How Cheap Are Stocks ?

One of our intermediate concerns about equities involves valuations. While we recently made a short term buy call, that is merely a trade that could possibly run for a couple of weeks or months — but not much longer. That call certainly wasn’t made because stocks are such screaming bargains.

As to valuations: Its hard to really say that stocks are cheap here. At best, I believe we can argue that — assuming that historically high earnings do not fade — that stocks are not terribly expensive. But that is very different than saying they are cheap.

Have a look at this lovely table from Dow Jones Market Data Center:

Stocks Are Not Cheap


You can also see the Yields On Dow Stocks here (Thanks, Tim!)


These are not the sorts of valuations you find at the end of Bear markets.  And, James Montier points out a factoid that makes the above even worse: Analysts lag reality. James adds the damning observation that "They only change their minds when there is irrefutable proof they were wrong, and then only change their minds very slowly."

Have a look at his chart below: It is a linear time trend out of operating earnings and the analyst forecasts of those earnings (so the chart simply plots deviations from trend in $ per share terms).



As the red line in the chart shows, the earnings recession is just now beginning. But as the black line reveals, analysts have yet to lower their earnings numbers. Montier notes that the downgrading of estimates has been highly constrained to the financials (and to a lesser extent the consumer sector in the US for 2007).

Roughly speaking US earnings ex financials have been revised down by 1.5% compared to nearer 4% for the market as a whole.

Thus, our contention that markets have only priced in a short, shallow recession . . .


Asleep at the wheel, or, How I learned to stop worrying and love the bomb
James Montier
Apr 07 2008, 02:52 PM

Schwab Asks Who Needs Analysts After Biggest Flub    
Michael Tsang and Eric Martin 
Bloomberg, April 7 2008

Print Friendly, PDF & Email

What's been said:

Discussions found on the web:
  1. bonghiteric commented on Apr 9

    Cycles have troughs. Don’t need a graph to see that all the external signals, (i.e., consumer sentiment, decreased spending, rising unemployment, higher input costs) cannot help but curtail earnings for several quarters at least. I believe stocks aren’t cheap enough because I don’t see earnings power in the future. Good article on a subject that has been perplexing me recently.

  2. BG commented on Apr 9

    Thanks for the charts. Good stuff from you as always.

    One off topic comment. It’s a little bit spooky how all (3) of the past and present Fed heads are out there talking publicly. That tells me this thing is nowhere near being over. Ignoring the man behind the curtain no longer works.

    I think all (3) are scared shitless in regard to what the economic/financial future holds. The evil genie has escaped its bottle and there is no getting it back inside.

    In regard to this post, I’m surprised how apparently the bottom has dropped out of the earnings for the Dow Industrials relative to other indices. That can’t all be financials.

  3. Steve Barry commented on Apr 9

    I would argue that a trailing P/E of 21 for the S&P, at unheard of high profit margins (which must mean revert), staring at a major recession and housing meltdown is outrageously expensive. Dividend yield is 2% where historically, before Mr. Bubbles, 3% was a major market top and 6% was a bottom.

  4. Eric Davis commented on Apr 9

    Always possible the turn could be just around the corner….

    I’m suspecting we are going to need “Bear Capitulation” before we can turn. Followed by the notion that “sideline money” isn’t going to come in.

    Dougie said “time to get short the other day.” he is usually 1-2 weeks early.

    Or Monday.

  5. yourkillingmelarry commented on Apr 9

    Curious, if the downgrading of estimates has been highly constrained to the financials, what sectors are the most vulnerable here using the assumption that the recession will be more pronounced than is being forecasted?

  6. wally commented on Apr 9

    The chart says it: forecasts are really backcasts. They are as useful as yesterday’s weather report.

  7. Todd commented on Apr 9

    How do the p/e ratios and dividend yields of our major indexes compare to similar indexes in Europe? Or in Brazil?

    I’d like to see that comparison in empirical data.

