« Best Book of 2007 | Main | Michael Shermer on the Mind of the Market »
Latest Stories
- Excel Wankers and Recession Averages
- Sorry, New York is Closed. Check Back Later.
- Catching Falling 2009 Earnings Estimate Knife
- Survivorship Bias in Global Markets
- Talking Positions on a Lazy-ish Retirement Portfolio
January 30, 2008
The Earnings/S&P 500 Connection
What is the relationship between corporate earnings and stock market performance? The naive view says that higher earnings -- or at least the anticipation thereof -- leads to a higher market. But is it so?
Well, let's just that it's complicated. I'll post some more slices of the data later, having spent some Excel airplane time today looking at S&P 500 earnings and market performance data back to 1960. Here, just to get you started, is binned annual earnings data for S&P 500 companies compared to average same-year stock market performance over the period.
Sphere It
| 
Digg it
| 
Bookmark it
| 
Stumble it
| 
Facebook it
Does your data break things out by quarters? It would be interesting to search for the lag between stock moves and quarterly earnings that gives the best correlation and then see whether that is still negative.
My guess would be not really a useful relationship. They key is--changes in earnings, are they rising our falling. projected and actual. That is a variable you can time and the correlation is very close to 1.
Do you have any confidence intervals for these numbers? My guess is that the intervals are pretty broad over such a small data set. It gets even worse when you factor in that the rest of the world is able to leverage the same data going forward.
A graphical representation of earnings growth and annual return could show a leading or lagging correlation OR more likely, none at all!
Joseph Ellis showed that corporate earnings are a lagging indicator for stock prices. See his 2005 book. I believe he found that very often the peak in corporate earnings comes after the economy has already taken a significant downturn, and in some cases when it is in recession. But certainly earning can still be increasing into the onset of a bear market, and they are usually still going down during the early days of a bull market.
Comparing percentages has a problem with the magnitude of the numbers. If IBM earnings went from $1 to $1000, they have 1000% growth but IBM is very sick, it is running at break-even. If IBM earnings went from $1B to $1.1B they have 10% growth. The 10% growth IBM is the healthier stock.
I don't have the proof, but the rule-of-thumb is:
1) Falling interest rates, rising earnings is best
2) Rising rates, falling earnings is worst.
3) Falling rates, falling earnings & 4) Rising rates, rising earnings is in-between, for equity returns.
Doesn't this table just suggest that buying low and selling high produces better average returns?
Companies with negative earnings growth and strong balance sheets are strong buy candidates. Although you've teased Ford on your blog in the past, given its potential for growth overseas and its relatively strong balance sheet (and off-balance sheet asset) position, I would say it's a prime example of the kinds of companies that fuel the 15% average returns you show in this chart.