This is the first part of a two part series asking the question, "How cheap is the market right now?" (part II is here).
The answer might surprise you. It certainly raises some very interesting questions as to what cheap is, the importance of having a long term perspective. It also begs the question of how much patience long term investors have when it comes to thinking about various metrics.
The question itself involves a combination of data analysis and opinion. To fully explore this issue, we will listen to two different perspectives on the subject: One says the S&P500 is cheap, the other asks, how much cheaper might it get?
For part one, we go to Eddy Elfenbein of Crossing Wall Street: Eddy observes "S&P 500 is now trading at just under 16 times trailing operating earnings. The P/E ratio hasn’t been this low since October 1995." Note that he references actual trailing earnings. This is more accurate than using forward forecasts, which tend to be very wrong at key turning points.
>
SPX 500 Trailing Earnings at 10 Year Low
click for larger chart
>
Eddie also notes that "What was unusual about the rally that began in March 2003 is that it came well after the bottom in earnings."
Note, however, the P/E continues to decline, even as the market rallies. This is consistent with my own views of P/E multiple compression (see this 50 year P/E chart).
>
Tomorrow, we hear from a hedge fund manager who will point out what this means historically.
>
Source:
The S&P 500’s P/E Ratio is at a 10-Year Low
Eddy Elfenbein
Crossing Wall Street, May 26, 2006
http://www.crossingwallstreet.com/archives/2006/05/the_sp_500s_pe.html
I would like to see the same chart ex-oil companies.
Matt–
It would be interesting to see a number of overlays by sector to see how p/e multiples have faired over the same time period–not just absent the oil stocks.
Leisa
It would be interesting to see S&P 500 sector weightings over the same time period.
Andrew
It seems that the period when S&P 500 has below 16 P/E ratio is 70’s and 80’s, at the time the inflation and interest rate were very high. You may want to overlay S&P P/E ratio with interest rate or inflation. I suspect with 5% interest rate, 16 P/E ratio is indeed cheap.
If I remember correctly, S&P changed the formula they used for calculating the p/e ratio (which lowered the ratio), and at the time said they weren’t going to go back and restate historical data. If so, this chart is pretty meaningless.
The P/E could rise sharply if earnings fall. Earnings have been hitting record levels these past few quarters, but they have been very volatile in the past.
S&P 500 earnings have gone from 13.7 in 2000, to 3 in 2002, back up to 17.3 in 2005. I dont think 1 quarter or year of earnings mean much. Maybe a 5 year rolling average of earnings? That would make a nice chart.
After all, if you were going to buy a business, would you only look at the past 1 years profit? or average of the past 5 years?
For the period from 1996 onwards, S&P has published “core” earnings, which includes employee stock option expenses, restructuring charges for ongoing operations, write-downs of assets (but not gains or losses from asset sales), pension costs (but not gains), purchased research and development expenses, and excludes income or expenses related to hedging, mergers and acquisitions, and litigation and insurance.
The idea is that “core” earnings are more comparable to historical earnings, particularly up to the mid-1990s, when “as reported” and “operating” earnings became increasingly manipulated.
Currently, the S&P 500 is 19 times trailing core earnings of $67.36.
Core earnings haven’t been published for 1995, but comparing “core” to “as reported” and “operating” earnings in 1996, the PE for core earnings may have been below 19 during 1995.
But does this mean that the market is cheap? Defintely not. Historically, a PE of 19 has almost always been considered expensive. The exceptions have been for the periods that have resulted to bubbles, such as the late 1990s.
S&P 500 index:
http://www2.standardandpoors.com/spf/xls/index/SP500EPSEST.XLS
Measures of Corporate Earnings:
http://www2.standardandpoors.com/servlet/Satellite?pagename=sp/sp_article/ArticleTemplate&c=sp_article&cid=1054325046344&b=10&r=1&l=EN
(Note that Warren Buffett is included in the acknowledgements.)
This is too shallow an analysis to merit serious consideration. A chart of P/Es from 1996 to present is presented as a historical chart? Show me a chart going back at least to the beginning of the post-WW2 period. That is a historic chart. Showing P/E ratios for the peak of a wild bull market is an attempt to mislead readers into thinking that we are at the bottom of the P/E ratio range.
