Too Late to Short SPY? An Historical Perspective

by: Ron Nelson October 10, 2008 | about stocks: SPY    

Ron Nelson

About this author:
Become a Contributor Submit an Article
  • Font Size:
  • Print

So, the S&P 500 (SPY) is down over 40% from a year ago. All the talking heads on CNBC have their stories on what to do now. Scary times for sure, but what is the perspective that historical data suggest? Is it time to jump back in or do we have much further to fall? Should we go long on SPY, or short it?

The housing bubble has burst and we all buy into the idea that there’s another 10-25% to fall. Prof. Robert Shiller of Yale University has published historical data going back to 1890 (below) that shows a small variation in inflation adjusted home prices over 100+ years prior to 1995.  Variations in fundamentals like interest rates, population and building costs have no correlation to the tremendous ramp up in housing prices after 1995. Common thinking held that real estate always went up has been broken. The real estate tipping point is behind us.

click to enlarge images

What about stocks? Panic is all around us. Must be close to a bottom, right? The stock bubble has burst (again). So what are the fundamentals? Shiller’s data on S&P500 P/E ratios dating back, again, over 100 years is shown below. Note that to take the “noise” out of the data, Shiller uses earnings over the previous 10 year period.

The market closed Wednesday, October 8, 2008, at a PE ratio of 14.6 for the S&P500, a tremendous drop from nearly 45 in 2000 and 25 earlier this year. This may feel like we’re close to a bottom, but the historical average going back to 1880 is 16.3. We reached PE lows close to 5 in 1921, 1932, and the early 1980s. I wasn’t around for the first two of these, but my dad was. He says the news today sounds almost as bad as it was in 1921 and 1932. I recall the early ‘80s and this feels worse to me. And earnings are likely to go down substantially from here. 

Further, financial stocks over the last 10-15 years have contributed 30% of the earnings for the S&P500, well above their historical contribution of 10-15%. So let’s suppose that financials drop down closer to their historic averages, in the best case. And let’s also assume we’re heading into a timeframe of lower corporate earnings. Both of these take the earnings denominator down and drive the PE ratio up for fixed stock prices. 

So what would a cold hearted analysis of this fundamental data suggest? If we keep the 10 year earnings constant but the PE ratio drops well below the historic average and hits 10, the S&P 500 would have to drop from today’s 909 down to 622. If the PE ratio were to drop down to 6, then the S&P 500 would hit 374. If we take the 10 year earnings average down by 20% and the PE ratio to 6, then the S&P 500 would have to hit 299, a 67% drop from the Oct. 8, 2008 close. 

Markets usually are driven more by emotion and “common thinking” (as erroneous as that may be), but if one believes in fundamental measures, we could be looking at precipitous drops from here.

Further, common thinking is that the market always bounces back relatively quickly from bear markets. Just listen to those talking heads.  But historical data (see the chart below which shows the inflation adjusted S&P 500 index, also from Shiller’s web site) clearly shows that it takes 20 years to recover from major dips (1906 to 1928, 1929 to 1958 and 1969 to 1993). 

One can never predict “common thinking”, emotional reactions, tipping points, or the market, but this look back at historical data suggests we could see a 50% drop from here on the S&P 500. Believe that? Then short SPY. The tipping point regarding bounce back time for stocks may be here (again). I’m short SPY and when the PE drops well below 10, I’ll cash out. And I won’t be in a hurry to jump back in.

Disclosure: Author holds a short position in SPY

Related Articles

This article has 17 comments:

  •  
    Oct 10 09:11 AM
    Good analysis. The inflation adjusted P/E chart shows that recent stock market history has been more of a bubble than most would care to admit. Reversion to the mean as you suggest could take the market much lower than current levels.

    Waiting for a falling market to stop falling before buying is usually prudent advice. Even after the bottom is in there will probably be retests of the lows that offer good buying opportunities. No need to be in any rush.
    Reply | Link to Comment
  •  
    Oct 10 10:36 AM
    Agreed Smaty_pants.The market was in the green a few minutes ago and now is again in the red. The only thing for sure at this point is volatility. Certainly buy into this panic is not prudent, patience is the name of the game.
    Reply | Link to Comment
  •  
    Oct 10 10:46 AM
    Dow: 6260
    Reply | Link to Comment
  •  
    Oct 10 11:33 AM
    Very insightful article. Combine this with my technical analysis post
    Where Are We in the Stock Market Cycle?
    to put your arms around what the possibilities are for the market.
    Reply | Link to Comment
  •  
    Oct 10 11:38 AM
    The link to my technical analysis post:
    seekingalpha.com/artic...
    Reply | Link to Comment
  •  
    Oct 10 02:08 PM
    For more on this topic, check out today's nytimes

    economix.blogs.nytimes.../
    Reply | Link to Comment
  •  
    Oct 10 10:10 PM
    As painful as it is, your analysis has a lot of weight and the time frame matches with what Danielle Park mentioned on her blog:

    www.jugglingdynamite.c...

