Had to read it twice -- too many Grosses.

Good point! Some other comments:
1) "The fact that Buffett, who has oodles of cash to put to work, hasn't found many--and has instead been nibbling on distressed properties--shows just how overvalued stocks still are..." This of course assumes that Buffett is reading the market correctly (ok I think he is, but just to point out the assumption).
2) The Dow at 5000 is not necessarily a reason to buy bonds - maybe the bonds known as Treasuries, but not *all* or even *most* bonds - since the rate of default on anything but Treasuries (e.g. corporates, municipals, etc.) is likely to increase should the market go that low. One also has to wonder what will happen to MBSs, where (I believe) Pimco has a large stake. Surely with the Dow at 5000 the U.S. housing market will tank significantly, and we will see what the implicit government guarantee of the Fannies is worth?

What I am most puzzled about is the current price/earning ratio for the S&P of about 21, even after a fierce 30 month bear market. In looking through data for the past 30 years, I can find no bear market that was remotely this expensive. P/E ratios are high if losses are excluded from the data, with losses the ratios are above 30. Price to book ratios are just as high.

What is happening? Are the ratios statistical oddities at present? Will the market soon begin to look quite cheap? Are stocks going to trade at elevated p/e ratios from now on? Well, why? What sort of market growth can be expected given current ratios?

Has earnings growth for S&P companies really been as low as Bill Gross argues? Why does a John Bogle argue that 7% is a proper figure for long term earnings growth?

Warren Buffett has argued in the NYTimes, that options expenses and excess pension fund investment gain provisions have far more seriously distorted earnings than the open scandals. Can we estimate a cost to such distortions?

>>Has earnings growth for S&P companies really been as low <<
Try this paper by Arnott & Bernstein:

http://papers.ssrn.com/abstract=296854

>>Can we estimate a cost to such distortions?<<
S&P is rolling out a "core earnings" calculation that computes earnings after subtracting options costs, ignoring pension income and perhaps one or two other adjustments. I believe core earnings are about 20% lower than GAAP earnings.

Question for Brad or other economists:

When looking at market p/e ratio, which is better, "including negative" or "excluding negatives." Given that shareholders are not liable for corporate liabilities, I'd think "excluding negative" earnings is a better approach.

I asked the same question about including or excluding negative earnings as Richard did on USENET and got very little response.

It's important to note that the difference in the w/ w/o negative earnings figures are very high these days. Thus, the question is timely.

I don't agree with Richard's rationale that limited liability is a reason for not using the including negatives figure, since I assume for many of these cases, the company won't go bankrupt, and there will be quarters with positive earnings.

On another topic, I don't know the details of Bogle's argument for 7% earnings growth, but I assume it's history. And history seems like a pretty good guide these days. (An EXTREMELY interesting article related to this can be found at http://www.efficientfrontier.com/ef/702/2percent.htm

sjfromm

I have traced monthly p/e ratios from Janaury 1975 to August 2002. The levels of p/e ratios both with and without losses are astonishingly high given how much prices have fallen. The differences between p/e ratios with and without losses are larger far than at any other time. Month after month there are massive losses being recorded by S&P companies.

Remember, we have not been seeing options expenses nor realistic estimates of returns on pension funds.

What is Intel or Cisco or Oracle worth when options are expensed? How do we invest intelligently when valuation is such a mystery?

Though we may use index funds, it is necessary to have corporations provide transparent accounts if we are to know if we are receiving a fair share of earnings.

John Bogle generally uses data from Jeremy Siegel - "Stocks for the Long Run." Bogle too has long complained with Buffett about corporate earnings mysteries.

Jeremy Siegel suggests that a p/e in the low 20's is realistic given structural improvements in the markets. Brad DeLong has commented on the tax bias against larger dividends. However, stock buybacks as a substitute for dividends have long been rather a smoke screen according to John Bogle. We are more dependent than ever on our stock and bond investments for retirement, and much in need of transparent accounting. We can not rely merely on home value appreciation for retirement.

What earnings growth can we expect going forward? With a 21 p/e, even though we do not kn ow options expense, we might expect a market return of 2 percent dividend and 7 percent on earings or 9 percent if the p/e stays about 21. A 9 percent return for an index investment would not be at all bad.

