We estimate the causal real economic effects of a randomized anti-corruption crackdown on local governments in Brazil using rich micro-data on corruption and firms. After anti-corruption audits, municipalities experience an increase in the number of firms concentrated in sectors most dependent on government relationships and public procurement. Through the estimation of geographic spillovers and additional tests, we show that audits operate via both a direct detection effect as well as through indirect deterrence channels. Politically connected firms suffer after the audits. Our estimates indicate the anti-corruption program generates significant local multipliers which are consistent with the presence of a large corruption tax on government-dependent firms.

We study the inner workings of internal capital markets during the 2008–09 recession using a unique dataset of loans between business group firms in an emerging market. Intragroup loans increase quickly during the recession. Firms that are more central in the ownership network simultaneously increase lending and borrowing. Acting like simple intermediaries, central firms do not increase net lending. Our results imply that formal control rights are essential for intermediation in internal capital markets, particularly during distress. In line with previous results on winner-picking, receivers of intragroup loans are high-Q, financially constrained firms, which also perform significantly better than providers during the recession.

In all modern bureaucracies, politicians retain some discretion in public employment decisions, which may lead to frictions in the selection process if political connections substitute for individual competence. Relying on detailed matched employer-employee data on the universe of public employees in Brazil over 1997–2014, and on a regression discontinuity design in close electoral races, we establish three main findings. First, political connections are a key and quantitatively large determinant of employment in public organizations, for both bureaucrats and frontline providers. Second, patronage is an important mechanism behind this result. Third, political considerations lead to the selection of less competent individuals.

We study the protest behavior of teenagers linked to a student killed by a stray bullet coming from a policeman in Chile. We use administrative data to follow the schoolmates of the victim and those living near the shooting on hundreds of protest and nonprotest days. We find that police violence causes lower protest participation in street rallies but more adherence to test boycotts. These effects appear among schoolmates of the victim and not among students living near the killing. Negative educational consequences suffered by the schoolmates combined with previous results suggest that psychological mechanisms are a plausible explanation.

This paper provides micro evidence of labor mobility inside business groups. We show that worker flows between firms in the same group are stronger than with unaffiliated firms. Moreover, the reallocation of top workers between group firms is more sensitive to international shocks. Top workers that move within the group in response to shocks reach higher positions and earn higher wages. We find suggestive evidence that productivity increases when firms receive same-group top workers. Our results are consistent with the hypothesis that, in response to changing opportunities, joint ownership eases the redeployment of workers endowed with general management skills.

Policies based on prohibition and repression to fight the war on drugs have largely failed in a variety of contexts. However, incentive-based policies may also fail and have unintended negative consequences if policymakers do not properly anticipate behavioral reactions. This is an particularly important concern in the case of policies announced prior to their implementation. In this paper, we show that a naive and untimely policy announcement generated an unprecedented escalation in cocaine production in Colombia, offsetting almost 20 years and billions of dollars of U.S.-backed efforts to stop drug production and cartel action.

We study how the disclosure of corrupt practices affects the growth of firms involved in illegal interactions with the government using randomized audits of public procurement in Brazil. On average, firms exposed by the anti-corruption program grow larger after the audits, despite experiencing a decrease in procurement contracts. We manually collect new data on the details of thousands of corruption cases, through which we uncover a large heterogeneity in our firm-level effects depending on the degree of involvement in corruption. Using investment-, loan-, and worker- level data, we show that the average exposed firms adapt to the loss of government contracts by changing their investment strategy. They increase capital investment and borrow more to finance such investment, while there is no change in their internal organization. We provide qualitative support to our results by conducting new face-to-face surveys with business owners of government-dependent firms.