Facultad de Economía

Ineffective fundraising lowers the resources charities can use to provide goods. We combine a field experiment and a causal machine-learning approach to increase a charity's fundraising effectiveness. The approach optimally targets a fundraising instrument to individuals whose expected donations exceed solicitation costs. Our results demonstrate that machine-learning-based optimal targeting allows the charity to substantially increase donations net of fundraising costs relative to uniform benchmarks in which either everybody or no one receives the gift. To that end, it (a) should direct its fundraising efforts to a subset of past donors and (b) never address individuals who were previously asked but never donated. Further, we show that the benefits of machine-learning-based optimal targeting even materialize when the charity only exploits publicly available geospatial information or applies the estimated optimal targeting rule to later fundraising campaigns conducted in similar samples. We conclude that charities not engaging in optimal targeting waste significant resources.

This paper studies the effects of the minimum wage on the life cycle of firms. Using matched employer-employee and census data, I analyze the large increase in the federal minimum wage in Brazil between 1999 and 2010. Based on a difference-in-differences approach exploiting exposure of both firms and local labor markets to the minimum wage policy, I find that an increase in the minimum wage: (i) decreases the growth rates of small and young establishments, and (ii) increases the growth rates of old and large establishments. Moreover, establishment exit in affected regions increased following the increase of the minimum wage, although I found no effect on the entry of new firms. While traditional monopsony models cannot account for this result, a neoclassical monopsonistic model where firms face financial frictions can rationalize this finding.

Fiscal rules are a promising tool to address waste in public administration, but their effectiveness and political feasibility remain unclear, particularly in weakly institutionalized settings. We leverage exogenous variation across Colombian municipalities in exposure to a fiscal rule that limits the operating expenditures of local governments. Our difference-in-differences analysis yields three main findings. First, the fiscal rule is highly effective at reducing operating expenditures and the probability of a current deficit. Second, there is no meaningful impact on local public goods, which suggests wasteful administrative spending before the reform. Third, the fiscal consolidation does not cause a political backlash. Electoral support for the party of the incumbent mayor increases and the incidence of protests weakly decreases. These findings suggest that fiscal rules can align fiscal policy with the preferences of voters in settings, like Colombia, with weak political parties and limited career concerns for local politicians.

This paper explores the relation between agricultural shocks in illegal crops and schooling decisions. I focus on the case of Colombia where coca leaf fields were increasing at an average annual rate of 25 percent. Coca leaves are the main input to produce cocaine and Colombia is its main producer. The country's main strategy to eradicate coca crops was aerial spraying of herbicide until 2015, when the policy was suspended. I exploit the removal of the policy and use it as an instrument for the presence of coca fields. This temporal variation along with the cross-sectional variation of the spraying campaigns lead to an instrumental variable difference-in-differences. I use data on coca presence, eradication missions, and school outcomes at the municipal level from 2012 to 2018 to test whether a change in the presence of coca crops has an effect on schooling decisions. I show that my setting does not meet all the assumptions of the traditional difference-in-differences strategy but it fits those of Fuzzy Difference-in-Differences. My findings suggest that a one percent increase in the area cultivated with coca crops increases the high-school dropout rate by 0.3 standard deviations. This effect is equivalent to an increase of 0.9 percentage points, which represents an increase of almost 30% relative to the 3.15% mean high-school dropout rate. On the other hand, I do not find a significant effect on the primary- or middle-school dropout rate. Moreover, among the high-school students that decide to stay in school, there is a decrease in the share of students that fulfill the requirement for their grade in areas with coca presence. I also find some evidence that mathematics test scores in primary-school improve in such areas. Finally, I rule out the possibility that coca presence affects the area cultivated or harvested of any other crop in the municipality. Taken together, I interpret these results as suggestive evidence that high school-age individuals are leaving school to work on coca related activities.

Can the way students navigate admissions processes contribute to gender gaps in college major choices? We study the major choices of applicants to Colombia's largest public university who marginally miss the cutoff for their intended major. We find that women scoring just below the cutoff submit additional less-preferred majors and enroll sooner compared to similar men who intend the same major. Men instead reapply later for admission. As a result, a gender gap in potential salaries of 6.8-9.5% emerges among marginal applicants. Our findings show that differential approaches to admissions processes generate gender gaps in enrolled majors and earnings’ potential.

How important are liability dollarization in the transmission of commodity shocks on business cycles? To address this question, we developed a small open economy DSGE model with a banking sector and financial friction. The banks collect funds in the international capital markets in the form of foreign debt, which is denominated in dollars. Faced with a commodity shock, the results suggest that the presence of liability dollarization generates a greater response output, compared to the model that ignores this financial mechanism. In fact, after simulating a series of shocks in each period, bank balance sheets amplify expansions and deepen economic recessions. Thus, the volatility of business cycles is strongly affected by the presence of liability dollarization. However, the application of countercyclical macroprudential policies, based on Basel III capital requirements, absorbs the impacts of commodity shocks. In addition, this type of policy allows greater gains in social welfare. In summary, it can be pointed out that the bank balance sheets are quantitatively important because they provide information to understand macroeconomic fluctuations.

Much of the recent work on gender economics explains gender gaps pointing to differences in preferences, societal gender norms and prescriptive stereotypes about what women and men should be good or bad at and what they should like or dislike. But what happens to the individuals that do not behave/choose like the gender-typical person? We address this question using longitudinal data that allows us to measure gender conformity during childhood—using a seemingly irrelevant questionnaire—and tracks individuals through their life. We find that female nonconformers have substantially better school careers, labor market outcomes and are more likely to end up in male dominated occupations as compared to their gender conforming counterparts. We find the opposite pattern among male gender nonconformers; they perform substantially worse at school, earn less as adults and are more likely to be unemployed relative to their gender conforming male peers. We also see a positive correlation between gender nonconforming boys and mental health and addiction issues later in life. Furthermore, we find that gender nonconforming girls tend to delay childbearing in adulthood and face higher incidence of divorce.

A novel method for the joint estimation and identification of conventional and anticipated monetary policy shocks is introduced. The presence of an anticipated shock, such as forward guidance, renders the model dynamics noninvertible and invalidates the traditional SVAR based methods. To address this issue, invertibility-robust identification methods have gained popularity recently. Contrary to the current standard practice, we highlight that the proposed identification strategy based on noninvertible structural VARMA models does not require external information, such as instruments for the monetary policy shock. To distinguish between invertible and noninvertible model structures, non-Gaussianity of the error process is necessary, with this assumption solving also the standard structural identification problem. The main findings are: First, when purged from the anticipation component, the effects of a conventional monetary policy shock are less pronounced. Second, the anticipated monetary policy shock is shown to have large effects on the real activity in line with the current literature. These results lend support to the conclusion that the unconventional monetary policies are effective in stimulating the economy. Robustness to alternative data specifications and sample spans is checked with the main conclusions remaining unchanged.