Morelli investigates the impact of burocracy on the costs of populism

Populist leaders often campaign against bureaucratic constraints, yet their economic performance may depend on the very institutional environment they inherit. In the paper “Do Not Drain The Swamp! Populism, Bureaucracy and Economic Performance”, Massimo Morelli (Bocconi University and Baffi Centre); Dmitrii Petrukhin (University of Cambridge, Faculty of Economics, Students) and Matia Vannoni (King’s College London) examine whether bureaucracy moderates the economic costs of populism.

The methodology

The study uses two complementary settings: crosscountry synthetic-control evidence and a U.S. state-year panel spanning 1929-2023. The authors consider a balanced core sample of 28 populist episodes (drawn from their broader database of 51 populist national leaders) from 1900 to 2020. They add to this dataset information on the independence and capacity of bureaucracy from V-Dem Institute. 

Second, they test the effect of bureaucracy on the link between populism and economic performance in the context of the U.S. states, because of their similarity in terms of political systems, bureaucracies and economies. The U.S. state analysis relies on a state–year panel (1929–2023) constructed from State of the State speeches and administrative data. The authors assemble a corpus of 3,355 gubernatorial speeches and measure populism with LLMs and the holistic-grading prompt of Tamaki et al. (2025). They complement this with data on party affiliation, civil-service reforms, and per-capita income.

The findings

The crosscountry analysis of the paper offers suggestive evidence that the output decline following populist takeovers is attenuated in countries with stronger pre-existing bureaucratic institutions, particularly where bureaucracies are more independent. The U.S. state analysis provides sharper, within-country evidence. Leveraging LLM estimates of gubernatorial populism derived from State of the State speeches, the authors find that populist governors are associated with significantly lower per-capita income in states lacking an independent civil service, while no comparable penalty emerges-and the association is more than offset-where such protections are in force.

Heterogeneity-robust event-study estimates point in the same direction, with the post-entry income decline concentrated in states without civil-service reform. Together, the evidence is consistent with independent bureaucracies acting as a partial buffer against the economic costs of populist governance: the "swamp" appears to dampen the damage populism would otherwise cause.