Recommended
Blog Post
What’s the latest in international economic development research? Last weekend was the North East Universities Development Consortium annual conference, often called NEUDC. With more than 135 papers presented (and almost all of them available for download), it’s a great way to see recent trends in the international development research by economists and to learn about new findings.
The studies come from all over the world, as you can see in Figure 1 below. Just like last year, the plurality of studies take place in India (30 studies). Kenya is next (12), then Bangladesh (8), Brazil (7), China (7), and Indonesia (7). More than 40 countries are represented overall, from almost all regions of the world. (For an analysis of where submitters and presenters are based, as well other details of the conference, see Jillian Stallman’s blog post.)
Figure 1: Where are recent development economics studies focused?
Source: This map draws on a sample of 139 studies from the NEUDC 2022 conference. Studies that covered more than three countries (often broad global or regional analyses) were excluded.
Researchers draw on a wide range of empirical methods. Nearly a third of studies reported on the results of a randomized controlled trial (43 studies). Other commonly used methods include difference-in-differences, fixed effects, and instrumental variables.
Figure 2: What empirical methods do recent development economics papers used?
Source: This chart draws on a sample of 139 studies from the NEUDC 2022 conference. Some studies used more than one method.
Below, we provide a quick takeaway from every paper in the conference for which we could find a digital copy. As you read our takeaways, keep the following in mind. First, we can’t capture all the nuance of a paper in a couple of lines. Second, our takeaway may not be the authors’ takeaway. Third, some of the papers are marked as preliminary and not ready for formal citation (you can see which if you follow the paper links). Fourth, we largely take the findings of these papers at face value: most have not yet been through peer review, so feel free to dig into the data and analysis to decide how confident you are in the results.
Our takeways are sorted by topic. If your principal interest is in a country or region, you can also read the takeaways sorted by country. We provide some indication of the empirical method used (for empirical papers) with hashtags at the end of the takeaways. Some papers fit into more than one category: for example, is a paper about the impact of free childcare on mothers’ careers about labor or about gender? It’s about both! In those cases, we’ve repeated studies in multiple sections below so if you’re focused on health, you’ll find all the health-related papers in the health section. The second or third time a paper appears, we put an asterisk after the summary so you can skip it if you’re reading straight through.
Happy learning!
Guide to the methodological hashtags
#DID = Difference-in-differences
#FE = Fixed effects
#IV = Instrumental variables
#LIF = Lab in the field
#PSM = Propensity score matching
#RCT = Randomized controlled trial
#RD = Regression discontinuity
Households and human capital
Education and Early Childhood Development
- How critical are family conditions in early years for child development? Better weather (which means more agricultural income and better nutrition) at age 2 in Indonesia leads to higher adult cognitive ability. When households face hard times at earlier ages, they compensate with prolonged breastfeeding. (Webb) #FE
- Data from Indonesia suggest that parental education and parental income are the main drivers of differences in skills once kids grow up. (Thomas) #Other
- “Many teachers [in India and Bangladesh] underestimate the share of low performers in their classrooms, and...they believe that those students will perform better than they actually do. These results are not driven by less educated, trained, or experienced teachers or explained by biases against female, low-income, or lower caste students.” (Djaker, Ganimian, and Sabarwal) #Other
- A 1985 change in Indian law discouraging the payment of dowries led to a 24 percent drop in dowry payments, but it also led to an 18 percent reduction in girls' education attainment (with no impact on boys' education). (Jha) #DID
- Providing information about a learning app in Bangladesh didn't lead more people to use it, but it did lead some parents to arrange more tutoring, resulting in "lasting math learning gains, concentrated among richer households." (Beam, Mukherjee, and Navarro-Sola) #RCT
- Students in Kenya often apply to secondary schools with little information about the available schools. Providing information to students "led them to apply to" schools that were closer to home "without compromising school quality." Adding parents to those information meetings "led students to enroll in lower cost schools." (Bonds) #RCT
- Among students in 9th grade in India, student test scores rose similarly whether they were exposed to "rigidly defined remedial lessons that take time away from the curriculum" and "teacher determined remedial lessons," which allow teachers more flexibility. (Beg et al.) #RCT
