Ask Colin

For your fundamental scan do you still use PE Ratio, Dividend Yield and Price/NTA filters?

I now scan only for PE ratio. I was strongly influenced in this by David Dreman's brilliant book Contrarian Investment Strategies - The Next Generation. His work suggests that PE works best of the three measures I used before.

As always, when you change things, there is a trade-off of pros and cons.

It is a simple scan and can be done very quickly by hand if you do not have a computer (in our market - it would be a very long job in the US market, because there are so many more stocks).

A disadvantage is that I now have to do the dividend yield scan manually on those stocks. You would be right in saying that I could put the dividend yield into the scan. The thing I wrestle with is that computers do what you tell them, rather than what you want them to do. So, you say - find all stocks with lower than 10 PE ratio and more than 5% dividend yield.

As it goes along, the computer finds a stock with a PE ratio of 8 and a dividend yield of 4.99%. The computer rightly says that Colin did not ask for that one, so it does not list it. But that stock may be fine and I would have put it on the list if I had been doing it manually.

The problem is that where you put the rule is arbitrary, especially when there are several parameters that might be to some extent trade-offs. There is a system where you put all three parameters in an equation that comes up with a single number. However, this is at best a partial solution.

Finally this choice of parameters becomes a question of personal style. I like to be able to use my judgement and experience. I think it is one source of the edge that makes my methods work. Someone else might be quite happy to program it closely, so no judgement is called for. That is going to be fine for them. I just think that investment (even more than trading) is difficult to boil down to a simple set of rules.

I generally do not pay much attention to price/NTA at the scan stage any more. However, I may and do look at it in my fundamental research. But even there it is not a critical test for me any more.

Finally, this change is also partly a reflection of the times. In the bull market, I tended to scan for value stocks and then look at the trend manually. That gave me a workable list. However, in the bear market that gives too many stocks, so I scan for trend and then assess the value ratios manually.