This paper presents the first comprehensive analysis of how firms respond to carbon taxation in emerging economies. Our evidence builds on exhaustive administrative data from South Africa, the 13th largest emitter worldwide. First, we establish stylized facts on the types of firms that are affected, how much revenue is generated from which sector, and which share of national emissions the tax can capture. Second, we study the dynamic impacts of the carbon tax on firm-level outcomes such as profits, sales,  capital, and labor inputs. We show that the design of the South African carbon tax leads to substantial heterogeneity across sectors in terms of how strongly firms are affected. Contrary to the concern that carbon tax may hinder economic growth, we measure no negative effects on firm performance on average. We show that firms respond to the carbon tax announcement and try to anticipate future production costs, but we do not observe an increase in R&D. We also show that firms may be importing more of the same machinery but not new ones.

This paper examines the electoral impact of unverified narratives. In 2006, a famous journalist published an article in a widely read Brazilian newsmagazine. He stated that, according to a Brazilian Federal Police investigation, members of the left-wing Labor Party committed bio-terrorism by deliberately spreading a fungus that devastated cocoa production in Northeast Brazil and displaced nearly 250,000 workers. Although never substantiated, and the investigation ending without any criminal guilty verdict, we estimate that exposure to the allegation reduced left-wing vote shares in subsequent presidential elections, with effects persisting for up to 12 years across four elections. We also show that the news release increased the vote share for right-wing parties in gubernatorial and mayoral elections. We argue that the narratives activated collective memories of the agricultural crisis, shaping long-term voting behavior. Our findings highlight how previously hidden information can revive traumatic historical experiences and alter electoral dynamics.

This paper examines the impact of an affirmative action policy implemented by a top university on students' post-college labor outcomes. The policy reserved 45% of all slots in each major for Black and mixed-race candidates from public high schools. The findings indicate that the policy did not harm students' likelihood of securing formal employment. However, affirmative action students earned wages that were 22% lower than those of non-affirmative action students, a gap that emerged in the first year after enrollment and persisted for up to 14 years post-enrollment. This suggests that the policy is insufficient to close the wage gap for disadvantaged students. The negative effects on wages were primarily driven by disadvantaged female students, as well as students in technology and social sciences majors. Moreover, network formation failed to improve labor market outcomes and instead reduced the probability of job attainment for disadvantaged female students. Lower academic performance and higher dropout rates among affirmative action students may help explain the observed wage disparities.

This paper studies the impacts of mobile money on tax revenues in developing countries. The main results suggest that once we account for the staggered characteristics of mobile money deployment between countries, it does not increase tax revenues, contrary to the existing evidence. Additionally, this paper has three other contributions. First, using Kenya as a case study, we show that taxing mobile money also did not increase tax revenues. Second, we show the effect of mobile money deployment on other macroeconomic indicators, such as mobile phone subscriptions, consumption, and inflation. Third, using firm-level data for 11 sub-Saharan countries, we show that mobile money usage is associated with lower sales levels. It reinforces that mobile money benefits small firms, particularly those in the informal sector, the most.

Abstract: Mozambique is highly vulnerable to climate change. It faces frequent cyclones, floods, and droughts while having limited revenue collection capacity and social protection programmes. This paper assesses the distributional effects of climate shocks on household consumption and explores adaptation strategies using consumption survey data from 2008 to 2022, combined with district-level climate data. We find that extreme rainfall and dryness shocks during the growing season significantly reduce household expenses for essential goods, with impacts most pronounced at the bottom to the middle of the consumption distribution. However, we found that the consumption of self-produced goods mitigates the expense losses. Our findings suggest that the main social protection programme in the country, PSSB, also helps mitigate these negative effects. We do not find evidence that domestic family transfer and international remittances cushion consumption losses because of climate shocks. This study underscores the importance of developing targeted policies to protect households in vulnerable regions as climate risks intensify. 

Low-income countries face the combined challenges of climate shocks and limited domestic revenue mobilization, yet these issues are rarely studied together. This paper provides new evidence on the impact of climate shocks on firm performance and tax revenue in a low-income country context, using firm-level data from Zambia. We find that extreme weather events, such as excessive rainfall and high temperatures, significantly reduce firms' sales, input purchases, and tax collection, particularly in sectors like manufacturing, retail, accommodation, and construction. Firms respond by reducing employment and wages, reflecting a decline in productivity. Our findings highlight the need to consider the combined effects of climate shocks on both formal sector productivity and government revenue in developing countries, where the taxation on services and goods responds to a larger share of the government budget.

This systematic review examines the effectiveness of social assistance programs in protecting households in Africa – a region highly vulnerable to climate change, conflict, and other shocks – during periods of crisis. Despite the significant increase in the number of these programs over the past two decades, the need for emergency aid remains. Our review focuses on quantitative studies using microdata to assess the impact of these programs on household income, poverty, assets, and food security outcomes during crises. We find that large, predictable, and consistent transfers are effective in protecting households and building resilience, as they enable households to accumulate buffer stocks and better plan for future shocks. However, challenges with timely implementation often hinder their effectiveness. Cash-plus programs, which integrate consumption support with livelihood interventions, show particular promise in fostering long-term resilience, though the evidence base remains too limited to draw broader conclusions.

This paper exploits several waves of two major nationwide representative surveys to document the impacts of climate shocks on individuals and households in Zambia. We merge these datasets with historical precipitation and temperature data at the district level. First, we show the gendered effects of the shocks, which have a higher negative impact on women. Women have a lower probability of being in the labour force and fewer hours of work when experiencing shocks. Second, we show that households affected by climate shocks have 16% lower consumption, which is mainly explained by female-headed households. We show that social protection policies mitigate income reduction but not consumption, which may suggest that climate shocks affect households directly by reducing their income, but also indirectly by raising food prices. 

This paper investigates the impact of the Science without Borders (Ciência sem Fronteiras - CSF) program on participants' post-graduation enrollment, employment, and entrepreneurship. The program was launched in 2011 to increase students' human capital and interest in science and postgraduate education studies through a substantial increase in scholarships for Brazilians to carry out part of their undergraduate studies abroad. We exploit variation in the approval rate across CSF selection calls for the same destination country and year and combine seventeen public and private administrative records to track CSF candidates' outcomes up to eight years after the call. The main results suggest that the program did not achieve its goals of increasing approved student enrollment in postgraduate education programs in Brazil. Even though the program could have improved student skills and acted as a market signaling, we do not find effects on the probability of working in the formal labor market, or as formal entrepreneurs. Using detailed data from one top university, we show that approved students graduate more often, but take longer to graduate, which may have negative impacts on their labor market outcomes.  Finally, although we cannot rule out that students moved to a foreign country after the program, we show that the likelihood of this event may have decreased over time.