Should a p/e of 21 be about fair value, then we can expect about a 9% return on the S&P over the coming decade if earnings growth averages 7% since dividends are currently 2%.

This will make the S&P a much better investment than a 10 year treasury note at 4%. The tax treatment of an S&P or total market index and the low expense will leave an investor far better off than in bonds unless there is a long term decline in the p/e or earnings growth is far less than we have seen for 30 years.

Wasn't the PE of the S&P at 21 and the PE of the DOW at 17 in October, 1929? Wouldn't that mean that stocks currently are ABOVE the level that they fell FROM in 1929? I think people don't understand just how high the stock market was in this recent bubble.

Bear markets historically bottom at PEs of 7-8, and bear cycles bottom at PEs of 13-14 or so. Why do people think that this time is different? I think it is wishful thinking. You have to either raise the earnings or lower the price to adjust the ratio. Unless you can tell me where the huge growth in corporate earnings is going to come from to reduce these PE ratios, I think the market's price is going to need to go down a lot more.

Housing prices dropping is said to be the trigger for deflation and I see signs that housing prices might be about to drop. I'm in Silicon Valley and a lot of people here have run through their savings and their unemployment benefits and are starting to sell their houses. I read that foreclosures nationally are at a 30 year high.

I wish the Bush Administration would DO SOMETHING to get job growth, like real jobs programs etc. and not rely on ideological tax cuts for the rich (mostly kicking in way in the future) and hopes for seizing oil fields as a way of getting us out of this.

Two points; there is nothing intrinsically contradictory in being bearish on both stocks and bonds, and although stockholders are not liable for liabilities, it is nevertheless not appropriate to ignore negative earnings in most cases, as they either reduce retained earnings which might otherwise be distributable or necessitate dilutive equity issues. The only case in which the without-negative number might be relevant was if you believed that the loss would eventually be borne by creditors, and I don't expect that anyone seriously believes that a material proportion of the loss-making SP500 companies are actually insolvent.

It is interesting to see Gross (Daniel) write an entire column about Gross(Bill) without offering a single coherent argument to counter Gross (Bill)

The only (semi) argument that I could find was a reference to Business Week!

"It is...not appropriate to ignore negative earnings in most cases, as they either reduce retained earnings which might otherwise be distributable or necessitate dilutive equity issues."

If Mr. Davies is correct, then the S&P p/e ratio on Aug 31 was 31.94. This would not seem to allow much room for market growth for a while unless we assume there really is no limit to valuations relative to bonds.

NB of course that, as is always the trade-off for simon-purity of analytical constructs, the with-losses PE is a much more volatile number than the ex-losses, and could come down very fast for no other reason than that losses are usually driven by one-offs. I'd never put my faith in a single number.

Still, the looses being recorded are very high by historical standards. The price to book ratio is very high, the dividend yield is very low, the p/e ratio excluding losses is very high.

Just how do we make sense of all this. Will earnings suddenly rocket, leaving the maket looking cheap. Do low interest rates mean we can simply ignore high valuations?

Thanks for introducing me to me, but I think your charges of (1) character assassination; and (2) flip-flopping are more than a bit off-base. In neither article did I make any comments about the personal characters of either Buffett or Gross. And nowhere in my article on Gross did I suggest that stocks are somehow undervalued. I stick by my contentions that, (1) given the consensus, Dow 5000 is a somewhat outlandish call, and that (2) bond fund fund managers would like to see people put more money into bonds.

Thanks for introducing me to me, but I think your charges of (1) character assassination; and (2) flip-flopping are more than a bit off-base. In neither article did I make any comments about the personal characters of either Buffett or Gross. And nowhere in my article on Gross did I suggest that stocks are somehow undervalued. I stick by my contentions that, (1) given the consensus, Dow 5000 is a somewhat outlandish call, and that (2) bond fund fund managers would like to see people put more money into bonds.

Thanks for introducing me to me, but I think your charges of (1) character assassination; and (2) flip-flopping are more than a bit off-base. In neither article did I make any comments about the personal characters of either Buffett or Gross. And nowhere in my article on Gross did I suggest that stocks are somehow undervalued. I stick by my contentions that, (1) given the consensus, Dow 5000 is a somewhat outlandish call, and that (2) bond fund fund managers would like to see people put more money into bonds.