Substitutability of Capital-Labour in the Presence of Unions in the US Postal Services Industry

This paper studies the effects of labour unions on elasticity of substitution between production inputs in the US Postal Service industry. The study uses data from the National Income Product Account (NIPA) and the Current Population Survey (CPS). The paper estimates elasticity substitution between production labour, nonproduction labour, and capital using CES production functions and translog cost function. The CES production function is estimated by using seemingly unrelated nonlinear least square regression of inverse labour demand on the union density and input prices. In the translog cost function, seemingly unrelated linear regression is employed to run regression of cost share of operation labour and cost share of non-operation labour on union density and input prices. Further, the paper uses an instrumental variable of union density in the federal sector to solve the endogeneity of the union density. The analysis indicates that the labour union did not reduce elasticity substitution between production inputs. However, the union was able to maintain inelastic substitution between production labour and capital and between non-production labour and capital. The results of this study suggest that the strength of production labour and nonproduction labour to remain inelastic to substitute with capital was due to low competitive industry.