Advice?

Well, let me just be argumentative. Forget about the DCF basis for P/E that I spoke of in a previous post, and let's say that P/E is totally arbitrary. Let's also say that it is random -- with an unknown distribution, and we have no knowledge a-priori of how the random value is distributed.

It is very well and sound (but limited) reasoning -- given these assumptions, which include lack of knowledge of what drives P/E -- to observe historical P/E's over time, and a.) suppose that it is more likely for them to take values close to the historical mean, and b.) suppose that they will tend to move closer to the historical mean if they are "far" from it. The basis of the first being Bayesian-motivated reasoning about a distribution of a random variable, and the second being regression to the mean.

So I think the argument about basis and lack of basis for "15" is getting a bit specious, but back to finance 101...

If you believe DCF analysis driving the fundamental stock price, then P/E does in fact take into account _expectations_ of future earnings.

And it's common sense too. What would you pay for a stock that is going to earn $10 / share in 2009, and then there will be no future earnings after that, ever -- guaranteed.

I really think you are trying to use epistemology to discuss issues here. You seem bent on declaring we cannot really know anything. This is true, but it is also in conflict with the very premise of this newsgroup: Planning.

I think you are determined either (a) to refute using anything from the past to help make decisions in the future; or (b) to get me to accept the idea that astrology, numerology and investing is all for recreation and entertainment, like roulette. Otherwise, I do not know where you are going with all this. Maybe you'd do better to state why you do or do not think a P/E of 15 is or is not totally arbitrary.

Tman wrote

I think it only addresses the magic (namely, in the mathematics) of means. I do not think it addresses what is magic about 15. Why not 50? Why not 5?

No, I'm not "bent on declaring that we cannot really know anything." Planning for the future means to get ready for what you expect the future to bring, not what the past has brought.

That would be silly. By all means, use something in the past to help show that P/E or stocks will go up (thereby making the current price "cheep" by comparison,) anything will do, other than their past performance divorced from any other facts. I'm just saying (like every prospectus warns) that the fact that they happened to be higher in the past is not a valid indicator that they will be in the future.

I'm not going anywhere. I'm asking where is the evidence that stocks will go up? Is there something about the companies themselves? Something about the US economy? Something about human nature? What?

I don't know if the P/E of 15 is arbitrary or not, I'm trying to find out if that is, or is not, the case.

Finally, someone who is using something other than past performance.

Tell me just one thing you are planning for the future that does not rely on anything from the past.

Is he? Isn't he using the knowledge that past performance of awful earnings brings rising prices? Still past performance, just a different perspective. Everything humans do relies on past performance. Even when we learn something entirely new, we employ methods used in the past or knowledge acquired in the past and put together in a new way.

I think you're caught up in the "past performance does not guarantee future results" disclaimer in a perpectus. Just because a mutual fund gained an average of 5% a year in the past does not guarantee it will gain an average of 5% a year in the future. It might and it might not. It might lose 3% for a year (or more) before gaining 10%. The mutual fund company does not want you to rely on that past performance. Because the future cannot be predicted, mutual fund companies must post that disclaimer so you can't successfully sue them when they have a down year or two.

Elizabeth Richardson

New word

It's simpler than that. If E falls, then P/E rises even if P is flat. No history involved. Unstated is my belief (with a low level of confidence) that prices have bottomed. -- Doug

Douglas Johnson wrote

I do not see how you can be so confident about P/E changes here. As I know you and most everyone here are aware, earnings are anticipated. The financial media makes much fuss about predictions of earnings. Hence nearly simultaneously and in a virtual continuum, stock prices respond even to these mere earnings predictions. An actual decline that happens to correspond with an anticipated decline is for the most part already going to be built into a stock's price. Hence P/Es should already have declined for the fourth quarter at this point. Once actual Q4 earnings are announced, stock prices will respond to the difference between what was anticipated and what actually happened. These differences can be positive or negative, with the result that what P/Es will do is rather unpredictable.

But as I understand it, P/E isn't based on *anticipated* earnings, but past earnings. Isn't that correct? If the P has already been adjusted for anticipated earnings, then it won't change when the earnings are declared (assuming the predictions were correct.) When the E goes down the P/E will go up.

The most publicized P/E numbers are the trailing twelve months ones that use the last 12 months of claimed earnings per share, so yes.

