Siegel vs. Standard & Poor's 30 comments
-
Font Size:
-
Print
- TweetThis
WSJ Feb 25, 2009: “The S&P Gets Its Earnings Wrong, Stocks are cheaper than they look”, by Jeremy J. Siegel
Economist Jeremy Siegel (associated with WisdomTree funds, and professor at Wharton) is an important figure. When the Wall Street Journal published his editorial (read full article) claiming that Standard & Poor’s had a flawed method that massively overstates the S&P 500 P/E, that became something that can’t be ignored. In fact, Siegel said if S&P had calculated index earnings correctly, investors would see that stocks are historically undervalued.
The blogosphere came alive with comments that spanned the gamut from agreement to disagreement in both polite and impolite tones.
Several of our clients called to inquire, since we have been telling them that stocks are expensive and that the index needs to fall significantly to come in line with current realities, and to be in the lower range of historic P/E valuation.
Rather than joining the throng of “top of mind” commenters, we decided to ask S&P how they do their work.
Of course, they disagree with Siegel. They referred us to their publication on the mathematics of their indexes (download it here).
Our Conclusion:
- Siegel may or may not be correct in asserting that there is a better way — we have not attempted to analyze that.
- The S&P method does not appear to be inconsistent or illogical, as Siegel claims — not surprising after decades of publication and significant competition from MSCI, Dow Jones, FTSE and others who had incentive to point to any logical flaws in the S&P method.
- The S&P method, flawed or not, has been consistently applied for a long time, making historical comparison of current P/E ratios with prior P/E ratios useful.
- Siegel has not provided historical P/E ratios calculated on his basis, rendering his method useless in terms of historical analysis until he restates and publishes historical results using his method.
- Siegel created his own logical flaw by attempting to prove his point by comparing the current P/E, based on his alternate index earnings calculation, to historical P/E ratios based on the S&P method, which he claims is wrongly calculated.
- Our 81-year analysis of monthly P/E ratios, based on the S&P method (see our Nov 3, 2008 article), finds the historical range of P/Es to be 10 to 20, with 15 as a median — and based on that, the S&P projections of index P/E’s in the vicinity of 20 and up makes the current S&P 500 index expensive.
- Expensive markets in bad economies — and this is a bad global economy — are more likely to fall than to rise.
S&P Earnings and P/E Calculations:
This table is adapted from a spreadsheet available on the S&P website (download it here). (Click images to enlarge.)
S&P Index Calculation Mathematics:
The images that follow are a relevant collage of extracts from the S&P methodology document which you can download above and read in full.
The Siegel Opinion Letter Excerpts:
Standard & Poor’s recently shocked investors with an announcement that reported earnings for its S&P 500 Index for the fourth quarter of 2008 are forecast to be negative for the first time since such data were calculated in 1936. …
What this dismal news actually reflects is the bizarre way in which S&P (and most other index providers) calculate “aggregate” earnings and P/E ratios for their indexes. …
If one applies market weights to each firm’s earnings using the same procedure that S&P employs to compute returns, the results yield a more accurate view of the current profit picture. Market weights produce a reported earnings estimate of $71.10 for 2008 — nearly 80% higher than the unweighted procedure. …
No one can deny that the recent economic downturn has badly hurt corporate earnings. But let’s not fool ourselves into thinking that this is an expensive market. When computed accurately, P/E ratios show that this market is much cheaper than is currently being reported by the S&P.
Fundamentals and Chart Pattern Perspectives:
Fundamentals suggest applying a 10 to 15 multiple (the bottom 1/2 of the main range) in this negative environment — better closer to 10 than to 15. Combining that with an S&P 2010 index earnings projection of about $40, makes a 400 to 600 price range for the index more plausible than an 800 to 1200. Either one could happen, but we think odds favor the lower more than the higher range.
Visible chart patterns can be the arbiter of the truth. The prices are clearly and visibly going down, as the charts below show. Until the charts show signs of pointing up, the lower 400 to 600 range seems more likely than the upper range of 800 to 1200.
