(With Imran Rasul, Daniel Rogger and Martin Williams) (Submitted)
Abstract: Organizational culture is an important driver of organizational performance, but evidence on how to improve performance-oriented organizational cultures is scarce – especially in the public sector. We partnered with Ghana’s Civil Service to design a new innovation training module geared towards such culture change and deliver it on a randomized basis to mid- level bureaucrats in central government. The intervention was delivered at full scale by integrating it into the Civil Service’s standard training routine for one year. We find that the training improved organizational culture and performance 6-18 months post-training. Our design was split between an individual-focused training arm and one in which officials from an entire organizational unit were trained together. Our results are completely driven by the individual-focused arm, with the team-based treatment arm having no impact on culture or performance. We discuss potential explanations for this difference in effectiveness. Simple and scalable training interventions can thus have significant impacts on culture and performance, but their design matters.
[NEW VERSION COMING SOON]
Abstract: Do political connections distort the allocation of public goods across regions? This paper answers this question in the context of partisan connections between Brazilian municipalities and national level ministries approving grants for cities. Ministries are occupied by different political parties over years, providing exogenous variation in connections within a given city, across ministries. Also, administrative data on the universe of grant requests and approvals between 2009-16 from all ministries allows for evidence for many types of public goods, and for direct and indirect effects of connections over grant approvals and requests. Reduced form evidence from a triple differences design within city, ministry and year shows that when a minister changes, cities with mayors in the same party of the new minister request 15% more money and get 14.8% more money approved from that ministry. Additional evidence suggests these are unlikely to be fully explained by ideological alignment, soft information or differences in enforcement. A structural model of grant requests and approvals estimates small ministerial biases towards connected cities and almost linear returns in project size. Due to the linear returns, connected cities request significantly larger grants. If political connections are the only source of distortion in the ministry’s decision function, the small bias implies small amounts of misallocation, and the welfare losses due to political connections are of only 0.24% of the average ministry’s budget.
Coverage: VoxDev
Work in Progress
Economic Networks, Variable Markups and Intermediate Good Taxation: Theory and Evidence from India (with Arun Advani, Lucie Gadenne, Tushar Nandi)
Summary: A key principle of optimal taxation is the attainment of production efficiency, which often implies that under perfect competition, we should not tax intermediate input purchases (Diamond, Mirrlees (1971)). This has influenced 165 countries to adopt VAT systems. Yet, there is increasing evidence that firms are not in perfect competition, that they are charging rising markups, and that these are production efficiency wedges that can react to cost shocks such as taxes. With this in mind, this project evaluates alternative tax reforms for intermediate goods in economies with imperfect competition. To do so, we build a quantitative model of an endogenous network of firms under monopolistic competition, where firms have flexible and endogenous markups. We use this model to characterise (i) when do different tax reforms affect markups (a production efficiency wedge), (ii) what is their effect over consumer welfare, (iii) with this, discuss the optimality of these alternative tax reforms. We quantify this model then using data on transactions between firms in India, matched with survey of industries data.
Cash-in-hand, Liquidity Constraints and Sanitation Adoption: Theory and Evidence from India (with Britta Augsburg, Bet Caeyers, Bansi Malde)
Summary: A set of health, environmental and development policy priorities involve encouraging households to undertake lumpy investments - such as sanitation, cookstoves, household electrification, income generating assets, and so on. Yet, especially when facing liquidity constraints, poor households might have difficulties to undertake these investments. In such settings, a key question is how should we design effective policies to encourage household investments; in particular, what is the effect of cash-in-hand and credit over household uptake of these investments. We develop new methods to answer these questions. We start by developing methods to estimate household policy functions for discrete investments when household assets are endogenous and subject to many unobserved shocks (such as past temporary and permanent income shocks, past asset shocks, and so on). We further develop methods to use these policy functions to back out parameters from a dynamic structural model of household consumption and durable discrete investments with uncertainty and liquidity constraints. These tools are then applied to the context of sanitation investments. In particular, we use data on consumption, household cash-in-hand, sanitation investments, and temporary income shocks due to weather; collected from an RCT of microfinance for sanitation in India. We use these to directly evaluate the effect of cash-in-hand, versus policies on microfinance on sanitation investments; and use the structural model to discuss counterfactual policies.