When actual Q4 earnings are announced, what the buyers and sellers consider is the difference between actual and what was expected. If actual earnings exceed expected earnings, then the price shortly after the time of announcement will tend to rise. If actual earnings are less than expected, then the price will tend to drop. Since we do not know which way the difference is going to go, we do not know the way the price will go. Thus predicting P/E's movement based only on estimates of what actual earnings are likely to be (as Douglas proposes) is a crap shoot. Fact is it has been long anticipated that Q4 earnings by and large are going to be down. But the real question is whether the actual Q4 earnings will be as bad as anticipated.

This is pretty well known stuff (information already built into price type stuff) and I feel a little arrogant/pedantic/your choice repeating it when I know so many here know this.

Surely you can tell the difference between someone who says, "X will go up because it has been higher in the past." and someone who says "X will go up because it is inversely tied to Y and Y is going down."

Notice also that he is *not* saying the profits are going to be bad this quarter because they were good in the past.

Right, because the mutual fund's past performance has little to do with what its future value will be, other factors predominate.

So, I ask again, what is there *other than its own past performance* that you use to determine when, or if, stocks will go up? What are the other factors that you use to determine what the future value of stocks will be?

Because, as I've said in several prior posts, people perceive stocks to be on sale. They will buy. That will cause the price to go up. They perceive them to be on sale only because, in many respects, the price used to be higher. As I've also said, I don't know when this price rise will occur. I don't know that it will be in 2009, though others here have indicated we've already hit a bottom. I just know that it *will* occur. Market prices of broad indexes is not based on black and white facts, but, rather, intuition.

Elizabeth Richardson

Which is exactly why I expect P/E to go up. Lousy fourth quarter earnings are fully accounted for on the P side, but not on the E side, assuming we are talking about twelve month trailing earnings. As long as Q4 2008 earnings are less than Q4 2007 earnings and are close to what the market anticipates, P will stay roughly the same, E will decrease, and P/E will increase. Right?

The only way P/E will decline is if we get a big sell-off. Always possible. That's why I use weasel words like "almost guarantee". -- Doug

No I would not say they are fully accounted for. Someone has punched some numbers, estimated earnings, and then reported the same to the media. But these are only estimates. It is very common for these estimates to be shown to be off, sometimes by a lot, when actual earnings are reported.

Well you are now qualifying your statement to include consideration of the difference between anticipated and actual earnings. This distinction is important, as I explained in my other posts.

There tends to be a sell-off if earnings are worse than anticipated.

I encourage careful qualification. But in this case, I think you initially failed to note that the difference between anticipated and actual earnings is unknown, and hence we cannot say whether prices will tend to rise or fall. Thus even knowing that earnings will decline and so act to drive P/Es up, it is also quite possible earnings will be worse than anticipated, driving stock prices down. If prices are driven down, then P/E does not necessarily rise.

I think that the point is just calling things by what they are: it's not planning as much as guessing and, in the case of investing, betting that things will go one way.

Happy 2009!

the worst is yet to come?

intervals to ensure dollar cost averaging be a wise move?

Many seem to equate indices with investing in stocks. Indices include stocks I wouldn't touch. There are some very good quality stocks with solid earnings histories and solid earnings prospects trading below 6 times earnings. AT&T is one.

True, stocks correlate with the market, but not all of them, not perfectly. Stocks do correlate with earnings. I wonder if the mass- marketing on the part of funds whose disciplines are "top-down" have influenced any of the many posters here? It's relevant to talk about the broad market and its PE and yield and book value, and earnings, but the broad market is made up of individual companies. That's (I guess) a "bottom-up" approach, or fundamental analysis.

"Numerology" is important, too, but what makes a business successful is a primary consideration. The product line is what drives the sales and earnings. Like the Buffet story before he invested in American Express - he stood at a restaurant and observed that people paying with credit cards were exceptionally proud. So, I guess he figured it would be a popular product.

At year end the hypothetical portfolio I set up is +12.72% v. a Vanguard broad based index +7.6%. The website "Stockalicious" was temporarily hijacked, but is back online. My collection is just stocks I thought had gotten badly beaten up even though they are solidly profitable companies. My disappointment, really, is CHRW - I kind of cheated a bit on that one, since it didn't get badly beaten up. Its earnings are too good, and are somewhat insulated (the company does logistics for trucking). It's the percentage leader in the collection.

On Jan 1, 7:26 pm, dapperdobbs wrote: There are some very good quality stocks with

Sorry - wrong - I blew it. T is above 12 times earnings. IR (Ingersoll Rand) is just over 5 times earnings.