20-Year Monthly S&P 500 Prices
6-Months Daily S&P 500 Prices
Summary
We are sensitive to the anxiety many investors feel sitting in money market funds with some or all of their stock allocation, earning less than 1% (in some cases a lot less than 1%). That’s hard to do. However, a 1% gain looks really good to us in comparison to a probable significant capital loss.
Some investors may believe that they can ignore the possible lows between here and new highs. Some investors are early in the investment accumulation stage of life. For them, perhaps risking the stock allocation now and waiting for gains may be acceptable.
Other investors cannot ignore the in-between period, because they are done accumulating assets, possibly living off of their assets, or are not able to replace their capital with new earnings. For them, there is no reason or need or capacity to suffer losses though the in-between period (which could be long, although it may be short). For them, we don’t believe risking now is appropriate.
The S&P 500 is fundamentally expensive, the Siegel opinion notwithstanding. The S&P 500 price is trending down, not up. We completely exited the US stock market in July of 2008 and will not re-enter until it makes visible price chart sense — not before — not now.
We don’t buy what is falling. We buy what is rising.
We will gladly change our view when, and only when, the charts show credible evidence of a reversal of the obvious current downward trend. We don’t intend to risk capital on what “ought to”, “should”, “might”, “could” or “someday” will happen to prices, or on what spokespersons for asset gathering / asset fee dependent enterprises say we should do. We will risk capital on what “is” happening to prices, and are more than willing to give up the earliest upward movements to be assured that prices are in fact “trending” up, not just bouncing around in a sucker rally.
Our cash and our clients' cash is simply too precious to speculate on a situation where both fundamental logic and price charts are flashing warnings. Cash needs to be conserved and held aside until observable price action reality, preferably confirmed by fundamental logic, indicates a market trend reversal to the upside.
Related Articles
|
-
The Seeking Alpha 100
See allThe top 100 stock
market authors
selected for publication
Seeking Alpha 100 » -
Fastest Climbers
Authors with the largestSee all
increase in followers
in the last 7 days
Fastest Climbers » -
Top Commenters
Top commenters,See all
as ranked by their peers
Top Commenters » -
Top Instabloggers
The 100 most active
Instabloggers.-
1
David Fry
-
2
TraderMark
-
3
Don Dion
-
4
Cliff Wachtel
-
5
Mike Havrilla
Top Instabloggers » -
1
-
Top StockTalkers
TopSee all
StockTalkers.
Top StockTalkers »
This article has 30 comments:
Does this make the market cheap? With the Shiller calculation, secular lows around the 1921, 1932, 1949, and 1982 bottoms have been in the single digit range. Here's a link with charts to illustrate: dshort.com/articles/20...
"Suppose on a given day the only price changes in the S&P 500 are that the largest stock, Exxon-Mobil, rose 10% in price and the smallest stock, Jones Apparel Group, fell 10%. Would S&P report that the S&P 500 was unchanged that day? Of course not. Exxon-Mobil has a market weight of over 5% in the S&P 500, while the weight of Jones Apparel is less than .04%, so that the return on Exxon-Mobil is weighted 1,381 times the return on Jones Apparel."
"Yet when S&P calculates earnings, these market weights are ignored. If, for example, Exxon-Mobil earned $10 billion while Jones Apparel lost $10 billion, S&P would simply add these earnings together to compute the aggregate earnings of its index, ignoring the vast discrepancy in the relative weights on these firms. "
This is as it should be. Siegel goes way off thinking that earnings should be weighted like price. In his example, Jones Apparel with that enormous loss vs market cap would have an equally enormous negative EPS, which then would require weighing. If done properly, you'd come up with the same number had you simply done what Siegel accuses S&P of 'improperly' doing.
Or, you could choose to eliminate price weighing on the index. Exxon gains $30 billion in market cap, great, the index goes up by 30 billion points (around 10% move upward). Jones loses $20 million in market cap, great, the index goes down by 20 million points (around 10% move downward).
