Individuals and Organizations as Sources of State Effectiveness
NBER Working Paper No. 23350
Policymakers do not implement states’ policies—bureaucrats do. How important are bureaucrats in determining the productivity of the state enterprise? To what extent do the tradeoffs between different policies—and hence optimal policy design—depend on the effectiveness of the bureaucracy tasked with implementation? We investigate these questions in the context of public procurement. Using data on 16 million purchases in Russia during 2011–2016, we first show that over 40 percent of the variation in quality-adjusted prices paid—our measure of performance—is due to the individual bureaucrats and organizations that manage procurement processes. Our estimates imply that ineffective bureaucracies massively reduce public sector output: moving the least effective quartile of procurers to 75th percentile effectiveness would save the Russian government USD 13 billion each year. To explore the implications of bureaucratic effectiveness for policy design, we analyze Russia’s adoption of a ubiquitous procurement policy—bid preferences for domestic suppliers. Using a generalized difference-in-differences strategy, we estimate the impact of the policy separately for procurers of different effectiveness. Consistent with a simple endogenous-entry auction model, we find that bid preferences save the government 17.5 percent when implemented by the least effective quartile of bureaucrats, but only 0.7 percent when implemented by the most effective quartile. Overall, our results demonstrate that the often overlooked bureaucratic apparatus is critical for state effectiveness, which helps to explain why many policies work well only in some settings; and that policy designed with bureaucratic context in mind can partially offset the cost of bureaucratic ineffectiveness.
Document Object Identifier (DOI): 10.3386/w23350
Users who downloaded this paper also downloaded* these: