Someone on Kudlow’s show misquoted me on Monday (Larry?), and I wanted to get this specific data correct and distributed.

The year-over-year earnings indicator is based on the work of Marty Zweig; Its something I have mentioned for several years now, and it has proven to be an accurate and reliable tell. Its based on data since 1927, and includes the great crash, as well as recent bull markets.

Specifically, double digit earnings growth are not a Sell Signal. What they have historically signified, however, was a future period of market underperformance.

Nor are really awful earnings a buy signal. Very bad numbers lead to major market losses.

>>
Market Performance
Let’s start out assuming an average SPX return of about 11%; we can make
that number higher or lower depending upon which time periods we use,
whether we reinvest dividends, etc., but for this illustration, lets
work with 11% as out average annual market return.

Next, we breakdown earnings returns into 5 separate groups of Year over year percentage gains or losses. They are:

Gains Above 20%

Gains between +10 and 20%

Flat +/-10%  (between losses of10% and gains of 10%)

Losses between -10% – 25%

Losses worse than 25%

Let’s review these:

The best year-over-year earnings gains of over 20% produce only mediocre subsequent market performance — over the next 12 months, S&P 500 gains on average a mere 1.9%. Rather underwhelming. It is not a sell signal, but it does suggest that other asset classes might be preferable.

Conversely, when earnings are at their worst — when they fall 25% or more – subsequent market performance is the worst of the group. Markets are down nearly 30% over the ensuing 12 months. Ouch.

When earnings are between 10% and 20% — where we are today — the subsequent market gains are positive, but are  historically mediocre — about 5.8% over the next 12 months. Compare that with the risk free 10-year Yield of 5.1%, and again, you can see why this is underwhelming.

The remaining two groupings produce good, and excellent returns: When earnings are plus or minus 10% from flat year-over-year, we get subsequent returns of 9.3%. That’s pretty close to our historical average. One would expect these types of gains in the early to mid part of the business cycle.

The best market performance is in a negative range: down 25% to down 10% earnings; Subsequent market gains are a fabulous 28.6%.

I never like to rely on a single variable, and my work with the present cycle of P/E multiple contraction implies that we may be looking at numbers somewhat below thes average market performance.

Hence, why I am so unenthused by double digit year over year earnings gains.   

>
Why Do Earnings Gains Forecast Market Performance?

My explanation for this is simple: Earnings reveal where we are in the broader market cycle, and also provide some insight into market psychology. I hope to have more on this issue soon . . .

>

Here’s the same info, in chart form:

earnings_vs_market02132004173402.gif

 

>

Sources/Prior References:

The Sweet Spot ?       
http://bigpicture.typepad.com/comments/2004/02/the_sweet_spot_.html

How to Use Earnings as a Buy Signal                     
http://bigpicture.typepad.com/comments/2005/02/how_to_use_earn.html 

Earnings and Subsequent Market Gains                      
http://bigpicture.typepad.com/comments/2004/02/chart_of_the_we_2.html   

Earnings or Multiple Expansion?                
http://bigpicture.typepad.com/comments/2005/08/earnings_or_mul.html   

Single vs. Multiple Variable Analysis in Market Forecasts            
http://bigpicture.typepad.com/comments/2005/05/single_vs_multi.html

Category: Earnings, Financial Press, Investing, Markets, Technical Analysis

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

6 Responses to “Earnings and Subsequent Market Performance”

  1. spencer says:

    Historically, one of the most dangerous times for the market is right after the y/y Ch. in eps growth turns postive. A prime example of this is 1987.

  2. royce says:

    Barry, it’s odd that a -10 to -25 drop correlates with huge gains while -25 correlates with huge losses. That is counterintuitive. Any idea why it works like that?

  3. bob3 says:

    It would be good to know how many observations are in each of those categories. It seems a bit odd that only 1 category produces a greater than “normal” (ie, 11%) average return. Must be a fair number of years that fit into that category.

  4. Not really counter-intuitive. Really huge drops indicate the economy has just hit a wall and that worse is ahead. However large drops afterward generally mean the worst is behind you and gains are going to turn up and have soft comparisons.

  5. links for 2006-04-28

    Earnings and Subsequent Market Performance Ritholtz lays out one method of how earnings can be used to forecast market performance (tags: Earnings Forecast MarketHistory)…

  6. Obi Wan says:

    Excellent work. I take this as evidence that as of June 12, 2006, the market will NOT lead into a bear market sicne earnings projections are so positive for 2006-2007 See http://www2.standardandpoors.com/spf/xls/index/SP500EPSEST.XLS.

    So, why so much gloom and negativity on the market right now since projected earnings are so good?