Think of it this way - If GE's financial division reported a $30 billion loss, while its manufacturing reported a $15 billion gain, GE as a whole would report a loss. Even if you weighed the loss of the financial division vs its size compared to GE, you'd still get a negative P/E for the firm. There's no way to avoid this simple fact.
Siegel's point is utterly ridiculous. Middle school math students could do better.
I somewhat agree with dshort, but what role does interest rates play in the P/E comparisons? In '82 treasury yields were sky high; now they're scraping the bottom. Therefore we can and should have higher P/E ratios at the final bottom of the market compared with '82. I will admit I don't have the talent to come up with this nomalized P/E for '09 and I don't think anyone else does either.
I expect we will see another shakeout. Could be nasty. Could be this week. My downside target is SPX 600. My shopping list is ready.
> 1. Siegel may or may not be correct in asserting that there is a better way — we have not attempted to analyze that
Well Sir, please DO attempt to analyze it. Siegel's argument is self-evidently right.
But just to play Devil's Advocate, I noticed on the S&P spreadsheet that operating earnings in 1997 were around $40 and the Index was were it is now. While tech stocks and their high multiples skewed the P/E to the upside at that time, I wonder if the financial sector isn't doing the same on the downside. When I see a companies like Proctor & Gamble, J&J, and so forth selling at 12x next year's earnings, it makes me wonder if this isn't a good time to be a selective buyer.
Debate is useful and welcome. Declaring an argument to be "self-evidently right" is either a statement of faith or of arrogance, but it is not a logical argument. Only logical arguments are of use to readers. Perhaps you can share the reasons you find something "self-evidently right".
Richard Shaw
Note that Shiller's P/E10 used the 10-year average of real (inflation-adjusted) earnings. The use of real numbers helps to normalize for the period of high inflation (and accompanying interest rates) leading up to the secular market bottom in 1982. Thus it is reasonable to compare today's P/E10 with the comparable ratios in both inflationary periods, such as the early 1980s, and deflationary periods, such as 1921 and 1932.
On Feb 27 11:04 PM dshort wrote:
> Excellent article. I prefer the valuation method of Yale Professor
> Robert Shiller over the Siegel's approach. Shiller smooths the P/E
> calculation by using the earnings average of the previous ten years
> as the divisor. With this method, the historic P/E10 average is
> 16.3. The February 2009 close saw a P/E10 of 13.
>
> Does this make the market cheap? With the Shiller calculation, secular
> lows around the 1921, 1932, 1949, and 1982 bottoms have been in the
> single digit range. Here's a link with charts to illustrate: dshort.com/articles/20...
>
>
>
"The S&P method, flawed or not, has been consistently applied for a long time, making historical comparison of current P/E ratios with prior P/E ratios useful"
There are enough upheavals in the investing world without moving the goal posts on historical comparison benchmarks
Anyone know what the P/E is without the financials? I see plenty of single digit P/Es everywhere outside of them especially in infrastructure and energy.
On Feb 28 10:36 AM pelican wrote:
> Just to clarify, I meant the financial sector's lack of earnings
> contribution may be masking a cheaper market than what it appears
> to be based on total reported operating profits.
In 1974, P:E ratios were single digits, and the S&P 500 was back to the mid 60s.
The important question is whether earnings will recover to their long term rate of increase of 6 per-cent per year.
As reported earnings for the S&P 500 peaked in 2006 at $81.51. Will they never reach that point again?
Forward earnings, operating earnings and all the rest have no place when trying to compare today's earnings with stats from the past. "As reported", trailing earnings are what should be used.
> I agree - this bear market is unlike all others(they always are)
> in that financial earnings were completely destroyed and it's skewing
> the average P/E.
>
> Anyone know what the P/E is without the financials? I see plenty
> of single digit P/Es everywhere outside of them especially in infrastructure
> and energy.
Good comments and questions but will you accept the answers?
You can see earnings right from S&P at
www2.standardandpoors....