- Parental aspirations for their children matter, but they may not be enough on their own. “In rural Gambia, families with high aspirations for their children’s future education and career, measured before children start school, go on to invest substantially more than other families in the early years of their children’s education. Despite this, essentially no children are literate or numerate three years later. When villages receive a highly-impactful, teacher-focused supply-side intervention, however, children of these families are 25 percent more likely to achieve literacy and numeracy than other children in the same village.” (Eble and Escueta) #RCT
- A ten day increase in the overlap between school days and peak farming periods in Malawi translates to children losing about a third of a year of schooling. (Allen) #IV
- Eliminating school fees for secondary school in Tanzania led not only to increased secondary school enrollments; it also increased primary school pass rates. (Sandholtz) #DID
- Indonesia's major school construction program from the 1970s led to eight percent overall higher national output forty years later, and much of that comes through migration from rural to urban areas. (Hsiao) #DID
- Phone call tutorials during COVID-19 were effective at boosting learning in India, Kenya, Nepal, Philippines, and Uganda, whether implemented by government teachers or non-government organization instructors. (Angrist et al.) #RCT
- An affirmative action policy in Brazil was effective at redistributing university spots to low-income students, with little drop in average achievement. "The policy also reduced the gap in applications to selective majors" between poor and rich students by more than 50 percent, but note that those are applications: many of those major-choice changes were among students unlikely to be accepted into a highly selective major. The policy worked, but it could work even better. (Melo) #DID
- Peru shut down a bunch of low-quality universities in 2015. Graduates from surviving universities experienced an increase in wages and higher employment rates. (Vivar, Flor-Toro, and Magnaricotte) #DID
- An influx of Syrian refugees in Jordan reduced school enrollment among Jordanians, particularly boys and kids with less-educated parents. More young Jordanians went to work instead. (Almuhaisen) #DID
- An edutainment program in Bangladeshi schools to trace school-to-home transmission of handwashing find that children are induced to wash more at school but less at home, yielding a net negative effect of the program (Hussam and Oh) #RCT
- Removing English language study from pubic primary schools in West Bengal, India, increased private school enrollment and---for those still in public schools---increased private tutoring among the richest households. (Nandwani and Sen) #FE
- In utero exposure to high ocean salinity levels (induced by climate change) reduces a child’s height-for-age z-score in Bangladesh, and increased prevalence of stunting and severe stunting due to nutritional deficiencies by age five. (Guimbeau et al.) #FE
- In Indonesia, “remittances increase household consumption, reduce poverty, and stimulate growth. Households send more children to school, and district governments increase public schools at the primary and junior secondary levels.” (Hilmy) #FE
- The tariff reduction from the U.S-Vietnam Bilateral Trade Agreement decreased school attendance and increased children’s work, mainly in non-wage and household business jobs. Effects were stronger for boys, older children and households where the head had little education. (Nguyen) #DID
- Giving a widely known award "to top performers on a mandatory nationwide exam in Colombia" boosts their earnings by between 7 and 12 percent, and the effect endures for 5 years after graduation. It helps students graduating from low-reputation colleges the most. (Busso, Montaño, Muñoz-Morales) #RD
Health (including mental health)
- A national vaccine program in Burkina Faso in 1984 boosted the country’s child vaccination rate (against measles, yellow fever, and meningitis) from 17 percent to 77 percent in a few months. Child mortality fell, primary school completion rose, and—when the children reached adulthood—employment and agricultural productivity rose. (Daramola et al.) #DID
- School-based deworming in western Kenya—nearly a quarter of a century later—reduced under-5 mortality of the beneficiaries’ children by 24 percent! (Walker et al.) #RCT
- In Ecuador, letting employees use work time to get a flu vaccine boosted vaccination rates, but employees got sick about as much. Why? Some evidence suggests that employees engaged in “riskier health behaviors after getting vaccinated.” (Hoffman, Mosquera, and Chadi) #RCT
- In Kenya, “both patient subsidies and pharmacy incentives for diagnostic testing significantly increase usage of testing and may encourage malaria positive individuals to purchase high quality antimalarials.” (Dieci) #RCT
- Women who were babies in utero during a cholera epidemic in Peru in the 1990s were nearly 20 percent more likely to die of COVID-19. (Ritter and Sanchez) #DID
- After four years of using iron and iodine fortified salt in school lunches in India, children have lower likelihood of anemia, higher hemoglobin levels, but no differences in cognitive or educational outcomes. (Grafenstein et al.) #RCT
