One of my first economics lessons contrasted perfect competition, which was judged to be a good thing, with monopoly, which was not. There are worse places to begin than by being shown the difference between championing the miracle of the free market and favouring the depredations of dominant businesses.
But monopoly power has often seemed like yesterday's issue. Standard Oil was broken up in 1911; AT&T in 1984. To the extent that we economists worried about companies being too big, we were thinking about the systemic risks from banks that were too big to fail. But we are starting to notice again the risks not of corporate failure but of corporate success.
The most obvious examples are the big digital players: Google dominates search; Facebook is the Goliath of social media; Amazon rules online retail. But, as documented in a new working paper by five economists, American business is in general becoming more concentrated.
David Autor and his colleagues looked at 676 industries in the US - from cigarettes to greeting cards, musical instruments to payday lenders. They found that for the typical industry in each of six sectors - manufacturing, retail, finance, services, wholesale and utilities/transportation - the biggest companies are producing a larger share of output.
For example, in the early 1980s the largest four players in any given US manufacturing industry averaged 38 per cent of sales; three decades later the figure was 43 per cent. In utilities and transportation the typical market share of the biggest four companies rose from 29 per cent to 37 per cent. In retail, overshadowed by Walmart and Amazon, the rise was dramatic: 14 per cent to 30 per cent.
This is surprising. As the world economy grows, one might expect markets to become more like the perfectly competitive textbook model, not less. Deregulation should allow more competition; globalisation should expose established players to pressure from overseas; transparent prices should make it harder for fat cats to maintain their position. Why hasn't competition chipped away at the market position of the leading companies? The simplest explanation: they are very good at what they do. Competition isn't a threat to them. It's an opportunity.
What Professor Autor and his colleagues call "superstar firms" tend to be more efficient. They sell more at a lower cost, so they enjoy a larger profit margin. Google is the purest example: its search algorithm won market share on merit. Alternatives are easily available, but most people do not use them. But the pattern holds more broadly: superstar firms have grown not by avoiding competitors but by defeating them.
This is not entirely bad news. But it's not entirely good news, either. The superstar firm phenomenon is the best explanation we have of a little-noticed but worrisome trend: since 1980, in the US and many other advanced economies, workers have been getting a steadily smaller slice of the economic pie (the distribution of this labour income also became much more unequal during the 1980s and 1990s).
Workers, from shelf-stackers to chief executives, have seen their total share of economic value-added fall from about 66 per cent to about 60 per cent in the US since 1980. This decline in "labour share" is often blamed on international trade making life harder for workers and easier for footloose capital. Prof Autor and his colleagues find little evidence for this idea.
Superstar firms, instead, seem to be the cause. The story is simple. These businesses are highly productive and achieve more with less. Because of this profitability, more of the value added by the company flows to shareholders and less to workers. And what happens in these groups will tend to be reflected in the economy as a whole, because superstar firms have an increasingly important role.
All this poses a headache for policymakers - assuming policymakers can pay attention to the issue for a long enough. The policy response required is subtle: after all, the growth of innovative, productive companies is welcome. It's the unintended consequences of that growth that pose problems.
Those consequences are not easy to predict, but here are two possibilities. Either the US economy ends up like Amazon, or it ends up like Microsoft. The Amazon future is one of relentless competition, a paradise for consumers but a nightmare for workers, and with the ever-present risk that dominant businesses will snuff out competition as the mood takes them.
The Microsoft future epitomises the economist John Hicks's quip: "the best of all monopoly profits is a quiet life". Microsoft in the 1990s became famous as a once-brilliant company that decided to pull up the drawbridge, locking in consumers and locking out competitors.
In either scenario ordinary people lose out, unless they can enjoy returns from capital as well as returns from working. In the very long run a superstar economy could become a technological utopia, where nobody needs to work for a living. That would require quite a realignment in our economic system; I wouldn't bet on such an outcome happening by chance.