I searched for a chart of historical S&P 500 P/Es and found one. Surprise, surprise, I found one at this website. This too is a chart of trailing P/Es, but goes back 50 years. Let REAL analysis begin! http://bigpicture.typepad.com/comments/2005/12/pe_vs_sp_500_50.html
Good point about the commodity stocks,
The value is in the oil and the nat resource stocks,
the other sectors are doing nothing like GE, IBM, MSFT,
etc, yet we hear the non-commodity bulls pumping
the “value” of US stocks, yet not admitting that
the value is in the resource sectors.
Just to play devil’s advocate for a minute, how well do historical stats actually apply to this market?
Not trying to be sarcastic or pointlessly combative… the question goes back to “reversion to the mean.”
Most of the time, historical norms have value as reference points ’cause market valuations are historically mean reverting. But when you get a true paradigm shift–an underlying structural change in the basic nature of things–historical norms can go out the window. Think of all those economists convinced that $50 oil would grind the global economy to a dead halt, or all those traders who thought copper was a sure-fire short at $3, etc etc.
Food for thought from Marc Faber, via Bloomberg last month:
“If the Dow Jones goes up three times in the next 10 years, I think gold prices will go up by a minimum 10 times to something like $6,000 an ounce,” said Faber, 60, who founded Hong Kong-based Marc Faber Ltd. and manages about $200 million.
Point of the quote being that stocks as an asset class are best measured relative to other asset classes, kind of like the dollar vs the euro (or all fiat currencies vs gold). It’s possible to juice the market’s returns via credit stimulus, as long as the *winners* pretend not to notice that inflation has them running in place. With a little sleight of hand, stocks can keep going up even as their intrinsic value goes down.
A “managed decline” in the dollar is looking all but ordained from on high, with the only question being whether that decline will turn into a rout. Furthermore, maintaining the perception of a healthy equities market, if not the reality of such, is very important for all the players that matter: the big hedgies and mutual funds, the government, foreign central banks, the federal reserve itself.
Against this backdrop, with all the powers-that-be invested in the illusion of a healthy equity market to maintain a fragile psychology of optimism, what are the odds that something like what Faber speaks of could come true: the stealth-reflation machinery continuing to pump the equity markets like an inflatable air mattress, as hard assets call BS louder and louder with no one listening.
Not really sure what the alternative scenario is, other than letting this leveraged leviathan slowly self-implode… in which case stocks won’t just be cheap, they will be brutally mauled and left for dead. Hard to see the shot-callers and lever-pullers going for that.
Either way, we seem destined to seriously overshoot the old PE measures or drop way, WAY under in the next few years…
The P/E is only cheap if you believe that future earnings are higher than the present. If earnings do not increase then the P/E is actually more expensive than it appears to be.
trader75, here is an article with some charts you might enjoy. Talks about the themes you touched on (fed, pumping money, inflation etc): http://www.contraryinvestor.com/mo.htm. If someone happens to click on this link at a later time, be sure to go the May 2006 monthly commentary. That is what I’m pointing to.
Seems like there are about 50 different PE comparisons floating around out there. For mine, I prefer Hussman’s site and his analysis. It’s consistent, historically accurate and long term. I guess that makes me a grumpy old Bear but the rest of these measures I find to be The New Math. I think J. Hussman started to see these floated around also (at this inflection point), and decided to weigh in.
http://www.hussman.net/wmc/wmc060530.htm
Good article bt
I like how the liquidity discussion is evolving. It all seems to be pointing in the direction of blaming the central banks with their tolerating asset bubbles, and then pumping liquidity into the mistake when it deflates.
In the long run I think this is going to mean the Fed in order to unwind this will lean into restrictive territory throughout this expansion. Short term bills and cash seem to be the sweet spot for this.
Here is another article on excess liquidity
Submit any email address to get in.
http://www.2000wave.com/gateway.htm
PE is low because earnings have ballooned. All you have to do is look at the chart of long term earnings for a few major stocks on CNN Money and it’s pretty clear that Americans have done 20 years of shopping in 5!
I’m betting this level of earnings is unsustainable and I’m probably not the only one.
example:
http://money.cnn.com/quote/chart/chart.html?pg=ch&symb=WMT&time=all&compidx=aaaaa%7E0&comp=&ma=0&maval=60&freq=1dy&type=2&uf=0&lf=67108864&ind_compind=
just FYI as a public service, tinyurl.com is a great (and free) antidote for those super-long links that extend off the edge of the page