An International Perspective on Labor.

STANFORD, Calif.--(BUSINESS WIRE)--July 23, 1999--

Once upon a time, unions were powerful players in many economies.

They could affect price levels, employment, even legislation. But that was when the world was a smaller place and the industrial engines of most countries were located inside their own borders. In a broad paper examining the economic impact of collective bargaining around the world, Stanford University Graduate School of Business labor economist Robert Flanagan explains why the kind of bargaining structure a country has is less important than it used to be.

Flanagan, who is the Konosuke Matsushita Professor of International Labor Economics and Policy Analysis, traces the economic thinking about unions over the last 30 years. Up until the mid-1970s, economists had a general sense that unions affected macroeconomic performance and they knew that union workers earned more than nonunion employees.

They found that changes in union power could affect wage levels once and for all, but would not be an ongoing source of inflation. After all, if you keep trying to raise wages, higher costs passed on through consumer prices will squash demand and reduce employment. But once the oil shocks hit in the 1970s, economists began to analyze which countries bounced back faster and what types of collective bargaining systems they had. Until then, no one had given much thought to the effects of different institutional arrangements on macroeconomic performance. A union was a union.

But in fact, the shape of collective bargaining varied widely around the world. In the United States and Japan, negotiations occurred at the corporate, or even plant, level. In Europe, bargaining took place at the industry level, as if an association of autoworkers negotiated with all the car makers at once. In Scandinavia, bargaining was even more centralized. A group similar to an American Federation of Labor negotiated with the equivalent of a national chamber of commerce to strike a wage deal for the entire economy.

To explain the effects of different forms of collective bargaining, three hypotheses emerged in the 1980s. One "corporatist" view suggested certain combinations of government and labor market institutions produce the most wage restraint.

A second hypothesis proposed that highly centralized bargaining created the most wage restraint because centralized unions must recognize that the benefits of excessive wage increases may be cancelled by the price increases they provoke.

A third proposal held that both very centralized and very decentralized setups produced wage restraint, but industry-based bargaining structures, such as those in Europe, resulted in the highest wages and would ultimately create unemployment. This view stressed that union power is relatively weak in decentralized bargaining. Yet empirical support for these hypotheses proved fragile. "Relationships that may have existed in the late 1970s or early 1980s seem to have disappeared by the 1990s," says Flanagan.

So what's going on? Some studies suffer from significant methodological flaws and, Flanagan argues, many studies have ignored significant changes that undermine the early theories. The most important of these is the increase in international trade. "Opening economies to the rest of the world means employers can't simply pass on wage costs because they'll price themselves out of international markets," says Flanagan. Once the economy is open, consumers can go to foreign producers, which weakens the unions.

A second development in recent years is the growth of the nonunion sector, which undermines unions and makes the type of bargaining structure a country uses less relevant. Union decline has been most dramatic in the United States and the United Kingdom but is also evident in Japan, New Zealand, and elsewhere. Driving this trend are concerns over the usefulness of unions as well as legal changes. For example, U.S. unions fought hard for federal laws that protect all workers. Ironically, that success diminished the need for unions.

Bargaining structures have also been slowly decentralizing. In Scandinavia, the move has been from centralized to industry-level bargaining or lower. In the United Kingdom and New Zealand, bargaining has moved to the enterprise level. While Germany nominally has industry-level bargaining, fewer employers are participating in official bargaining arrangements.

By the 1990s unemployment and real wage levels no longer seemed to depend on a country's bargaining structure. "You can't say centralized bargaining will always produce superior economic outcomes," says Flanagan. "The effect of specific bargaining institutions will depend on how open the economy is to trade, the extent to which it's a union or nonunion economy, and the nature of certain government institutions."