Do the names Armstrong Whitworth, Bleachers Association or Brown & Co mean anything to you? They would have done 100 years ago, because these were among Britain's largest companies back then. They were the Unilever, Diageo and Capita of their time. But they are long forgotten.
This is typical. Howard Gospel and Martin Fiedler, two economic historians, have compiled a list of the UK's largest firms in 1907. Of the top 100 then, only three are independent stock market quoted firms today: WH Smith, GKN and Prudential.
The message here is that firms die. Recent official figures have quantified this. They show that last year 11.9 per cent of companies ceased trading - a death rate 18 times higher than that for men. Such high mortality is not merely due to the recession. Even in the best year recently, 9.4 per cent of firms died.
Such mortality rates imply that there's little chance of a company making it to old age. They suggest there is only a two in a thousand chance of a company surviving for 80 years, and that two-thirds of firms will be under 10 years old.
Research by Alex Coad at the Max Planck Institute shows that this prediction fits the facts quite well. He's found that corporate age is roughly exponentially distributed around the world. In Ireland, Italy or India, for example, there are many young firms and very few old ones.
There are, however, two deviations from this simple pattern. One is that young companies are more likely to die than older ones. Dr Coad estimates that, in the US since 1980, only 79 per cent of firms survived to see their first birthday, whilst 91 per cent of four-year-olds lived to see their fifth. Once a company has survived infancy, then, its chances of a long life improve.
This explains why the death rate for listed firms is less than the ONS's estimate for all firms. Before a firm arrives on the stock market, it has usually traded for several years and so overcome the high probability of an early death. Unfortunately, though, this does not guarantee their survival - as investors in Rok or Connaught will know. (And those early, pre-quoted years are often times when the firm earns its best returns, but this is another story).
What's more, found Dr Coad, extremely old firms are less rare than the exponential distribution predicts. Italian gunsmith Beretta, for example, has been going since 1526; even with a death rate of only two per cent a year, there's only a one in 18,000 chance of making it so far. However, says Dr Coad, such firms tend to be family-owned ones, for whom survival and tradition matter more than profit maximization. This suggests that the pursuit of financial efficiency on its own is not conducive to longevity. This might explain why companies don't live as long as other organizations; whereas we think a 100-year-old firm is remarkably stolid, a 100-year-old university or army regiment is as fleeting and insubstantial as this season's fashions.
Paul Ormerod of Volterra Consulting points to another quirk about corporate death rates. If you look at their pattern over time or space, he says, it looks a lot like the pattern for the extinction rate of species. There are long periods of a low death rate, interrupted by mass deaths. If we couple this to the fact that corporate survival rates seem to stop improving after the first few years, we have a disturbing implication - that mature firms cannot learn enough to allow them to survive, any more than species have been able to learn to avoid extinction.
But what is it that firms can't learn? Research by Boyan Jovanovic at New York University provides one answer. Firms' organizational capital - their ideas and technologies - has, he says, a specific vintage. Companies embody and develop the technologies that were new when they were new; Disney got into the film-making business when it was quite new in the 1920s, Microsoft got into the new software industry in the 1980s, and so on. And firms carry on in businesses they began in - you can't teach old dogs new tricks - so, for example, the Prudential has always sold insurance, WH Smith always stationary. Exceptions to this - such as Nokia which transformed itself from a cable-maker to a phone company are rare. And some firms that try such transformations get into a horrible mess: just remember what happened when GEC - another of 1907's giants - became Marconi.
This fixity means that firms are vulnerable to technical change. The classic example is Polaroid, which was swept away by the rise of digital photography. But every day, businesses lose out to cheaper, often newer, rivals or find their business model and products badly adapted to changing tastes and technologies.
Company deaths, therefore, are an inevitable result of competition and technical change. There's a reason why Joseph Schumpeter put the word "destruction" into "creative destruction."
This matters for investors in two senses.
First, it means that one common way of valuing firms - the Gordon growth model, the simplest version of the dividend discount model - does not apply. It tells us that the value of a firm today should be equal to next year's dividend, divided by the difference between the required return on equity and the expected long-term growth rate. So, if dividends are expected to grow at five per cent, and the required return on equity is eight per cent and next year's dividend is 10p, the share price should be 333p: 10p divided by (0.08 - 0.05).
This, however, overstates the company's value because it assumes it will live forever. Relaxing this assumption makes a big difference. If, in the above example, we assume the firm suddenly becomes worthless after 30 years, its price today should be just 190p - 43 per cent lower. And 30 years is the life expectancy of the median company if the death rate is just two per cent a year. Small probabilities of death, then, might make a big difference to valuations.
I don't say this to claim that most firms are in fact so grossly over-valued. Perhaps one reason why shares are cheap now - in terms of price-earnings ratios if not dividend yields - is precisely that Mr Market is implicitly pricing in death risk.
Secondly, this represents a case for loss averse investors - those who are more than averagely worried by the small chance of a big loss - to hold tracker funds rather than individual stocks. Most firms that existed 100 years ago have died. But the aggregate market has thrived. This tells us that the chances of a massive loss - 100 per cent or close thereto - are greater for individual stocks than for the overall market. This means that an investor wishing to avoid tail risk - the small chance of massive loss - should prefer the market portfolio to individual stocks.
MORE FROM CHRIS DILLOW...
Chris blogs at http://stumblingandmumbling.typepad.com