence. Because of that restriction, the chart has to stop in the late 1990s.

From the 1929 peak, it took the market 24 years to earn its money back. In 1942 and 1982, it took less than seven years.

Comments

[AuGod!:]

For some reason lately I have been thinking about negative P/E ratios. I am told by Investopedia that it is possible but probably unreportable. I quote:

"Mathematically, there are only two ways for a ratio of this form to have a negative value:
1.The numerator falls below zero
2.The denominator falls below zero.
In the case of the P/E ratio, it is impossible for the numerator to fall below zero because this represents the price of the asset. However, the denominator, which is equal to the earnings of the company, can become negative. EPS values below zero mean that the company is losing money and is the reason why it is possible to have a negative P/E ratio.

Negative EPS numbers are usually reported as "not applicable" for quarters in which a company reported a loss. Investors buying a company with a negative P/E should be aware that they are buying a share of a company that has been losing money per share of its stock." (End quote)

So, I ask two questions.: What are the implications of this? Does it matter?

Sep 03 08:00 AM

[Pauly B:]

Good article.

Another question, What is the PE of Congress and their insane spending? That should offset any positive earnings in the S&P.

Sep 03 09:10 AM

[Isnt_It_Jus...:]

This article is another way of saying that the "PEG", P/E divided by average earning's growth over the next 10 years of the S&P, should equal 1.

Sep 03 10:15 AM

[Mad Hedge F...:]

xvcnm. “The total breakdown of the system is ahead of us. It may come in four, five, or ten years, and it will devastate the world economy. By bailing out the issuers of derivatives, the Fed actions have only postponed the day of reckoning,” said Marc Faber, publisher of the Gloom, Doom & Boom Report.

Sep 03 10:43 AM

[Mad Hedge F...:]

zxcbn. US stocks are now the most expensive they have been in seven years, and never really got cheap during the March low, just fairly valued. At least I have some good company in my views, which are also shared by David Rosenberg of Gluskin Sheff, the former economist at the late Merrill Lynch. The “faith based” rally is now discounting a GDP growth rate of 4.0%, which has a snowball’s chance in Hell of actually occurring. This is up dramatically from the 2.5% growth rate the S&P 500 was discounting when the index was at 667. The best stock market rally since 1933 added an unprecedented eight PE multiple points to stocks, and there is now more risk in the market than the 2007 peak. Underweight portfolio managers and momentum driven day traders are to blame. It’s what happened after the 1933 rally that scares me. Needless to say, stocks offer no value here. You can sign up for David’s well thought out research for free by going to his website at www.gluskinsheff.com/.

Sep 03 10:44 AM

[Keep your p...:]

Interesting analysis. However, an even better analysis would be to include dividends in your "forward payout stream". Three forms of this adjusted stream would be to:
1. just keep the dividends in cash form and add back to your forward earnings stream.
2. reinvest those dividends quarterly or yearly in the S&P500 index which in turn would "earn" more earnings and dividends.
Alternative #2 would provide a more real earnings payback.
Finally, one could take those future yearly earnings and dividends and determine the present value by using either the Treasury bill rate or the actual CPI inflation. This last method would provide an apples and apples comparison in todays dollars.
If an investor then compared that payback duration with the payback duration of a 10 year bond using a similar methodology, you would end up with an obective measure of the relative attractiveness of bonds vs. stocks historically. The trick is to then apply your earnings growth forecast and inflation forecast to see whether today's S&P500 valuation is attractive relative to bonds. If it then appears attractive relative to historic payback durations, you probably are using different assumptions than the market. ... and around and around we go.

Sep 03 11:14 AM