- Gold mining in the Philippines created new bodies of stagnant water, which boosted malaria cases by nearly a third (relative to provinces without gold deposits). (Pagel) #DID
- Giving households a flyer about mobile health services in rural Bangladesh didn't get them to use it more, but offering to save the access numbers in the participants' phone boosted take-up by 22 percent in the succeeding 2 months and reduced health expenditure, since households were less likely to go to "informal providers who usually overprescribe medicines." (Sardar) #RCT
- A drug procurement program in China "brought down the prices of 10 chronic condition drugs by an average of 78" percent. As a result, "drug adherence was improved for the uninsured who had poorer adherence" before the price reduction. (He and Yang) #DID
- In Dakar, Senegal, it can be hard to find someone to desludge your septic pit. Providing subsidies to use a government run call center to connect households with desludgers increases use, and that use continues for a while after the subsidies end. Later, a city-wide subsidy increased adoption most in those communities that had received subsidies earlier. (Deutschmann) #RCT
- How critical are family conditions in early years for child development? Better weather (which means more agricultural income and better nutrition) at age 2 in Indonesia leads to higher adult cognitive ability. When households face hard times at earlier ages, they compensate with prolonged breastfeeding. (Webb) #FE *
Fertility and family planning
- "Learning about government mistreatment of citizens undermines trust in institutions. In Perú, “disclosure of information about illegal sterilization reduced usage of contraceptive methods, prenatal and delivery services, and the demand for medical services, resulting in worsened child health."" (León-Ciliotta, Zejcirovic, and Fernandez) #DID
- During the colonial period in the Congo, greater exposure to Catholic nuns increased women’s fertility (as opposed to exposure to Protestant or male Catholic missionaries). Catholic nuns likely promoted the image of an ideal Christian woman which explains the results. (Guirkinger and Villar) #DID
Households and marriage
- Households in Bangladesh reduced their monthly residential electricity use by 15.8 percent (≈37 kWh) when they switched from postpaid electricity metering system to prepaid metering. (Das) #IV
- A new method to infer causal effects on choices that exploits relationships between choices and hypothetical evaluations “can recover treatment effects even if the treatment is assigned endogenously and standard estimation methods are poorly suited, or if the treatment does not vary.” (Bernheim, et al.) #Other
- In the Democratic Republic of the Congo, 73 percent of households provided with access to childcare centers use them, with significant gains in early childhood development outcomes, particularly for younger children. Women reported increases in their concentration and sense of control. Both parents “increase their engagement in commercial activities, leading to gains in agricultural productivity, household income and women’s subjective well-being.” (Donald and Vaillant) #RCT
- Can commitment-saving ahead of a lean season alter consumption downfalls among the ultra-poor? In Bangladesh, a temporary savings subsidy doubled formal savings, and resulted in increased food and non-food expenditure by 8.6-12.6 percent during the lean season, with no lasting post-lean season impact. (Takahashi et al.) #RCT
- An “edutainment” intervention designed to reduce child marriage in rural Pakistan, significantly reduces marriage of girl adolescents. Targeting men alone reduced child marriage in sample households, while targeting women or men & women jointly reduces child marriage at the village level. (Cassidy et al.) #RCT
- In Kenya, workshops and couples’ therapy sessions to decrease alcohol consumption lowered prevalence of sexual intimate partner violence (IPV) by 0.21 standard deviations, with smaller or no effects on physical and emotional IPV. (Castilla, Aqeel, and Murphy) #RCT
Migration and refugees
- Can temporary foreign work permits “throttle human smugglers’ businesses? “Combining internal and external controls with a regulated market for temporary visas alleviates the policy trade-off between migration control and ending human smuggling.” Data from migration between Senegal & Spain and the Democratic Republic of the Congo & South Africa. (Auriol, Mesnard, and Perrault) #Other
- Massive exodus of Venezuelans in Colombia had a larger negative effect on the lower tail of the natives’ wage distribution, increasing inequality in the host economy. Due to formal restrictions, immigrants ended up working in more routine and low-paying jobs. “A large-scale amnesty program reduced the magnitude of downgrading, mitigating the unequalizing impact of the exodus.” (Lombardo et al.) #IV
- In Mexico, children in households with return migrants (from the U.S.) “benefit from an increase in school attendance and a decrease in the probability of schooling delay relative to children in non-migrant households.” However, females in return migrant households are likely to complete a lower grade relative to non-migrant households. (Chakraborty, Bucheli, and Fontenla) #IV
