Panning for good, undervalued companies

No two companies are alike. Investing means choosing between apples and oranges: Megacorporations and small enterprises, companies that derive profit from physical assets and companies that derive profit from intellectual property, companies that account for their financial performance one way and companies that measure it another.

Databases massage data to make companies more comparable. They can't iron out many of the wrinkles, but they can iron out a few. Most of the important things about companies, though, cannot be boiled down to statistics.

My Decision Engine is a giant spreadsheet. Today, it contains data about the financial performance, market valuations, and strengths and weaknesses of 56 businesses, but that number is increasing as I add new companies. I do some tricks to normalise the data, but in other respects the Decision Engine is more than a data depository. It contains opinions.

Here's a snapshot (here's the spreadsheet in a window of its own):

It tries to incorporate everything I think about when judging a long-term investment, including soft factors like the quality and motivation of management, how the company makes money and how it plans to make more (its business model and strategy), and the extent to which financial performance is susceptible to external forces (like recessions or raw material prices).

I have a ten-point checklist, and if a company satisfies enough criteria to convince me a profitable company (earning good returns on capital*) is capable of remaining profitable indefinitely I give it a positive rating. At the right price, these companies are worth buying and holding for long periods.

If I think profitability won't be sustained, the company gets a negative rating. I can equivocate too. If a company has impressive qualities, but faces significant challenges, I give it a middle rating. To form these opinions, I ask questions of annual reports, of companies (principally at Annual General Meetings), and other investors.

The return an investor might expect from a company depends only partly on a company's return on capital. It also depends on how much the investor pays for their share of the return. I use two metrics: the earnings yield, which divides the market value of the business by profit earned in the company's most recent full-year, and a similar measure that uses a normalised profit figure, one derived from average return on capital over at least six years**.

The earnings yield is the hypothetical return from the investment. An attractive yield is a matter of judgement. Despite our best analysis, a company may not perform as well in the future as it has in the past, so it should be considerably higher than risk free investments like the interest on a bank account. I consider an 8% yield to be obviously good value for a decent company.

Lower yields require more thought. Companies that should continue earning more profit than they need to stay in business can often invest in new business and grow. The promise of higher profits in future means we can accept a lower yield now. In practice though, I find it hard to buy a share if its earnings yield is below 5%.

The Decision Engine ranks all of the companies it tracks according to their earnings yields on a scale of 1 to 10 (An earnings yield of 1% scores 1, and earnings yield of 10% or more scores 10). It also ranks the companies by my opinion on their ability to sustain profitability (they get a score of 0, 5, or 10). The two rankings are combined and the spreadsheet sorted to show good companies at attractive valuations at the top (highlighted in light green) and dodgy looking business and/or shares on heady valuations near the bottom (highlighted in salmon pink).*

There are apples, oranges, bananas, steaks and cartons of milk in the Decision Engine. There are companies I've been following for more than a decade and companies that I've been following for a few months. I've gathered ten years of data on some of the companies, and only two for others. I can't hold all that information in my head, but I've found that once I've poured it into a spreadsheet and let the sediment settle, the market (at least for the kind of stable, long term investments I'm interested in) becomes a bit clearer.

I don't believe that company five is necessarily a better investment than company seven, but I do believe that the top twenty are better investments than the bottom twenty and if I were to start a new portfolio today, I would draw from the top of the list.

About 50% of the Share Sleuth portfolio is invested in the top 19 shares that score more than 15 out of 20, and that proportion is gradually and purposefully increasing.

One of the joys of sharing the Decision Engine with (87 so far) other shareholders has been learning that many have their own decision engines. If you would like access to the Google Spreadsheet and you have a Google account, please send the email address associated with your account to richard.beddard@iii.co.uk.

*By profitable, I mean a company must be able to earn a return on capital of at least 8% on average for a long time. This is not an easy hurdle to measure, let alone meet! Return is profit, which is susceptible to all manner of accounting manipulations. To calculate return on capital you divide profit, the return, by the funds the company needs to stay in business, capital, a figure it's only possible to determine inexactly from the accounts.

Very broadly, I believe a company capable of earning an 8% return on the funds invested in it is a viable business. Because capital has a cost (interest, if it is borrowed), a company earning less than 8% return on capital may be just recycling money it's borrowed or raised directly from investors rather than making it. It may be creating less profit than the cost of the funds it uses.

**Profit in any single year may be unusually high or low. The "normalised" figure is smoothed, usually it's a more conservative measure. If the Decision Engine incorporates enough data to calculate the average return on capital over more than six years for a company, it uses the average of both normalised and current earnings yields. If it doesn't, it uses the current earnings yield.

***Market valuations are updated automatically as share prices move (delayed by 15 minutes). I update the financial data as soon as I can after a company publishes its annual results.

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.