    But it doesn't mean we can't have cyclical bull markets between now and when S&P500 P/E bottoms out (around 6 or so).
    Reply | Link to Comment
  •  
    Oct 11 12:22 PM
    I may not understand the statistics here, but, if we're using 10 year averages of earnings, surely 1 or 2 years of bad earnings won't bring the avg down very quickly. I don't see how long term trends project short term behavior.
    Reply | Link to Comment
  •  
    Oct 11 07:48 PM
    Sundays NYT has a great interactive chart comparing every bear market as to depth and duration. Very Interesting. More yet to fall looks like
    Reply | Link to Comment
  •  
    Oct 14 02:26 PM
    Has anyone found a reliable and timely source of S&P500 trailing-ten year data? Robert Shiller updates his excel spreadsheet from time to time but the latest data is to August 2008...

    According to the latest article about this topic at online.wsj.com/article... the TTY SPX PE was 15 as of Oct 10. It was 20 in August.

    Many thanks in advance..

    Cheers,
    Matt Blackman
    Reply | Link to Comment
  •  
    Oct 18 11:37 AM
    "but if one believes in fundamental measures"

    Wait, are you seriously trying to imply simply looking at past historical P/E lows and making the assumption out of thin air we will revisit lows is a "fundamental measure"? No valuation metrics, no deep thoughts, just simple "it's happened before thus logically it must happen again, and probably this week". You're reading charts and are trying to fool people that this is "fundemental analysis". Unbelievable.
    Reply | Link to Comment
  •  
    Oct 22 11:36 PM
    Muzie,

    Thanks for the comment. As I pointed out, we all know that markets move on a combination of factors -- fundamentals, sentiment, earnings projections, overall world and local economics, etc. The fundamentals are that PE is still historically high and that earnings estimates are falling. Additionally, sentiment suggests things are as bad or worse that the early '80's and world economics are being bludgeoned by the credit and banking crisis.

    Given that, is it out of the question that we hit PE ratios similar to the '80's? No, and that suggests SP500 in the low 400's. If it's as bad as the '30's, then it's in the mid 350's.

    After writing the above article, I went long with ultra etf's for a while. Then went back to ultra short etf's as the "bear rally" looked to be running out of steam.

    Ron
    Reply | Link to Comment
  •  
    Oct 22 11:45 PM
    Matt,

    Sorry for the delayed reply. Here's the site for the latest SP500 earnings data:
    www2.standardandpoors....

    You can download an excel spreadsheet with the total SP500 earnings as well as the sector breakdown. Then use the same smoothing formula that Shiller uses in his spreadsheet. Note the time delay.

    Ron
    Reply | Link to Comment
  •  
    Oct 22 11:49 PM
    No Free Cake,

    You're right, lowered earnings for a year or two don't pull the 10 yr average earnings down by much. The main point of the article (which I may not have made a clearly as I should have) is that if the economic times we're heading into are really as bad or worse than the early '80's, then it's not unreasonable, as a "value investor", to think that the PE ratio could go as low as that of the early '80's, which would bring the SP500 down into the 400 range.

    Thanks for the comment,

    Ron
    Reply | Link to Comment
  •  
    Oct 22 11:56 PM
    Smarty Pants,

    I agree, no need to be in a hurry to go back in long, just play the short side for a while -- we're now below the mean, but it's unlikely to stop there. Perhaps one sigma below the mean? That would take us to 604. How about 1 1/2 sigma?

    Ron
    Reply | Link to Comment
  •  
    Oct 23 12:01 AM
    jlounsbury59,

    Thanks for the reference to your technical analyis. After what could be a bear rally, we're heading down rapidly toward your level 2 support level. What's your take tonight?

    Ron
    Reply | Link to Comment
  •  
    Oct 26 09:40 AM
    It is possible that the major indexes could move to single digit P/E’s but the comparison to the 1980’s is a weak one. In 1980 the inflation rate was about 13.5%; the Prime Rate was 21.5%. The Federal Reserve under Mr. Volker was determined to ring inflation out of the system. The big difference is that today the Federal Reserve is “one our side”, with low real interest rates, easy money supply, and a low inflation rate. In the early 80’s the Federal Reserve policies were the exact opposite of what we have today. As long as inflation is under control the Federal Reserve can cut key rates, and can have relatively easy money, just the opposite of the early eighties. This should have the effect of cushioning the economy to a softer landing. On top of that some early data suggests that the housing market may be forming a bottom here, news over the weekend about home sales in California. No one knows where market will bottom but the comparison to the 1980’s is a week one.
    Reply | Link to Comment
More Comments on SPY »
Top Rated Comment Streams:

Numbers are net rating-

See all Top 100 »
More by Ron Nelson

Investment Resources
Money Management
Research
Seeking Alpha provides these ads as a service to its leading contributors
Articles on related themes
Market Outlook
Short Ideas
Bonds
Dollar/Currencies