Union Density Changes and Union Effects on Firm Performance in Peru

The paper documents the sharp reduction in union density in Peru between 1986 and 1998, in a context of large macroeconomic fluctuations, structural reforms and changes in the Collective Bargaining Law in 1993. Using a pseudo panel of household surveys we find which worker and firm characteristics increase the likelihood of unionization in a context in which union density falls sharply. We find that a blue-collar job, a permanent contract, higher education and working in a large firm increase the likelihood of unionization, but only before the legislative change. There is evidence of a breakpoint in the reduction of union density probably related to the 1993 change in the Collective Bargaining Law. Most of the reduction in union density can be explained by within- categories decreases in union density. However, there is a small contribution stemming from the reduction in employment in the highly unionized public sector and from the increase in employment in low union density temporary and small firm employment. Using a panel of firms for the manufacturing sector for the period 1994-1996, we find a negative impact of unions on profits for all firm sizes. Within unionized firms profits are lower the higher the union density within the firm. In the econometric analysis, we find a significant negative effect even after controlling for firm and sector characteristics and firm fixed effects. There is some evidence that this effect diminishes over time, consistent with the reduction in union density during that period, but the reduction is not robust to different specifications. This impact of unions on profits is also negative and significant when we use within-firm union density. Labor productivity is negatively related to having a union in the firm, but the negative effect disappears once we control for firm characteristics.