5 traits of a perfect long-term investment

It only takes half an hour to roll your-own algorithm for investment success - though implementing it might take a little longer.

Common sense algorithms are better at making decisions in noisy environments - like the stockmarket - than expert intuition. Psychologist Daniel Kahneman invites readers to roll out their own in Chapter 21 of his book Thinking Fast and Slow. He says:

"The message of this chapter is readily applicable to tasks other than making manpower decisions for an army. Implementing interview procedures in the spirit of Meehl and Dawes [researchers who demonstrated the superiority of algorithms] requires relatively little effort, but substantial discipline."

This is a step-by-step summary of Kahneman's advice for recruiters. He says the preparation of the algorithm (steps 1 and 2) should take no more than half an hour:

1. Select a few traits for success, he suggests six. They should be as independent as possible from each other and you should be able to assess them by asking a few factual questions.

2. Make a list of the traits and think about how you will score them, say on a scale of 1-5. Decide what those scores mean.

3. Collect the information one trait at a time, to stop your enthusiasm or disdain for information pertaining to one trait from contaminating the others, an example of the "Halo effect".

4. Resolve to hire the candidate with the highest score, even if there is another you like better.

Discipline is often the first word off the lips of successful investors when explaining how they win. It's not much of a stretch to think of recruiting shares to your portfolio as a similar activity to recruiting employees to a company.

The process may involve interviewing executives, although it's more likely, at least in the first instance, to mean interrogating a company's annual reports. The algorithm gives you the questions you need to ask and puts the answers into a format (the score) that enables you to compare candidates.

My Decision Engine is to be a five-factor algorithm based on the traits of a good long-term investmentEver since I started publishing checklists I've been using them as crude algorithms to rank the shares I follow in a spreadsheet I call the Decision Engine.

When I wrote last week's precursor to this article I had intended to share the latest version with you. However, re-reading Kahneman in preparation, I realised I had over-complicated the algorithm. I need to sharpen it up.

What I'm about to describe is what the Decision Engine is still becoming: a five-factor algorithm based on the traits of a good long-term investment. Each factor is scored 0-2, so the maximum score is 10 and the minimum is zero. The business should be:

1. Straightforward. It's apparent how the company makes money. The business model should be comprehensible, and the accounting should be clear.

2. Generating excess returns. Over a long period, preferably at least a decade, the business should have earned a return on capital not less than 8%, and more in most years. It should also be generating enough cash to fund present and future activities.

3. Resilient. The business should profit from activities that are likely to be relevant in future. Preferably it should do these activities in a distinctive way that is costly for others to copy, making its product or service especially valuable.

4. Managed for the long-term. Judge management by their actions and their incentives. They should invest steadily to grow and adapt the business. They should profit by making the company better, preferably through their ownership of shares.

5. Valued attractively. I use a much adjusted cousin of the price earnings ratio, an average earnings yield, to compare company valuations. An earnings yield of 10% or more scores two. Less than 4% scores zero. For earnings yields between 4% and 10%, the score rises with the yield from zero to two.

I doubt those factors will surprise many seasoned investors, they're well established hallmarks of good long-term investments and I've sought them out for many years.

What stops more people from buying and holding shares like these is the discipline it takes to find them, screening out the many competing factors that make the stockmarket so noisy, and decision making so error-prone.

Selection bias

How I weigh these factors is changing, though. The longer you hold shares in a company that prospers and the more it appreciates in value, the less the initial price matters.

Originally, a share's valuation score was worth 50% of its total score with ten other factors accounting for the other 50%. Currently my algorithm has eight factors in total and the weighting of the share's valuation is 30%. The reduction in the number of factors to five means the valuation weighting now falls to 20%, if they are equally weighted.

That decision has also been influenced by Kahneman. The evidence shows, he says, that more complicated weightings introduce "selection bias".

About step 4 in Kahneman's process, which urges investors against using intuition to override the algorithm - even if it's strong. He doesn't shut the door completely. He describes a thought experiment proposed by the algorithmist Meehl.

Broken leg rule

Suppose you had an algorithm that predicted whether a particular person will go to the cinema tonight. It would be sensible to disregard it if you learned that the person had broken his leg earlier in the day. Apparently the term "broken-leg rule" is now used in the academic literature to describe a rare and decisive factor that justifies exercising judgement.

He's wary, though, of letting intuition in through this crack in the door, saying:

"...try to resist your wish to invent broken legs to change the ranking."

I suppose a cast-iron "broken leg" factor might be information that a share is about to delist, or a company is making an acquisition so big it will effectively become a very different company.

I'm not sure if doubts about management style, for example the way I feel about Games Workshop's board, constitute a clean break, or just a minor fracture. That's why I'm agonising about it so much!

Contact Richard Beddard by email: richard@beddard.net or on Twitter: @RichardBeddard

This article is for information and discussion purposes only and does not form a recommendation to invest or otherwise. The value of an investment may fall. The investments referred to in this article may not be suitable for all investors, and if in doubt, an investor should seek advice from a qualified investment adviser.