Oblique portfolio selection

I will soon publish the initial composition of the first Obliquity Portfolio, but before that, here are the nine criteria I’m using to assess stocks. Each individual criterion is either a pass or fail, and a stock is considered for admission if it passes 7 (out of 9) or more.

The criteria aims at identifying “good companies” that are likely to be long term profitable, in order to build a diversified “index less crap” portfolios. The aim is to rebalance as little as possible — to “never sell” in a Warren Buffett kind of way.

The criteria are primarily qualitative, and to a large extent valuation-free: we’re trying to find good sustainable businesses; not, as such, cheap ones.

1) Useful business

A business passes if it produces a generally useful product, that people need or enjoy, and don’t need to be tricked to buy. For example a wine producer or a mattress manufacturer are good. A casino operator or a financial intermediary whose success relies on obfuscating their fees are not.

I include there soft ethical criteria, like excluding the makers of weaponry, and tobacco companies, as killing people generally destroys value; even merely stockpiling killing gadgets is a dead weight on the economy. I guess that it is a debatable point in pure investment terms, this could mean missing a missile manufacturing bubble, but then it’s not a bubble I particularly want to be part of.

At a company level, a particularly well run or keenly priced tank manufacturer could escape,  but given that there are thousands of investment options to choose from, random but limited restrictions on the investment universe shouldn’t hurt.

2) Long term

Basically here we’re trying to answer the question: is there a high chance that this business is still here in ten years time? I don’t want to have to trade all the time, and bankruptcies are bad for your returns.

3) Employees

This is based on the idea that people who enjoy themselves at work are more productive that those who don’t. For most people, including some who wouldn’t necessarily admit it, work is more than just a way to earn a living, it occupies a good chunk of one’s awake time, and forms part of who people are; so it is valuable for work to be rewarding for its own sake. Not all jobs are fun, but for any given task, companies can get in the way, or they can be helpful.

Assessing that for companies with thousands of employees is well nigh impossible. I’m attempting to guess from, on the one hand, management attitude, trying to read between the lines beyond the “our people are important to us” platitudes that are part of every annual report template, and on the other hand, whatever I can find on the web about employee testimonies, from company assessment sites like glassdoor.com.

4) Executives

Here we are trying to find out whether the executives are reasonably competent, and if not how easily they could destroy the company. Large companies are made of a lot of individual activities that essentially run themselves. Belgium recently ran without a government who could make strategic choices for almost two years, and so could most large conglomerates. Still, sufficiently bad strategy can destroy even businesses that could run themselves.

So the questions to answer are what leeway the executives have to destroy the business, and then do they appear to be sensible guys who care enough about the actual business, and are not outright crooks or simply dickheads. In their communication, they should talk about what the business does more than about accounting metrics or silliness about “shareholder value”.

5) Low volatility

Historically over the long term, a low volatility bias tends to return slightly more than the market, and markedly more in risk adjusted terms (slightly more returns over markedly less risk). This is known as the “low volatility anomaly”. The world hasn’t reached a consensus on an explanation, other than that it seems very widespread, across time, geography and asset classes. Like any anomaly that has been documented, it may not persist, but it seems a good bias to have anyway, and the risk it reverses to become a negative risk factor seems low.

As luck would have it, most of the other oblique criteria tend to imply low volatility. Still a highly leveraged or too fashionable company pops up on occasion.

This is assessed by looking at beta, volatility and particularly shaky charts.

6) Price action

The second and last criterion that uses price charts, through a quick glance at the chart (on long and short time scales) rather than any clever technical analysis, aims at capturing funny price action: a sudden drop, or spike, or a parabolic trend. The portfolio is for the long term and therefore excludes stocks which are currently hot, e.g. subject to takeover proceedings or rumours, or have just crashed, for whatever reason.

Short term small jumps, like a couple of percent, either way, after say an earnings announcement can be mostly ignored. This kind of noise disappears via time and diversification.

7) Balance sheet

While we’re not trying to do valuation, a few accounting checks can be useful. This is a cursory look at whether the balance sheet of the company seems reasonably solid: not too much debt, a sizeable amount of equity, no excess amount of “goodwill”.

The usual metrics like Price to Earnings are also looked at, but just to check they’re within reason. We’re trying to exclude basket cases, not to find fine grained value in the accounts. Basically we assume that most of the things that can be read in standardised accounting numbers, on a medium term horizon, are more often than not priced in by the market, given the many people who crunch and worship such numbers.

8) Free float

When someone owns the majority, or close to, the majority of the shares of a company, there’s no market for governance: if the controlling owner, through malice or incompetence, runs the company into the ground, and they won’t sell at any price, there’s nothing the market can do about it. When a company has sufficient free float, the price collapses and someone takes it over. This has grown into quite a competitive activity so that the odds that the entity taking a failing company over is better than existing management — and not mere asset strippers — are reasonable.

Besides, even if the company is well run and valued, someone may still wish to buy it, e.g. to combine it with their existing business, or for empire building, and usually have to pay a premium for this, and sometimes overpay, which is like a free call option for existing shareholders. This possibility does not exist in the same form when there’s a sticky controlling interest.

So a free float or 80% plus is usually desirable. The free float numbers published on financial websites are not always meaningfully reliable, so the regulatory disclosures are also checked.

9) Diversified

I don’t plan to buy hundreds of stocks just to get enough diversification, so the stocks selected need to be sufficiently diversified, both in what they’re doing — commensurate with the size and sector of the business — and geographically, e.g. have a scope similar to that of the portfolio, e.g. world presence for a world equity portfolio.

Most sizeable companies who are a leader in their domain do, and this is usually a refreshingly easy one to assess.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s