Individuals and Organizations as Sources of State Effectiveness, and Consequences for Policy Design∗

How much of the variation in state effectiveness is due to the individuals and organizations responsible for implementing policy? We investigate this question and its implications for policy design in the context of public procurement, using a text-based product classification method to measure bureaucratic output. We show that effective procurers lower bid preparation/submission costs, and that 60% of within-product purchase-price variation across 16 million purchases in Russia in 2011-2015 is due to the bureaucrats and organizations administering procurement processes. This has dramatic policy consequences. To illustrate these, we study a ubiquitous procurement policy: bid preferences for favored firms (here domestic manufacturers). The policy decreases overall entry and increases prices when procurers are effective, but has the opposite impact with ineffective procurers, as predicted by a simple endogenous-entry model of procurement. Our results imply that the state’s often overlooked bureaucratic tier is critical for effectiveness and the make-up of optimal policies. JEL codes: H11, O2 ∗[email protected], [email protected], [email protected] We thank John M. Abowd, Daron Acemoglu, Nick Bloom, Richard Blundell, Raj Chetty, Allan Collard-Wexler, Francesco Decarolis, Pascaline Dupas, Ben Faber, Fred Finan, Matthew Gentzkow, Josh Gottlieb, Caroline Hoxby, Amit Khandelwal, Brad Larsen, Torsten Persson, Andrea Prat, Imran Rasul, Jimmy Roberts, Jonah Rockoff, Orie Shelef, Andrei Shleifer, Eric Verhoogen, and Danny Yagan for helpful discussions; Andrei Yakovlev for institutional guidance; Georgiy Syunyaev and Andrey Druzhaev for outstanding research assistance; and seminar participants at U.C. Berkeley, University of British Columbia, Central European University, Columbia, University of Copenhagen, Duke, the Econometric Society, Higher School of Economics, McGill, University of Michigan, NBER Development Economics Program meeting, NEUDC, Northwestern, University of Oslo, Santa Clara, Stanford, University College London, and University of Toronto for comments. The study has been funded within the framework of the Basic Research Program at the National Research University Higher School of Economics (HSE) and by the Russian Academic Excellence Project “5-100”. Any remaining errors are ours alone.