  8. Wayne Mulligan commented on Apr 9

    Haha, that’s a great chart…essentially proves why most buy side analysts aren’t worth the six-seven figure bonuses they make and should be more in the 2 figure territory. On aggregate they’re essentially doing the equivalent of a reporter rewriting last month’s news articles.


  9. N commented on Apr 9

    people may want to consider stock buybacks by companies in addition to dividend yield

  10. VennData commented on Apr 9

    BG’s point is interesting about all the Central bankers out there in the media. But instead of talking about solutions like others, Greenspan’s out that deflecting blame.

  11. Alfredo commented on Apr 9

    The earnings expectations on the Russell 2000 are downright bizarre heading into a recession. These PE’s suggest that earnings are forecasted to grow by 159% in the forward 12 months compared to the trailing 12 months. What a pipe dream.

    a) Trailing P/E: 53.50
    b) Forward P/E: 20.63

    c) Russell 2000: 711.92

    Trailing Earnings: $13.31 (c)/(a)
    Forward Earnings: $34.51 (c)/(b)

    Forward O/(U) % to Trailing: 159.3%

  12. Vermont Trader commented on Apr 9

    Bloomberg) — Trading on the New York Stock Exchange dropped to the lowest level this year, a sign that changing profit forecasts are making investors reluctant to buy or sell stocks before companies report first-quarter results.

    About 1.2 billion shares traded on the NYSE yesterday, the least since Dec. 31. Over the past four days, an average 1.24 billion shares changed hands, 27 percent lower than the 1.69 billion daily average this year. Volume on the NYSE this year had jumped to 1.72 billion shares a day through April 2 and was on pace to climb to a record, according to Bloomberg data.

    Watching the market, there are plenty of shares for sale at resistance.

    If shares were cheap here then they wouldn’t be here.

  13. Philippe commented on Apr 9

    Well substantiated reasons why to buy equities markets by minyanville all centered around the theme

    « Do not tell my mother that I work for the financial industry as she thinks I am a pianist in a brothel »

    They do not say if their mothers love them

  14. Ross commented on Apr 9

    Don’t get me started about stock buy backs.

    I’m eagerly awaiting the study about how much capital was squandered over the last decade on buybacks. Corporations followed their Wall Street masters and bought high only to sell low! Ridiculous.

    On cash dividends I have only one thing to say. GIVE ME THE MONEY.

  15. craig commented on Apr 9

    excellent discussion on long-term EPS growth, PE multiples, corporate profit margins, how to derive long-term probable market returns, etc. at, in the weekly market commentary section, go find the Feb 22, 2005 commentary. the link is:

    very good analysis, once read you will either have a good framework for the rest of your investing life for thinking about valuation, PE multiples, earnings growth cycles, etc, or you won’t agree with him. simple.

  16. craig commented on Apr 9

    disclaimer: i don’t own any hussman funds, i never have invested in them, i don’t know him, i have no pecuniary interest in the funds, etc, yada, yada. i just think it’s good analysis and he post a free weekly market commentary that i enjoy reading. nothing more.

  17. craig commented on Apr 9


    I was looking at the russell 2000 also and i’m wondering are you looking at the EPS including the neg EPS or excluding it for 07 and 08? because a lot of russell 2000 companies with net losses. and where are you getting the data? i’m having trouble finding updated data . thanks.

  18. Alfred commented on Apr 9

    I do not understand how most of you come up with this idea that we are going into a deep recession. We might not even have a negative first quarter. The OECD is just out with an update for Germany’s growth forecast in 08/09. It does not seem that Europeans strongest economy is falling into recession et all. Stocks have priced in already somewhat recessionary levels, judging by the chart they look cheap. I caution everybody to be too negative on the economy and fall into the trap of the Greenspan – Bernakenistas. What if the monumental reinfaltion has nothing to do with the real economy but everything with the ailing investor class on Wall Street? It would not be the first time that the Fed is wrong – see Greenspan interview on CNBC from yesterday.

  19. Pat G. commented on Apr 9

    Boeing is up 4% on an announcement that they will delay Dreamliner deliveries to 2009. I guess the market has already priced that in too.

