Job market paper
Quality of Governance and the Design of Public Procurement
The design of public procurement largely consists of choosing the extent of discretion allowed to procurement agencies. Discretion can improve the quality of projects but it can also be misused, leading to rent diversion or outright corruption. Using data from the Czech Republic, a novel measure of quality and quasi-experimental evidence, this study documents that procurement agencies use discretion only partially efficiently. Discretion inflates prices and this is not fully compensated by an increase in quality. I develop a structural model that explains what frictions inhibit the optimal use of discretion. Procurement agencies are corrupt, do not fully internalize quality and do not possess critical information about the quality of vendors. These frictions increase procurement spending by 17%. Switching to rule-based mechanisms is efficient in the Czech environment as it limits the misuse of discretion. Additional oversight is also beneficial as there is evidence that increased monitoring aligns the objectives of procurement agencies and taxpayers. Finally, this study demonstrates that a one-size-fits-all approach to procurement design is not optimal. Policies that enhance efficiency in the Czech setting would decrease efficiency in countries with better governance.
Award: Young Czech Economist of the Year 2020
Revisiting Cap-and-Trade in Presence of Publicly Owned Polluters: The Case of Italy 2006-2018
with Jakub Kastl and Federico Boffa
We use the example of the Italian electricity spot market to empirically document that carbon pricing schemes may not work efficiently when the major firms in the market are government-controlled. We show that government-controlled companies do not internalize emission prices implied by the European Union emissions trading system in their bids, which reduces pass-through of emission costs and decreases productive efficiency. A vast majority of electricity generators in the world are government owned and this is especially true for fossil fuel burning ones. We argue that, as a result, contrary to conventional wisdom among economists, carbon pricing is unlikely to be an efficient way to regulate and mitigate emissions in the electricity sector. Command-and-control approach, involving emission standards, might be more suitable, especially since reliable estimates of the production functions of electric generators are readily available.
Data Transparency, Public Oversight and Collusion in E-procurement
with Leon Musolff and Vitezslav Titl
In this paper, we study an e-procurement market in Ukraine. We develop a novel structural test to detect collusive behavior, document that bidding patterns in the data are incompatible with a competitive equilibrium, and identify pairs of colluding firms. We validate the soundness of our collusion detection algorithm on a large sample of prosecuted companies. In Ukraine, a broad policy reform created an unprecedented scale of data transparency. Data transparency, in turn, led numerous NGOs and educated volunteers to monitor the market for public procurement. We document that this new supervision is effective in reducing collusion and prices on the market; in particular, prices decrease by 20.6%. Finally, we estimate the deadweight loss and find a possible sizeable overall welfare gain from the additional oversight due to e-procurement of between 0.55% and 2.68% of the total procurement spending.
Grants: Weiss Foundation
The Cost of Favouritism in Public Procurement
with Vitezslav Titl
Are political connections in public procurement harmful or efficiency-gaining for the public sector and what are the costs of favoritism towards politically connected firms? Exploiting detailed data on firm representatives’ political affiliations in the Czech Republic, we find that favoritism to politically connected firms increases the price of procurement contracts by 8% of the estimated costs while no gains in terms of quality are generated. Interestingly, these adverse effects of political connections are not present for procurement contracts that get additional oversight from a higher level of the government, because they were co-funded by the European Union. Based on our estimates, the total procurement expenditures increased by 0.48% due to favoritism. Finally, we discuss and document channels of such favoritism and present suggestive evidence that politicians tailor technical specifications of projects to fit the comparative advantage of specific firms.
Awards: Young Czech Economist of the Year (runner-up in 2019), 3rd for the best paper (Wicksell prize) at EPCS Meeting (Rome in 2018), among the 3 best papers at Public Choice Meeting (2020).Download paper
Work in progress
EU Subsidies Shape Governmental Spending: Evidence from Spain
with Stefano Baratuche, Jeffrey Sun and Vitezslav Titl
The European Union spends over 50 Billion Euro each year to subsidize projects in poorer regions of Europe. A large portion of these resources is allocated to public entities. Using a change in eligibility for subsidies we show that Europen subsidies significantly alter the portfolio of projects purchased by the public sector. There is, however, only limited evidence that public entities now acquire projects that have high externalities, such as investments into green energies or R&D, which are among the main targets of the European policy. We develop a dynamic model of procurement investment and show that simpler policies such as a lump-sum subsidy for poorer regions enhance efficiency.
Renegotiations of Procurement Contractswith Vitezslav Titl
Renegotiations of procurement contracts are very common across the world. They constitute a significant share of procurement expenditure. It is, however, not clear whether allowing for ex-post renegotiations is efficient. Using data from the Czech Republic we show that expanding the scope of renegotiations led to lower initial costs. The initial saving is more than offset by creating a holdup of procurement agencies by private firms. Ex-post costs increase as a result of this policy.
Informing Politicians and Bureaucrats about Consequences of their Actionswith Michal Soltes and Vitezslav Titl
Burecreauts and politicians are primary agents in public procurement as they choose the allocation mechanism and control the critical steps of the procurement process. Previous literature discovers possible information frictions mitigating the optimal behavior of procurement agencies. We prepare a randomized controlled trial to test whether informing procurement agents about their actions‘ average quantitative consequences affects their beliefs and behavior. We estimate the impact of this information treatment on public expenditures.