If making money is all you care about, all that matters in stock market investment is valuation. Except in the case of a company going bust, any share can be a profitable trade at the right price.
Personally, I care about the kind of businesses I invest in, so I spend time making sure they do something I approve of.
But there's another reason why valuation is my fifth and final criterion, only considered after I'm satisfied that I understand what the company does, and that it's highly profitable, capable of remaining so, and run by executives who should ensure it stays on track.
I think about the companies first, because you can only value something if you know what it is.
Most investors use relatively simple metrics like the famous price/earnings (p/e) ratio to decide whether shares are good value. The 'e' in p/e refers to the earnings attributable to shareholders over the financial year. The 'p' is the market price of the company.
If the price is 15 times earnings, we say the company has a p/e of 15, which is another way of saying it would take 15 years of earnings at the current level to double our money if those earnings were paid out to shareholders in their entirety as dividends.
Every year we would receive 1/15 of our original investment, which is just under 7 per cent. Expressed as a percentage, the p/e becomes an earnings yield, which is like an interest rate.
A company that costs 10 times earnings (its p/e is 10) is better value than a company that costs 15 times earnings, because it promises a 10 per cent return as opposed to 7 per cent.
There is a complicating factor, though: the earnings of companies vary a lot from year to year. While a company might earn 7 per cent of our original investment one year, it might earn only 4 per cent another, and 14 per cent a year later. Depending on the company, the figures could be alarming.
For turnarounds, cyclical companies or start-ups, the p/e is an unreliable guide - a bit like a compass in the Bermuda Triangle.
Establishing a realistic value for future earnings amid a corporate restructuring or while a company has yet to establish a profitable business model is probably beyond most of us, even sometimes those very close to the situation.
One way around this problem is to stick to the most predictable companies. The purpose of my first four criteria is to find stable businesses where future profits are likely to be much like they were in the past, only somewhat better.
There are no guarantees of course, but over time most of them should earn returns higher than their earnings yields imply, because they can invest in their highly profitable operations and grow.
The excess returns of these companies, will, I expect, outweigh the disappointing returns from firms that are less predictable than I thought.
So how much do you pay for a firm that should perform much like it has in the past, but somewhat better? We answer the question by asking another.
How much do you want to earn? 12 times earnings, a p/e of 12, promises an earnings yield of 8 per cent. That's my benchmark for value.
I prefer not to pay more, but I'm prepared to pay up to 20 times earnings, analogous to an earnings yield of 5 per cent, in the expectation that growth will bring returns above the benchmark.
If that sounds arbitrary, it is; but only to a point. The joy of expressing a p/e as an earnings yield is that you can compare it with the alternatives.
My expected minimum return of 8 per cent average over at least 10 years compares very favourably with yields on bonds and the interest on cash deposits.
Shares are more risky, we're told. However, not only am I expecting a higher return, but the first four criteria should have directed me to the most stable companies. The least risky, in other words.
Subscribe to Money Observer Magazine
Be the first to receive expert investment news and analysis of shares, funds, regions and strategies we expect to deliver top returns, plus free access to the digital issues on your desktop or via the Money Observer App.Subscribe now