Local labor market concentration and union activity in the Philippines: descriptive estimates and implications

Abstract

Workers’ voice institutions traditionally counterbalance the persistent wage- and term-setting power of firms in labor markets, but can only do so if they exist. Do we observe less union activity in labor markets where employment is more concentrated? We address this question and estimate unionization and bargaining activity across broadly aggregated industry-region labor markets using two register data sources from the Philippines: an establishment-level census and a union and collective bargaining registration database. Our results suggest that local labor market concentration is robustly and negatively associated with new union registrations and new collective agreements. A one-standard-deviation increase in the employment Herfindahl- Hirschman Index (HHI) and the four-firm concentration ratio (CR4) is associated with a 12 percentage-point (pp) and 10.4 pp lower probability of having a new union registration, and a 9.7 pp and 15.3 pp lower probability of having a new collective agreement, on average, respectively. Further, concentration is also associated with lower union membership densities. These results highlight how employer power in input labor markets is plausibly prohibitive for collective action and may deter organizing and bargaining in contexts where only a handful of firms dominate employment. For antitrust authorities in developing country contexts with low union salience and decentralized bargaining systems, a closer examination of the repercussions of concentrated labor markets in merger reviews, competition enforcement, and policy research is a promising way forward.

JEL classification: J42, J51, J52