Concrete Thinking About Development

Misallocation is increasingly recognized as an important factor in explaining productivity differences across countries. This paper uses new micro-data on key input prices in the construction sector and market structure at a global level to study distortions in construction sector inputs and their consequences. We document that (i) there is large dispersion in construction sector input prices and that cement prices are particularly high in the poorest countries; (ii) cement prices are highest in countries with few firms; (iii) cement plays a significant role in construction sector expenditures, particularly in the poorest countries. To understand the reasons for price differences in cement we estimate a model of oligopoly using both a demand-based instrument and exploiting geological variation in the dispersion of limestone deposits within countries. Our results suggest that lower levels of competition lead to significantly higher prices. Firm-level financial accounts data point toward substantial economic rents. We then embed the oligopoly structure into a dynamic general equilibrium network model to analyse the consequences of distortions on the wider economy. We find that due to cement's network position, distortions have large effects on steady-state output: for every dollar increase in cement profits, steady-state output falls by two. Finally, we find that common ownership in cement is an important source of distortions accounting for between 75% and 85% of the wedge due to markups in cement.