by Alison L. Booth, Marco Francesconi, and Gylfi Zoega
by Thorvaldur Gylfason, University of Iceland and CEPR.
Which came first: oligopsony in labor markets, or oligopoly? In Europe, the historical record seems clear: oligopsonistic employers came first, and for centuries had the upper hand in labor markets, only belatedly to be confronted by trade unions that were formed as workers who felt they had been wronged by their employers turned their backs together in an attempt to undo the injustice and to assert their rights. As the 20th century progressed, however, labor market power began to shift in the trade unions’ favor until, in the early 1980s, the British government felt that the pendulum had swung too far and that, therefore, it was necessary to restrict union power by law in order to contain excessive wage increases, make labor markets more flexible, and thereby reduce unemployment. Today, many observers believe that France, Germany, and other continental European countries would benefit from following the British precedent in order to get their unemployment rates down to British, or at least more acceptable, levels. For a long time, the debate of the important role of trade unions in labor markets distracted attention from the original phenomenon that set the unions in motion, namely, oligopsony in labor markets (Manning, 2003). Even so, the problem never went away. To this day, many communities around Europe are company towns with one, two, or perhaps three factories that offer the only jobs to be had or thereabouts and use their market power to keep wages low and restrict the number of jobs on offer knowing that the workers have nowhere else to turn. In places where labor mobility is restricted, so that the workers are stuck at home, oligopsony among employers imparts a potentially significant bias to labor supply and human capital investment decisions. However, with increased labor mobility, the importance of this type of labor market imperfection seems likely to fade with the passage of time. Faced by oligopsonistic employers, some potential workers and parents may decide to forego education and work and to stay at home with their children, for example, as suggested by Booth and Coles (2004). Hence, in an oligopsonistic labor market, all three are perhaps best viewed as being jointly determined: work and pay, education attainment, and fertility. Typically, under oligopsony, too little is produced, and valuable education opportunities are lost to the parents, while their children may actually gain if the parents are able to provide them with better care and training at home than would be available outside the home. Empirical studies suggest that social gains from education exceed private gains especially during the first two years of life (Blinder, 1991; Heckman, 1999). Empirical evidence also suggests a strong negative correlation between education and fertility within and across countries.
This paper approaches oligopsony in labor markets from a different but no less interesting angle. Here the main emphasis is laid on oligopsonistic employers who offer too few jobs, too low wages, and also too little training compared with a perfectly competitive labor market. Using an elegant game-theoretic framework, the authors suggest that history repeats itself in that well-designed minimum wage legislation, training subsidies to firms, and welfare-maximizing oligopolistic industry-wide unions that care about training as well as work and pay can restore training intensity – that is, training per worker – to the optimal, competitive level, even if union activity tends to reduce the level of employment – that is, the number of workers with jobs – as earlier literature on oligopolistic trade union behavior suggests. The results for firm-level unions are less tidy, but they still suggest a role for unions in lifting the average level of training under oligopsonistic labor market conditions. Interestingly, the authors note an apparent difference between the results of empirical studies of American and European labor markets indicating that union workers in America are likely to receive less work-related training than nonunion workers whereas in Europe union workers are likely to receive more training than their nonunion colleagues. The authors present a simple cross-country regression of the percentage of the work force aged 25-54 in training in Europe in the 1990s against two different measures of trade union activity: (a) union coverage that reflects the extent to which labor contracts follow the precedents set in union contracts regardless of the degree of unionization, that is, of union membership, and (b) union coordination that reflects the degree of coordination between unions and employers at the industry level. The regression results suggest that union coverage – which is perhaps the most representative single measure of union influence available – may significantly reduce training from one European country to another whereas union coordination has a positive effect on training, but the latter effect is not significant; after all, the regression covers only twelve countries, so not much of value can be inferred from the exercise as it stands. More data and more detailed empirical work are needed to settle the issue. Even so, this is good and useful and well-made paper that throws new and stimulating light on some of the good things that well-managed trade unions can be expected to accomplish in imperfect labor markets, things that have tended to be overlooked or underrated in some of the recent literature on oligopolistic trade union behavior.
Blinder, Alan S. (1991), Growing Together: An Alternative Economic Strategy for the 1990s, Whittle, 1991.
Booth, Alison L. and Melvin Coles (2004), “Part-time Employment Traps and Childcare Policy,“ CEPR Discussion Paper No. 4357, April.
Heckman, James (1999), “Policies to Foster Human Capital,” NBER Working Paper No. W7288, August.
Manning, Alan (2003), Monopsony in Motion, Princeton University Press.