From action to analysis

Using your brain

This blog follows three previous articles on beginning value investing advocating:

1. Understanding the basic premise of value investing, that sometimes share prices are lower (or higher) than a realistic evaluation of the business would determine.

2. Buying shares that, statistically speaking, look cheap (see parts two and three).

Jumping straight from understanding (1) to action (2) by-passes the need to analyse companies, which is very time consuming, rather daunting and a potential block in the development of an investor.

Some value investors, mechanical investors, go no further. Relying on computer recommendations leaves them free to enjoy life without the stress of making decisions to buy and sell stocks that almost as often as not punish as much as they reward.

Traditional value investing, is ideally suited to a statistical approach because, when companies are on their knees, the precise state of their finances really matters.

The signals that indicate a company is failing, increasing debt for example, only really work with troubled companies. A growing company might be taking on more debt to finance new business and a highly profitable company might be able to afford more debt. If the company’s in trouble, as many really cheap companies are, then a dip in profit, or a rise in debt is unambiguously bad news and so the stats are better predictors of whether a company will survive and prosper.

But that’s not true for every company, it just works out that way if you invest in a lot of cheap companies with good finances over a long period of time.

To improve on the screens, the first step is to check against the original source, a company’s annual reports. The data used in financial software are culled from company statements and inevitably mistakes are made in transcription

Even an accurate screen is open to interpretation, though.

For example, during the credit crisis my Ultimate Bargain Screen filled up with property developers like Persimmon (PSN). The screen uses the price to net current assets ratio, Benjamin Graham‘s famous net-net measure, and, it seemed, shares in property developers were selling for less than the value of their most liquid assets (cash, stock and money owed to the companies by their customers) less all liabilities.

House builders tend to include land and properties they are developing, in the stock category, though, even if it’s likely to be more than twelve months before they are sold. Being doubtful these assets were liquid (i.e. easy to sell) during a property crash I ignored the screen and refused to add house builders to the Thrifty 30 portfolio.

The second highest ranking company on the Nifty Thrifty screen is currently drug company ImmuPharma (IMM), which appears to be enormously profitable. Its return on capital is 813% according to the screen. But Cephalon, a major US drug company has just exercised an option to license ImmuPharma’s drug Lupuzor, which treats Lupus, paying ImmuPharma $30m. The license could be worth $500m in total, depending on its progress through trials and Cephalon’s enthusiasm for the drug.

In a typical year, though, ImmuPharma makes a loss. Its high ranking is more to do with the fact that it received a big lump of revenue last year than the quality of its business.

Cephalon may be a value trap, shares that look cheap statistically but something about them not evident in the data will prevent the value from being realised. They may face technological obsolescence, like newspaper companies and book and DVD retailers like HMV, have a controlling shareholder squeezing out minority shareholders, like OPD, or large pension liabilities like Aga.

Checking for value traps, is almost as far as my analysis goes. It means re-calculating, augmenting and interpreting the data for each company, compiling a spreadsheet and ticking off a checklist as I go.

I *might* pay a *little* more for a company which seems likely to grow, though even that interpretation is rooted in a statistic: the company’s profitability (it’s return on assets and/or return on equity).

Real analysts estimate profit growth, but I don’t. I’m sceptical because:

1. It takes industry expertise and a lot of time
2. It’s easy to fit the assumptions you must make to the forecasts you want, and…
3. Maybe because of the first two, professional forecasters are generally bad at it.

For a year now I’ve occasionally gazed at a large textbook at the bottom of a pile of books to read and review on my desk (it’s at the bottom because it’s the biggest). It’s called Financial Statement Analysis and Security Valuation by Stephen Penman, a professor at Columbia University, and has the best first chapter I’ve read in any book on investment. Much of the rest is devoted to valuing growth.

Then I remember Benjamin Graham’s admonition that forecasting growth is just speculation and I wonder whether we really have learned anything since then.

-

When failure is the only option

Hedge fund manager Victor Niederhoffer on failure:

But also, most people have, in one way or another, a stop loss. If they go to Vegas with $10,000, they say I’m not going to spend more than $5,000. But they never say, "Hey, when I win a certain amount, that’s when I’m going to quit." I’d had this incredible string of successes where I made 50, 100 percent, year after year [but]… I didn’t have a stop-gain, if you will.

A study in Current Biology, suggests Jason Zweig, shows investors don’t just find safety going along with crowds. It gives them pleasure.

Computer models have three inherent problems, says Joseph Fuller, the modellers (who don’t understand finance), the managers of the modellers (who don’t understand the models), and the models (which don’t understand each other):

Each model executes its own strategy based on its calculus for maximizing value in a given market. But individual models are not able to take into account the role other models play in driving the markets. As a result, each program reacts almost in real time to the actions of other programs, potentially compounding volatility and leading to wild market swings.

Geoff Gannon recommends investors check a company’s vital signs first. They are its: Z-Score, F_Score, 10 year free cash flow margin of coefficient of variation, and 10 year real cash flow yield

Research from Arvid Hoffmann, Hersh Shefrin and Joost Pennings summarised on the CXO blog concludes that investors who say their motives are speculative earn lower returns than investors saving for retirement, investors who say they use technical analysis earn lower returns than investors using fundamental analysis.