- An evaluation of a large-scale migration loan program in Bangladesh revealed that capacity constraints at scale lead effort to be directed toward those already planning to migrate without a loan. (Mitchell et al.) #RCT
- A Zambian fertilizer subsidy program led to “some households to intensify their agricultural activity, and others to out-migrate.” The subsidy increased the share of households with outmigrants by 40 percentage points and doubled the number of outmigrants net of in-migrants. (Diop) #DID
- Clearance of slums in Santiago, Chile, and families’ relocation to public housing in low-income areas led to displaced children having 10 percent lower earnings and 0.5 fewer years of education as adults than non-displaced. (Rojas-Ampuero and Carrera) #FE
- In refugee camps and surrounding communities in Uganda and Kenya, refugee children can be up to three times more likely to be poor than adults. Child’s age, household composition, and access to sanitation and clean water, predict child poverty in refugee settlements well, often better than per-capita household expenditure. (Beltramo et al.) #ML
- In Indonesia, “remittances increase household consumption, reduce poverty, and stimulate growth. Households send more children to school, and district governments increase public schools at the primary and junior secondary levels.” (Hilmy) #FE *
- During WWII, nine ethnic groups were entirely deported from the Soviet Union to Central Asia. In the 50s, five returned to their former homeland, while the other four remained marginalized in internal exile. Locals in host regions had significantly higher levels of education two generations later. “A strong positive effect on higher education is found among returnees to origin regions, suggesting that these ethnic groups hedged against further negative shocks.” (Zimmermann) #IV
- An influx of Syrian refugees in Jordan reduced school enrollment among Jordanians, particularly boys and kids with less-educated parents. More young Jordanians went to work instead. (Almuhaisen) #DID *
- Pairing employers in Uganda with a refugee and providing an incentive to offer a free internship to that refugee "improves employers’ beliefs about refugees’ skills, but it does not change their willingness to hire new refugees," but certain types of matches (depending on employer and refugee characteristics) do result in more refugee hires. (Loiacono and Silva-Vargas) #RCT
Gender
- In India, Hindu women are subject to caste “purity” norms, while Adivasi, or Indigenous, women are not. “Having more Adivasi neighbors leads to: (i) higher rates of Hindu women’s paid work and lower perceived stigma of such work; and (ii) lesser adherence to a range of purity norms, including the practice of untouchability towards Adivasis.” (Agte and Bernhardt) #FE
- Does free childcare improve mothers’ careers? Yes. In Sao Paulo, Brazil, mothers in sub-districts with above median childcare availability have a persistent increase of 8 percent in earnings, driven by 1 percentage point higher labor force participation and 4 percent longer hours. (Garcia, Latham-Proença, and Mello) #FE
- Can cash transfers influence gender roles? In Chad, cash transfers increased women’s business profits (0.6 SD), and marital separation (by 13 percent). The program also “led to large improvements (0.3-0.7 SD) in a broad set of women’s subjective well-being, including self-efficacy, depressive symptoms, and perceived social status.” (Kandpal, Schnitzer, and Dayé) #RD
- In Nigeria, women prefer to defer budget allocation decisions to their husband even when deferral is costly and is not observed by the husband; the reverse is true for husbands. A randomized cash transfer receipt increases women’s demand for agency: if the decision is hidden from the husband, women want to make their own budget decisions, even if it is the same as their husband’s. (Bakhtiar et al.) #LIF
- Showing teachers in Pakistan a pro-women’s rights award-winning movie (the 2011 film Bol) increased their own and students’ support for women’s rights, being unbiased in gender Implicit Association Tests, and willingness to petition parliament for greater gender equality. (Mehmood, Naseer, and Chen) #RCT
- Profiles for women who signal on an online Indian matchmaking site that they want to work after marriage receive up to 22 percent less interest from men than those of women who have never worked. Women willing to give up work after marriage face a lower penalty. (Dhar) #RCT
- In Chile, informing outstanding students in mathematics and science about their relative performance and presenting STEM majors as a feasible option, led to women applying more, but only in health-related majors, and not in STEM majors. (Ramirez-Espinosa) #RCT
- In Brazil, union bargaining that prioritized women’s needs increased female-centric amenities (like longer maternity leave with job protection) at work. These led to women queueing for jobs at treated establishments and separating from them less, which are both indicators of firm value. (Corradini, Lagos, and Sharma) #DID