Earnings for Q4 of 2008 were NEGATIVE even excluding financials. Anyone who didn't already know that really shouldn't be f'ing around in the stock market. It's basic investing 101.
As for "financial earnings were completely destroyed" did you stop to think about why earnings were growing so fast as the bubble grew over the past two decades? They were growing in large part because financial geniuses were creating wealth where none actualy existed and now it is all deflating out of the economy in the form of lower house prices, lower commodity prices, lower wages & higher unemployment.
We have several years to go before this deleveraging is completed, perhaps a decade or two given the ineptitude of our government.
> To SteveTN and pelican
>
> Note that Shiller's P/E10 used the 10-year average of real (inflation-adjusted)
> earnings. The use of real numbers helps to normalize for the period
> of high inflation (and accompanying interest rates) leading up to
> the secular market bottom in 1982. Thus it is reasonable to compare
> today's P/E10 with the comparable ratios in both inflationary periods,
> such as the early 1980s, and deflationary periods, such as 1921 and
> 1932.
Inflation pushes stock prices down as it is happening but eventually any and all inflation shows up in stock prices and so "adjusting" for it is counting it twice.
"Considering" inflation when valuing stocks is smart but there is no reason to build it into a chart and then use that chart to compare hsitoric stock prices.
> Just to clarify, I meant the financial sector's lack of earnings
> contribution may be masking a cheaper market than what it appears
> to be based on total reported operating profits.
See my other comment about this or look at www2.standardandpoors.... but I wanted to point out that "operating profits" are not what has historically been used. You should use "as reported" earnings since that is what has always been used.
As you compare the two you will notice that "as reported earnings" and "operating earnings" are getting farther and farther apart this year and last as companies attempt to hide their true losses.
We will gladly change our view when, and only when, the charts show credible evidence of a reversal of the obvious current downward trend. We don’t intend to risk capital on what “ought to”, “should”, “might”, “could” or “someday” will happen to prices, or on what spokespersons for asset gathering / asset fee dependent enterprises say we should do."
++++++++
So, then, the P/E ratio is not a considertion. That being the case, and you write that you have not studied what Siegel suggests, What is the point of the article?
>When I see a companies like
> Proctor & Gamble, J&J, and so forth selling at 12x next year's
> earnings, it makes me wonder if this isn't a good time to be a selective
> buyer.
I gave your comments a thumbs up because you asked some good questions but let me address what you wrote above:
One of the things that got us into this overvaluation mess for both stocks and real estate was using next years' earnings. In a deflationary period, which we are clearly in, just about every company will have to fight just to maintain earnings and history and available data tells us that such a period is likely to be around for a few years.
Also, companies like J&J and P&G, along with Wal-Mart, do not historically carry high P/E ratios because they are slow growth, defensive type companies. In 1982 they would have probably been selling at 5-8x (as-reported) earnings. I recall that time very well and knew that we were in a very rare period of opportunity once the inflation spiral ended, but only once it ended.
The buying opportunity will come once we have unwound from this deflationary spiral and the key is to know when that is. Just keep in mind that you would have missed some nice appreciation if you had waited until 1933-1934 to start buying but you would have gotten hammered if you had started buying in 1930-1931. Much better to wait until there are some clear signs other than that stocks prices are lower than they were.
Keep in mind that inflation pushes the price of everything up; even stocks. Deflation pushes the price of everything down and is much more insidious in its destructive power towards stocks.
P/E is important. The P/E today is too high. A low P/E and a falling price is not a good buy. A low or moderate P/E and a rising price is a good buy. That would be one point of the article, but there are others, such as the key topic, which is that S&P publishes reasonable, consistent and useful number, and Siegel does not publish historical data to support his claims. There are even more points, if you care to think about what you read.
2. Stock markets look forward , so I don’t think looking at trailing earnings and PEs thereof makes sense.
3. Yes in the bear markets PE multiples go down (a lot) – as low as 6 in ’82. Multiples are a function of prevailing sentiment – it is of course very low at the bottom.
