Is Industrial Policy Worth the Effort?

The Letta and Draghi reports helped rekindle the European debate over the merits of industrial policy, including the promotion of national champions. Some, like Harvard’s Dani Rodrik, see this type of policy as a way to enable European firms to better compete internationally. Others, like former EC Commissioner Mario Monti, warn that industrial policy could face pushback from other countries and weaken competition enforcers’ power.

A recent study attempts to evaluate the value of industrial policy by taking the “economics textbook case” approach, namely the case when some sectors are subject to external economies of scale. When that is the case, the private marginal cost is lower than the social one. In other words, society’s benefits from the expansion of some sectors go beyond the benefits to firm owners in those sectors.

Methodologically, the paper’s main contribution is to develop a general equilibrium model of the economy; estimate key model parameters (supply and demand elasticities); and finally derive the optimal unilateral industrial policy (thus making the best case for industrial policy).

The authors’ estimates of sector-specific scale economies are significant, both statistically and economically speaking: the average scale elasticity is about 0.21. There is widespread heterogeneity, with estimates ranging from 0.11 in the coke/petroleum and basic metals sectors to 0.31 in the paper products sector.

Notwithstanding the finding of “substantial economies of scale,” the authors conclude that the “gains from industrial policy […] are hardly transformative, even among the most open economies.”

The table below, copied from the paper, summarizes the main numerical results. The authors derive the optimal industrial policy (subsidies to sectors with greater scale economies) as well as the optimal trade policy (optimal tariffs) from a unilateral point of view (i.e., ignoring the effects on other countries and their possible reaction). The first column considers the joint implementation of the optimal industrial and trade policies.

These numbers correspond to long-term gains of at most about 4% of GDP, with an average of about 1% across countries. In the authors’ own words, these are “hardly transformative.” There may exist, of course, other reasons why industrial policy works, but the “textbook case” based on scale economies does not seem to be the one.

Dominick Bartelme, Arnaud Costinot, Dave Donaldson, and Andrés Rodríguez-Clare

The Textbook Case for Industrial Policy: Theory Meets Data

Journal of Political Economy 133, number 5, May 2025