TMF: Lie, damn lies, and statistics / Investment StrategiesI was on the lookout for "good companies", and maybe not so good. "Good" is usually defined as "high returns on capital" - but I've looked at a few companies, and I'm amazed at how many cruddy ones have high ROCEs. Some ROCEs look bizarre - too true to be good, as my old Physics teacher would sometimes say.
I decided to have a look, instead of at ROCEs, which seemed to throw up some suspiciously high numbers, and looked at operating margines. I looked at median margins for, usually, the last 10 years, and the margins now. I also looked at the share price rise, the PE "then", and the PE "now". I also recorded the dividend yield. I selected a ragtag bunch of companies that had crossed my path for one reason or another. Well, here's some stats:
EPIC OPER SHARE PER YLD
MARGIN PRICE THEN NOW %
% CHNG
MED NOW %
AZN 27 35 2 29 8 3 AstraZeneca
BA. 9 10 -22 18 7 5 British Aerospace
BATS 10 31 404 9 13 8 Brit Amer Tobacco
BAY 5 -2 -47 32 NA 0 Brit Airways
BLT 22 24 595 12 12 3 BHP Billiton
BP. 10 10 -40 16 6 6 BP (who could resist including that)
BWY 17 6 -16 6 23 0 Bellway
CLLN 1 2 137 15 9 4 Carillion
FCAM 24 21 -82 23 8 11 F&C Asset Management
GSK 31 32 27 30 9 5 Glaxo
NICL 10 17 96 7 16 3 Nichols (they make Vimto)
TSCO 5 5 85 22 12 3 Tesco
XTA 23 19 -5 14 8 0 Xstrata (miners)
It's very interesting to get some historical perspective on this, and a mix of inductries.
It's a very mixed bag of companies and results, which makes it very interesting to look at. I think one thing is very clear: don't overpay for a business! Look at AZN and GSK. They've had huge operating margins - around 30% - but have been medocre performers (AZN is up just 2% in 10 years, and GSK only 27%. That's a measly capital appreciation of just 2% pa). This is explained by the drastic re-ratings that their PERs have undergone, going from 30 to about 9. Look at their operating margins, though. Despite the fears of cheap Indian copycat drugs, they are still earning operating margins in the 30% range. At PERs in single digits, they look quite undervalued looking at the kind of margins they earn.
The miners are quite interesting, too. BLT and XTA have operating margins in the 20's, which seems quite high considering that I heard that margins on miners are quite low. Special mention should also go to BP. It's margins are at about median for its decade. It's lost about 40% of its value over the decade, as the PE has gone from 16 to 6. No prizes for guessing for why that is.
What's also interesting to look at are the bad earners. Companies like BAY and CLLN are really bad at generating profits, and TSCO is no dynamo, either. Difficult to imagine that BAY was on a PE of 32 all those years ago. It doesn't look like it was an "especially" bad year for them, either; a possible source of distorted PE ratio.
BWY is a bit of an interesting company - it's a housebuilder. It's median operating margin is actually not so bad, at 17%. It's PE is currently 23 - but that's due to current low earnings.
The problem with this magic formula is that it does not prevent you from selecting losers. It’s a pure statistic method which, as far as we are concerned, is open to further improvement.
Determining which stocks are undervalued or which are justly beaten up, is a very difficult task, which we believe is best executed by people who know the company very well, the insiders. What’s even more important about inside tradings is that when executives bought shares in their own companies, the stock tended to outperform the total market by 8.9% over the next 12 months. Conversely when they sold shares, the stock underperformed the market by 5.4% (Nejat Seyhun, a renowned professor and researcher in the field of insider trading at the University of Michigan).