Working Papers



Abstract: This paper studies the impact of e-commerce platforms on firms’ market power. I present a model with firm heterogeneity, oligopolistic competition, and Input-Output linkages, in which firms use the platforms to lower their variable costs. The cost reduction can occur either (i) as a result of efficiency gains or (ii) via direct discounts on input prices. The increasing use of platforms contributes to the rise of market power because it transforms the cost structure of the firms. Platform users show lower marginal costs but higher overhead costs: since the benefits of the marginal costs reduction outweigh the increase in overhead costs, this gives them a competitive advantage, explaining the increase in market power. Once calibrated to US data, the model attributes one third of the increase in markups to the introduction of e-commerce platforms. At the sectoral level, the heterogeneity in platforms use explains up to 40% of the differences in market power trends across sectors.



R&R at the Economic Journal

[Previously titled "Heterogeneous Increase in Sectoral Market Power: the role of ICT"] [Dec. 2022] [Dec. 2021] [May 2021] [Aug. 2020]


Abstract: During the last four decades, the U.S. industries have experienced heterogeneous increases in market power. This paper argues that this heterogeneity can be explained by the different dynamics of turbulence across sectors. To show it, we model a sector where firms differ by their productivity level and compete under oligopolistic competition. Business dynamism is captured in the model by sequential idiosyncratic entry, exit, and productivity shocks. A sector-specific increase in turbulence accelerates the turnover of leaders and the mobility of firms over the productivity distribution. This leads to reallocation of market shares towards the most productive firms that charge the lowest price. Their cost leadership allows them to charge the highest markups and gain the steepest profits, driving the increase in sectoral market power. The model can explain between 35% and 57% of the cumulative increase in the observed markups, and its predictions are supported by the U.S. and European data.



[First draft soon! Slides available upon request]