  20. craig commented on Apr 9

    what is a “deep” recession, please define. you posit that stocks have priced in a somewhat recessionary level. please support that with data about forward earnings, the PE you put on the earnings, etc. i enjoy reading all viewpoints, however, viewpoints supported by a numbers driven analysis are usually the most compelling. your analysis was all qualitative about “looks like” the market is priced it in “somewhat”. i’m not flaming you, just want to see some data to flesh out your viewpoint.

  21. Alfred commented on Apr 9

    My comment about valuation is based on Barry’s excellent chart. Trailing earnings for DJI is 53.6, forward earnings are just 13.63, that’s down 70%. That looks cheap to me.

    I want to be cautious because I don’t have a crystal ball to look into the future. The evidence in form of data is just not there to support the outlook for a severe (prolonged) recession. I would recommend to read Anatole Kaletsky’s piece in the Times (source: Recession unlikely if US economy gets through next two crucial months
    Anatole Kaletsky: Economic view, TimesOnline

  22. Alfredo commented on Apr 9


    All of the data I used is from the WSJ table that is above (in the original blog posting). The only additional data is the Russell2000 price as of yesterday’s close. By dividing the Price by the Forward and Trailing PE’s, you derive the earnings number.

    The 70% decline in the P/E that you are referencing does not mean that stock prices are cheap. The key is how realistic are the earnings expectations and the main point of the original blog posting is that analysts always seem to miss turning points in earning cycles. The PE numbers for DJIA imply that DJIA earnings will increase by 293% in the next 12 months compared to the prior 12 months. Does that seem like a realistic expectation to you.

    IMHO, the market is waiting for earnings reports and guidance to come in before heading lower…..much lower.

  23. AH commented on Apr 9

    Looking at a very long term chart (approximately 80 years) of S&P and its PE will reveal a few very interesting data points.

    The market, over the long run, tends to pay an average 16 times the earnings. The key is ‘over the long run’. On top of that just because it has been doing so for the last 80 years (or may be longer, but I don’t have those data) does not necessarily mean it will always keep on doing so in the future, too.

    A closer observation will reveal some more interesting phases:
    (1)The market can deviate from the average for a significant period of time, both on the upside and the downside.
    (2)The market can also keep ignoring the changes in the earnings and decide to do its ‘own thing’ for a while.

    During early 1930s the market essentially followed a huge decline in the earnings for approximately 3 years; then followed the upswing in the earnings until late 30s. But during WWII and late 40s through early 50s it chose to ignore the upswing in the earnings. Essentially, during each of these deviations, the market PE declined from the top of the range (approximately 19) to the bottom of the range (approximately 8).

    Then suddenly, the market decided to go to a PE expansion cycle for more than a decade, and by early 60s its PE was hovering again at the top of the range. During this time the market grew at a much faster rate than the earnings. Then for another decade the market PE mostly remained in the upper half of the range, occasionally falling below 14 but never going beyond the peak (roughly 23) it had visited during early 60s.

    After that, market chose to do another price discovery in the downside, and in the process, pushed the PE down below 8 during 1973-74 and again during late 70s and early 80s. Interestingly enough, during the period of early 70s through early 80s the market’s earnings kept on growing (at least in nominal terms, I am not aware of the inflation adjusted earnings), and this time the nominal earnings grew at a rate faster than the market price.

    The bull market during the 80s and 90s pushed the PE of the market way beyond any historical norms (of last 80 years, at least) up in the mid-40 range. And now, apparently we may as well be in PE contraction phase, where the market may look for a new price discovery in its own way.

    So what does it tell us? I believe it warns us against giving a blanket ‘cheap’ sticker to the market if we consider that the notion of cheapness or dearness should be anchored into certain context. And just because the market is hovering in a range where it chose to during certain period does not mean it can not choose to reprice the fundamentals, be it rational or irrational.

    Finally, if this market chooses to pay a PE of below 10 (as it did in several occasions before) the current levels would definitely not look cheap when we look back. But again, the market may not do that, and instead wait until the earnings improve significantly over a long period, and then decide to reprice it on the upside.

