Join our community of smart investors

Easy pickings?

The rise of 'superstar' companies might mean that stock picking has become easier
October 31, 2018

Has it become easier for stockpickers to beat the market? For me, this is a question posed by the emergence of so-called 'superstar' companies.

To see why it’s a question, you need to know my prior beliefs. These were heavily shaped by two pieces of research. One, published in 2001, was by  Louis Chan, Jason Karceski and Josef Lakonishok. They concluded that “there is no persistence in long-term earnings growth beyond chance, and there is low predictability even with a wide variety of predictor variables”. The other was a study of corporate growth around the world by Alex Coad, which found it to be a “fundamentally random process”.

For me, this research suggested that picking stocks the conventional way is a waste of time, because we cannot foresee future earnings growth.

Recently, though, I’ve found reasons to weaken this prior. MIT’s John van Reenen and colleagues show that recent years have seen a rise in “superstar companies” – those that can use past success and efficiencies to achieve further growth.

Take Netflix, for example. It has enjoyed a virtuous circle: increasing revenues generate the cash to invest in good programming, which attracts more customers and so on.

Or consider Wal-Mart or Amazon. They have used their size to invest in IT to manage supply chains better, thus cutting the costs of transporting goods and holding inventories. This has enabled them to cut prices and offer greater product variety, thus growing at the expense of smaller, weaker rivals. These are merely the most eye-catching examples of what other companies have done: Boston University’s James Bessen has shown that a key determinant of corporate success is the ability to use IT systems well.

These are not the only examples of how size and success lead to further successes. Facebook is popular simply because so many people are on it, and this attracts advertising revenues. This is an example of a form of Metcalfe’s law: the value of a network increases with the square of the number of users. The same applies to many platform businesses such as Uber, Purplebricks or eBay.

Yet other big companies can benefit from herding effects. Much of the success of Apple, for example, comes from people wanting iPhones and iPads because others have them.

Mechanisms such as these, says Professor van Reenen, are causing not just increased concentration and monopoly power in many industries, but more persistence in this concentration: dominant companies in the recent past are still dominant now.

All this suggests life might be getting easier for stockpickers. Instead of corporate growth being random, maybe it is becoming predictable. We can buy today’s leading companies and see them use their past success to generate future growth. Paul Romer recently won a Nobel prize for contributing to endogenous growth theory – the idea that past growth can lead to more growth. What’s true of national economies, however, might be also be true now for individual companies.

Evidence for this doesn’t just come from the great long-term returns on so-called FAANG stocks: Facebook, Amazon, Apple, Netflix and Google/Alphabet. Some of Warren Buffett’s success comes from investing in companies with “economic moats” – sources of monopoly power that can be used to generate further earnings growth. And Tilda Nguyen at the University of Hawaii has shown that investors have in the past underpriced companies with lots of intangible assets. They have neglected things such as brand power and good ideas that can cause future growth.

Now, in one sense there is nothing new about 'superstar companies'. Hendrik Bessembinder at Arizona State University has shown that all of the S&P 500’s outperformance of cash since 1926 is due to just 4 per cent of all the companies that have ever been listed. Of course, in any one year around half of all companies will beat the market. But over the longer run many of one year’s winners have been another year’s losers. Only a few superstars have persistent success.

If this success is predictable simply on the basis of past success, though, life is easier for stockpickers.

Or is it? If market dominance and network effects were all that mattered, we’d still be using Netscape Navigator to visit MySpace and Friends Reunited. The fact that we’re not tells us that other things matter for corporate success.

Some of these are diseconomies of scale. Big companies become harder to manage: Facebook’s trouble with managing users’ data is just the latest example of this. Worse still, success can lead to overconfidence and hubris and to catastrophic decisions, such as AOL’s to buy Time Warner in 2000.

Another problem is creative destruction. FAANG stocks are today the beneficiaries of this. But they need not remain so. Dominant companies in the past have faded as new technologies emerged: AOL declined as broadband eclipsed dial-up; Kodak as digital photography emerged; Nokia as smartphones were developed; Netscape fell as Google rose; and so on. Creative destruction isn’t a near-term threat to incumbents, but it could be a long-term one. Given that the pace and direction of technical change are hard to predict, we can’t be confident this threat is absent.

And then of course there is the danger that investors have more than wised up to this and are overpaying for superstars. Even the best companies can be overpriced if investors price in even faster growth than they can deliver. The old question 'what is in the price?' is as important as ever. And perhaps it makes stockpicking as difficult as ever.