4. The current projected earnings are $40 (not consensus but projected today by Merrill Lynch today, S&P estimates are $48) – 10 multiple would lead to 400 – 480, a 12 -> 480 – 576. Any which way you look at it – lot lower from here.
5. Economic numbers are horrible, getting worse.
So brace yourself, there simply is no place to run or hide.
In his "The S&P Gets Its Earnings Wrong" (op-ed, Feb. 25), Jeremy J. Siegel claims that Standard & Poor's systematically understates the earnings of the S&P 500. In his view, the recent losses of the financial companies in the S&P 500 should be discounted because of their diminished weights in the index.
His argument, however, fails the simple tests of both logic and index mathematics. A dollar earned or lost is the same, irrespective of whether it is earned or lost by a big index constituent or a smaller one.
Prof. Siegel's example of Exxon-Mobil illustrates why S&P's method of calculating earnings works. If large Exxon-Mobil earned $10 billion and small Jones Apparel lost $10 billion, index investors collectively -- and individually -- would bear a proportionate share of both Exxon's earnings and Jones's loss, despite the fact that the value of Exxon-Mobil's shares in the index portfolio is about 1,381 times the value of the Jones's shares.
To use an analogy, we could hypothetically view the S&P 500 as a single company with 500 divisions, with each division having earnings and an implicit market value. The smallest of these divisions could have an outsized loss that wipes out the combined earnings of the entire company. Claiming that these losses should be ignored or minimized because they came from a less valuable division is flawed.
Prof. Siegel's approach -- applying the weights based on market values to the results based on a company's earnings -- effectively mixes apples and oranges.
David M. Blitzer
Managing Director, Chairman of the Index Committee
Standard & Poor's
New York
On Feb 28 12:09 AM Ricard wrote:
> I really don't get Siegel's point:
>
> "Suppose on a given day the only price changes in the S&P 500
> are that the largest stock, Exxon-Mobil, rose 10% in price and the
> smallest stock, Jones Apparel Group, fell 10%. Would S&P report
> that the S&P 500 was unchanged that day? Of course not. Exxon-Mobil
> has a market weight of over 5% in the S&P 500, while the weight
> of Jones Apparel is less than .04%, so that the return on Exxon-Mobil
> is weighted 1,381 times the return on Jones Apparel."
>
> "Yet when S&P calculates earnings, these market weights are ignored.
> If, for example, Exxon-Mobil earned $10 billion while Jones Apparel
> lost $10 billion, S&P would simply add these earnings together
> to compute the aggregate earnings of its index, ignoring the vast
> discrepancy in the relative weights on these firms. "
>
>
>
> This is as it should be. Siegel goes way off thinking that earnings
> should be weighted like price. In his example, Jones Apparel with
> that enormous loss vs market cap would have an equally enormous negative
> EPS, which then would require weighing. If done properly, you'd come
> up with the same number had you simply done what Siegel accuses S&P
> of 'improperly' doing.
>
> Or, you could choose to eliminate price weighing on the index. Exxon
> gains $30 billion in market cap, great, the index goes up by 30 billion
> points (around 10% move upward). Jones loses $20 million in market
> cap, great, the index goes down by 20 million points (around 10%
> move downward).
>
> Think of it this way - If GE's financial division reported a $30
> billion loss, while its manufacturing reported a $15 billion gain,
> GE as a whole would report a loss. Even if you weighed the loss of
> the financial division vs its size compared to GE, you'd still get
> a negative P/E for the firm. There's no way to avoid this simple
> fact.
>
> Siegel's point is utterly ridiculous. Middle school math students
> could do better.
Worst case scenario is probably S&P 500 at 400.
out their cash by the way of dividends, they would have a
higher stock price.
Siegal is a crack pot
dividends are fine, but long term stock values are more important
A great resource on understanding valuation tools, check out the book Active Value Investing by Vitaliy Katsenelson. He lays out a great framework for attempting to perform proper stock analysis.