    Who knows!

  24. Ptrck commented on Apr 9

    Since the price of the Dow a year ago was just about what it is now, your numbers imply that earnings for the Dow over the last year have fallen by 1 – (16.99 / 53.61) = 68%. That can’t be right.

  25. Alfred commented on Apr 9

    I think you are making a falls assumption in your model of calculating a likely/or not likely earnings increase in the next 12 months. If I understand your calculations correctly you are basing your forward DJIA earnings (293%) on the historical 12 months trailing data. That might work in a perfect future, but when you have likely deviation from the historical data series your model gives you the wrong results. The 70% reduction in the forward P/E compared to the trailing clearly expects deviation from the historical data series.

    I also think that what you are referring to is not the P/E but rather PEG (price arnings growth).

  26. SPECTRE of Deflation commented on Apr 9

    Goldman sells $500 mln of Chrysler debt at very deep discount (.63 Cents On Dollar)

    By Cynthia Koons and Serena Ng
    NEW YORK (Dow Jones)–Goldman Sachs (GS:Goldman Sachs Group, Inc
    4:00pm 04/09/2008

    Goldman Sachs placed $500 million of Chrysler Automotive’s loans at a price of 63 cents Wednesday to an investor group that included hedge funds, a person familiar with the matter said.


  27. n commented on Apr 10

    MONDAY, SEPTEMBER 10, 2007

    Buyback vs. Payout
    Will stock repurchases doom disbursements?

    COULD DIVIDENDS ONE DAY GO the way of the dinosaur?

    Douglas J. Skinner, a professor of accounting at the University of Chicago Graduate School of Business, thinks it’s possible. In a recent study entitled “The Evolving Relation between Earnings, Dividends and Stock Repurchases,” Skinner surveys the determinants of corporate payout policy and compares the merits of stock buybacks with those of cash dividends.

    Owing to a change in regulation, stock repurchases emerged as a viable form of shareholder reward around 1983. Aggregate net repurchases were consistently in the $30 billion range from 1985 to 1990, before declining briefly in the early 1990s. In 1998, however, the dollar amount of stock buybacks surpassed that of dividends for the first time — and did so again in 1999, 2000, 2004 and 2005.

    Skinner finds that stock buybacks are now being used as substitutes for dividends, even by companies that continue to make payouts. The reason for that appears to be the increased volatility of earnings.

    In some ways, says Skinner, repurchases are the superior choice for the issuer because “there’s total flexibility with stock repurchases, whereas you’re committed forever with a dividend.” For obvious reasons, companies want to avoid cutting or omitting their payouts. “With repurchases, you could pay out $3 billion…one period and zero the next.”

  28. n commented on Apr 10

    See dividend yield on Berkshire Hathaway Class A

    How about total return? (vs. dividend yield)

    Sometimes a stock with no cash dividend may be okay.

  29. TKL commented on Apr 10

    The trailing P/E of the DJIA is surely not 53.61. If valuations had tripled, we’d be at . . . hey, Dow 36,000 after all!

    Dow Jones itself shows the trailing P/E, including negative earnings, to be 14.64 as of 03/31/08 (see

    Even the financials-heavy S&P, where earnings have been decimated, is “only” a little over 20 times trailing.

  30. Turley Muller commented on Apr 10

    Regarding the 53 trailing P/E on the dow- It’s way skewed bc of GM- it had an one-time expense that resulted in EPS (ttm) being -$68. Excluded this one-time item, for this ONE dow component, trailing earnings would be around 15-16. Therefore earnings haven’t declined 70%.

    My thought on trailing multiples- most always they are useless. Share prices reflect future expectations, not past information.

    Trailing multiples capture historical performance which can vary significantly from future performance. Only thing that matters is future cash flows, and stocks are valued based on expectations, therefore, forward multiples are best indicator. However, those multiples are based on estimates, which are rarely accurate and subject to change.

    14x multiple is rather cheap (SP500), in itself, but considering that growth expectations and EPS estimates may be too high, then in actuality, the multiple is not too high.

Posted Under