The World’s Biggest Myth

It happens every time there’s a global credit squeeze, an international trade dispute, or an outsourcing outrage. Like clockwork, the two opposing forces of the enduring globalization debate go to battle on the world’s op-ed pages, in policy-shop salons, and sometimes in the streets. The pro-globalizers talk of leveling playing fields and freeing markets to spread wealth to every corner of the world. The anti-globalizers claim rich countries and corporations have set up the rules of the game in their favor, profiting from the hard work and low expectations of most of the world’s poor.

These two sides of the debate agree on almost nothing, with one exception: They both believe globalization leads to more market share for fewer players. In fact, they couldn’t be more wrong. This perspective can be traced back at least to Karl Marx, who wrote more than 100 years ago that "one capitalist always kills many… [leading to] the constantly diminishing number of the magnates of capital, who usurp and monopolize all advantages."

Today, Naomi Klein sees globalization as leading to the triumph of the bigger and blander: "Market-driven globalization doesn’t want diversity; quite the opposite." And pro-globalizers add fuel to the fire. As two marketing professors, Jagdish Sheth and Rajendra Sisodia, recently put it, "When a market… moves toward becoming truly global in scope… [that offers] some of the most powerful evidence for the Rule of Three" — the notion that a stable competitive market never has more than three significant competitors. It comes as no surprise, then, that many managers buy into this notion as well: In a recent online survey I conducted, 58 percent of the several hundred managers who responded agreed with the proposition that "globalization tends to make industries become more concentrated."

But this one "fact" that anti-globalizers and pro-globalizers seem to agree on turns out, on closer examination, not to be much of a fact at all. If globalization indeed placed profits and power in fewer and fewer hands, we would expect to see a broad number of industries dominated by a few firms that continually gobble up smaller competitors as they expand their market share. But the data on global production reveals the opposite. In the graph, data summarizing changes in the share of global production of the top five market leaders in a variety of industries are shown. The data span periods of a decade or longer and account for all mergers and acquisitions.

As the figure indicates, the data do not show much of a tendency at all toward increasing concentration among market leaders. Although the number of industries in the sample that experienced concentration increases is slightly greater than the number that experienced decreases, the average share of the market held by the five largest firms in each industry actually declined.

And these are data over a relatively short time frame. When one examines industries over several decades, the hypothesis of increased concentration turns out to be even weaker, because a number of the industries reported as experiencing (relatively modest) increases in concentration in the graph — specifically, automobiles, oil production, and aluminum smelting — actually experienced very large decreases in concentration over prior decades. In other words, during a long period of rapid globalization, several important sectors of the global business environment became more competitive, not less.

Consider, in more depth, the global automobile industry. Much-publicized mergers such as DaimlerChrysler (since undone, of course) and the disappearance of competitors like British Leyland seem to imply that the past 20 years or so have seen more cars being made by fewer companies. But, in fact, concentration in the auto industry has decreased more or less continuously since the 1920s, when Ford’s Model T alone accounted for half of the world’s automobile fleet. Likewise, intuitions about "globalization as Americanization" clash with the United States’ declining share of total auto production from 80 percent in the early 1950s to one sixth today, as well as with Toyota’s overtaking General Motors as the world’s largest automaker. Another common perception, that globalization means fewer options, is simply wrong. Look, for example, at the continuing explosion in the number of car models available. According to Ford’s former chief operating officer, Nick Scheele, there were 184 new models rolled out in 2002, up from 52 in 1992.

The other remarkable thing about the auto industry is the extent to which the false tide of increasing concentration has been taken for granted, even by those who should know better. At least since the 1980s, when Fiat’s chairman, Giovanni Agnelli, proposed that the industry would afford room for no more than six major players worldwide, most automakers have treated increasing concentration as axiomatic — and have predicated important business decisions on that idea. Thus, the DaimlerChrysler megamerger, which led to the loss of tens of billions of dollars, was justified by senior management’s appeals for a decrease in the total number of independent companies in the industry since the 1920s. This focus solely on the total number of companies ignores the most basic precept in industrial economics — that concentration measures should look not just at the total number of producers but at the ones that are large in relation to the market. In other words, Ford in the 1920s shouldn’t be viewed as on par with garage producers. And it isn’t just the auto industry where globalization has wrought more, not less, competition. The paper- and oil-production sectors also demonstrate decades-long trends that should calm paranoia about corporate concentration.

Globalization itself generally increases competition and choice instead of reducing it. In fact, given the differences that arise at national borders, many of the benefits of globalization are actually unlocked by the visible hand of corporations. More broadly, the false debate about the consequences of globalization is a reminder of the fact that much of the discussion occurs in a data void. There are real arguments to be had about globalization, particularly its consequences for inequality within countries. But the debate should not be cluttered, let alone inflamed, by ideologues spreading fears instead of facts.