Monday, July 20, 2009

Trailing One Year S&P 500 P/E Ratio: About 2000

No joke here, the S&P 500 one year trailing price to earnings ratio was actually about 1912 on 06/30/2009. Because the losses in the 4th quarter (of 2008) were so huge, the total earnings for the S&P 500 for the past year is extremely low relative to the current S&P price, especially now that this total no longer includes the first part of 2008. When you divide the 06/30/09 S&P price (919)(its even higher now) by the second quarter (2009) earnings you get an abnormally large number for the PE ratio. I first saw a PE of 134 quoted by reddit today from this page of the Standard & Poors website, and I was thinking holy shit, really? So I looked into it and I found this spreadsheet that gives the historical S&P PE from 12/31/1936 until 12/31/2008 and added in the more recent data from this S&P spreadsheet to get my plot:

Yep, that's completely accurate and note that I left out the most recent data point (1912 on 06/30/09) because it would ruin the scaling. Click here for a copy of the appended S&P spreadsheet, all I did was take PE quotes directly from the S&P website. In fact, according to the Standard & Poors, the trailing 1 yr PE is projected to go negative in the quarter ending 09/30/2009! Ok, the headline number (1912) is partially based on estimated earnings for the most recent quarter since not all companies have reported, but frankly, it doesn't matter that much because the PE is so ridiculously large. If the PE is really 1500 or really 2500, does it make all that much difference? Its a massive stock market bubble any way you slice it. Compare this to the historic highs of the dot com bubble (~50) or any other bubble and its way, way, way higher. If you want to work out the details on your own, start here and please let me know what you figure out.

I am appalled that anyone would actually try and justify this rally based on value when its clear that fundamental analysis is completely worthless in this market. I think the markets can go higher but I sure feel sorry for those sorry bastards who buy stocks anywhere near here for the long term. Please note that I still expect a substantial move higher in price from here before this bubble pops. Recall that the Nasdaq broke out last week and notice that the S&P came pretty close today.

This is posted on reddit here.


Anonymous said...

Q2 earnings are still being reported. on 6/30 most earnings weren't out

pythagoruz said...

Right but the money had been earned as of that date, the S&P uses the estimated + reported earnings divided by the 06/30/09 price to get that number. I didn't calculate it myself, although I checked their math. They have the 1912 number on the S&P website, I estimate its closer to 2000 now due to price appreciation of the index.

Anonymous said...

Then why does this group show a recent p/e ratio (still high) but much much lower than what you have?

pythagoruz said...

Hi, they say their PE "uses trailing 12-month diluted earnings per share from continuing operations." So they are using "operating earnings" whereas I am using "as reported earnings." The difference is in the "one time charges," which mostly came from bank write offs. In other words, I am using the real, actual earnings over one year, which in most cases everyone would use, except that no one likes the answer they get. So instead that site uses the "operating earnings" because the number they get makes sense to them. Maybe this is just another example of how investing based on fundamentals is silly because you can make the value look however you want if you try hard enough.

As a side note, I used the S&P's website for my data, they did not cite their source.

Thanks for the question.

david Perkins said...

Bravo to you for finding this data. I have searched all over the S&P website, save press releases, and cold not find 3/31 and 6/30 S&P 500 p/e data. Where did you find this spreadsheet on their site?