I have started to read "Financial Statement Analysis and Security Valuation" by Stephen Penman, and it is getting my juices going as to how to approach company valuation.
One model I have been thinking of is for defensive companies. My idea is that one can calculate the "return" on a company as the long-term growth rate + dividend yield. I think a test for a company's "defensiveness" is its R^2 of its EPS. With this in mind, let's look at Greggs, the retail baker.
It's EPS record looks like this:
Here's some stats produced by my Racket program:
EPS for 2009 actual is 34.1p. My program suggests it "should" be 35.5p. This is pretty good, and I don't think we need to read too much into it. I think it can be used to point out if a reported EPS is likely to be too low, or too high.
What's interesting here is that GRG has an R2 of 93.3% - in other words, the model I use is very explanatory of the EPS values seen. Maybe uncannily so. It's a useful number to look at, because it tends to confirm that we are, indeed, looking at a defensive company. This gives us some confidence that our growth rate is meaningful. In our case, the growth rate in EPS is 8.1% pa. A word of warning: EPS figures can be influenced by how capital is being expended. My model doesn't account for that. Never mind. GRG has a fantastic balance sheet, so I am going to take my growth rate of 8.1% pa as "good".
Now onto the dividend. GRG is currently trading at 450p, and its dividend yield is 3.9%.
So our expected return for this share is 12.0% (=8.1+3.9). That's a pretty good return - although not Buffett-like.