GM's earnings for the year have been terrible (not saying anything new here). Yahoo Finance has them as -$53.3. The S&P 500 aggregate P/E is calculated by adding all of the earnings of the companies together and then dividing the current price of the S&P 500 cash index by that aggregate earnings number. This is flawed since the S&P 500 index is priced according to weighting each company by its market cap. However, there are arguments for/against weighting the aggregate earnings by market cap, I'm not arguing here, I'm just trying to present a hypothetical situation and make sure my thinking is correct. The S&P aggregate earnings can be found here: http://www2.standardandpoors.com/spf/xls/index/SP500EPSEST.XLS?GXHC_gx_session_id_=5350992f205e73e4& . So, let's pretend that GM was removed from the index and replaced with a company that was not operating at a loss (just assume $0 earnings/share for the calculations). The S&P 500 actually changes to having a reasonable P/E, which doesn't leave the index looking inflated by much. Current earnings of the S&P 500 using spreadsheet: This is adding together cells D35,38,39, and 40 as they did for the P/Es on the spreadsheet. 7.32 - 23.25 + 9.73 + 12.86 = 6.66 (seriously, 666 again associated with S&P 500!) estimate if GM weren't there: I use Yahoo's -53.3 number here, let me know if there's a better source. 6.66 +53.3 = 59.96 S&P aggregate P/E: 832.39/6.66 = 124.98 without GM: 832.39/59.96 = 13.88 Questions/comments? Am I doing these calculations correctly? It seems that by declaring bankruptcy, GM could actually be giving a boost to the market fundamentals in a big way if I've done it right.

Changed my post, making further edits now, I just took the current quarter from the spreadsheet, thinking it was the year, I still don't see where Bloomberg's number is coming from, though, these #s are straight from Standard and Poors. I'll see if I can come up with Bloomberg's #, it's probably just using operating earnings or something similar.

This is the problem Siegel has pointed out and its correct http://macrospeculations.blogspot.com/2009/02/jeremy-seigel.html Robert Shiller also said there is some distortion going on in the SP earnings

noddyboy, I guess Bloomberg's numbers are based on actual earnings that have come in and maybe some of the spreadsheet's numbers still haven't been updated to reflect the most recent earnings announcements. Plus, the Bloomberg estimate probably uses operating earnings since the number they have is somewhat close to the operating earnings P/E ratios the spreadsheets have. Sorry for the confusion on the calculations. I'm mostly just wondering about the premise of it, though. It seems the ballgame completely changes once GM gets removed.

Yeah, I'd seen his article back in early March and that's why I started looking into it more. Then, I started realizing how much of the skew is caused solely by GM and how it is likely they get removed from the index, and how much of a game changer that is.

susannah the flaw in your calculations are earnings are collapsing at 30%-40% per qtr rate thus the p/e on future earnings is much greater. est's for s@p earnings for 2009 are $40-$50. So even taking the higher est's the p/e is 840/$50 earnings or around 17 times earnings. The mkt looks at future earnings not past.Its like looking at the homebuilders when the had 3 p/e's on current earnings as the cycle was topping at 50 times future earnings. MOST BEAR MKTS OUTSIDE THE 2002 BEAR END WITH P/E'S LESS THAN 10 AND WE'RE NOWERE NEAR THAT

joeyata, yes, they might be collapsing, but my point is that GM is putting a -53.3 into that calculation, so if GM is removed, take whatever # you have for aggregate earnings and add at least 53.3 to it, since I don't think they'll put a company in the S&P to replace GM that is losing $. It makes the calculation a lot different, -53.3 really makes a difference to the numbers we're talking about.