Do Unions Impact Efficiency?: Evidence from the U.S. Manufacturing Sector

Article excerpt

I. INTRODUCTION

The relationship between labor unions and firm performance has garnered a vast amount of attention from scholars, unions, and businesses, as well as from policymakers as extensive studies have explored both the theoretical foundations and the empirical evidence regarding the impact of labor unions on productivity, as well as on other aspects of business, such as sales, profitability, investment, and employment growth (Doucouliagos and Laroche 2003). Perhaps one of the most relevant public policy questions is how unions affect productivity. Although studies have empirically investigated union productivity effects for a variety of industries and public sectors, particularly in the U.S. and European countries, empirical evidence on this question, however, is mixed.

Brown and Medoff (1978) and Freeman and Medoff (1984), in their early and influential work, argued that labor unions can raise productivity by inducing managers to adopt more efficient production methods and policies, and by providing better communication channels between workers and management. Supporting this impression, Dworkin and Ahlburg (1985) argued that opening communication channels between management and workers can result in integrative, rather than distributive, bargaining. Allen (1984) used the example of the construction industry, and cited union hiring halls, apprenticeship programs, and managerial shock effects as the reasons for union productivity improvements. In recent years, many employers have adopted new ways of organizing work that emphasize employee involvement. Empirical evidence suggests that various human resources management practices have a positive impact on firms' productivity. Black and Lynch (2001) found that unionized plants with these new approaches to organizing work had higher productivity than other similar non-union plants. Recent studies indicate that improved productivity of firms with labor unions may reflect better human resources management practices (Bloom and Van Reenen 2007; Doucouliagos and Laroche 2006; Ichniowski and Shaw 2003; Machin and Wood 2005).

However, Lewis (1963) pointed out that one of the most well-established impacts of labor unions on firms is the ability to raise wages above competitive levels. Such a monopoly power may exert adverse impacts on firm productivity by distorting the labor market. Other unfavorable effects of unions are on R&D spending and investments, since by acting as a tax on the return on investment, unions limit innovative and investment activities (Connolly, Hirsch, and Hirschey 1986; Hirsch and Link 1987; Link and Siegel 2002; Lommerud, Meland, and Straume 2006; Menezes-Filho and Van Reenen 2003). These effects can have a detrimental impact on the dynamic productivity of firms. Thus, unions can both enhance and detract from firms' productivity performance. Because of the ambiguity over the net effect of unions, an assessment of the union's effects on economic performance hinges on empirical evidence (Doucouliagos and Laroche 2003). (1)

Although most literature focuses on the performance or productivity impacts of unions, a strand of literature also looks at the channels through which unions affect measured performance/productivity. Measured productivity can be decomposed into two important components: the adopted production technologies and the efficiency of utilizing the technology. The former represents a shift in the production function and the latter shows the distance between actual production level and frontier production level for given inputs. An increase in the measured productivity may result if a firm adopts a better production technology or improves "technical efficiency" by moving production toward the frontier level of a given technology.

The arguments for or against unions regarding productivity can trace their causes to the impacts on one or both of the productivity components. For instance, the communication effect of Brown and Medoff (1978) and Freeman and Medoff (1984) results in improved efficiency in using installed technology, whereas the outcome of changing production methods, also mentioned by the same authors, amounts to choosing new production technologies. …