- While both gender barriers to occupational choices and wage penalties persist across countries, the “reduction in wage gaps between 1980-2000 was primarily driven by economic channels while the more recent decline between 2000-2015 was driven by changes in gender barriers.” (Chiplunkar and Kleineberg) #Other
- “A program targeting ultra-poor women in Uganda” paired “business and entrepreneurship skills development with psychological empowerment.” It increased profits by 105 percent. (Lang and Seither) #RCT
- Expansion of the coffee mills in Rwanda led to increased “women’s paid employment, women’s and their husbands’ earnings and decreases domestic violence.” Decline in violence is driven by women’s increased bargaining power and their contribution to household earnings, not exposure reduction between couples. (Sanin) #DID
- Sharing a hyperlocal digital job search platform with couples as well as the wives' social networks in Delhi, India, increased husband's labor market outcomes (including working hours and total earnings), but only home-based self-employment among the women, potentially due to social norms. (Afridi et al.) #RCT
- New data from more than 90 countries demonstrates three things: (1) the shift out of agriculture that happens as countries grow richer is driven by whole households (not just individuals within households), (2) "in the poorest countries, the gap between female and male market employment is only large for married urban women," and (3) "countries where employment rates of urban married women are low relative to their rural counterparts also see low urbanization rates of married men." (Doss et al.) #Other
- In Kenya, workshops and couples’ therapy sessions to decrease alcohol consumption lowered prevalence of sexual intimate partner violence (IPV) by 0.21 standard deviations, with smaller or no effects on physical and emotional IPV. (Castilla, Aqeel, and Murphy) #RCT *
- A 1985 change in Indian law discouraging the payment of dowries led to a 24 percent drop in dowry payments, but it also led to an 18 percent reduction in girls' education attainment (with no impact on boys' education). (Jha) #DID *
Working and saving
Banking and credit
- In India "delinquent borrowers who are offered a debt moratorium by their lender are 4 percentage points (6.9 percent) less likely to default on their loan, while forbearance has no effect on repayment if it is granted by the regulator.” (Fiorin, Hall, and Kanz) #RCT
- What are the household welfare gains from financial inclusion? Welfare gains per dollar lent or saved (retirement savings in the United States, commitment savings in the Philippines, and microfinance in Mexico) are small as compensated demand elasticities are large, but still correspond to large aggregate welfare gains from financial inclusion. (Loeser) #RCT
- In Ghana, microenterprises receiving joint liability loans reported higher profits six to ten months after borrowing. Effects are driven by borrowers whose applications were not endorsed by political party operatives. (Boso, Burlando, and Abdul-Rahaman) #FE
- A self-help group lending program in rural Bihar, India, “significantly improved risk-sharing in regions where the program had greater institutional capacity and was better implemented.” (Attanasio et al.) #FE #IV
- In Kenya, “performance-contingent microfinance contracts can encourage investment and increase profits – and, as a result, increase household consumption.” (Cordaro et al.) #RCT
- In India, “plants exposed to banking shocks redistribute this liquidity through the supply chain. As a result, firms extending trade credit can increase their own sales as their customers are able to purchase on credit. Downstream firms are able to increase their own sales, employment, and productivity.” (Chakraborty et al.) #DID
- In India, “risk pooling creates a distortion in consumption such that food consumption is better protected from aggregate village shocks than nonfood consumption.” (Fafchamps and Shrinivas) #Other
Cash transfers
- Can cash transfers influence gender roles? In Chad, cash transfers increased women’s business profits (0.6 SD), and marital separation (by 13 percent). The program also “led to large improvements (0.3-0.7 SD) in a broad set of women’s subjective well-being, including self-efficacy, depressive symptoms, and perceived social status.” (Kandpal, Schnitzer, and Dayé) #RD *
Firms and microenterprises
- In India, “larger cultural proximity [by way of caste and religion] between a pair of firms reduces prices and fosters trade at both the intensive and extensive margins.” (Fujiy, Khanna, and Toma) #FE
- After close elections in India, entrepreneurs from the same social group as the winning candidate are more likely to start businesses. (Bhalla et al.) #FE
- Politically connected firms in India were more likely to get access to short-term credit from banks and to be able to delay short-term payments to suppliers and creditors during the surprise demonetization of 2016. (Chen et al.) #Other
- Variation of COVID lockdowns over time and across parts of India reveal that inputs delivered by suppliers within the same industry are complements (rather than substitutes), which means that "shocks propagate through supply chains," increasing the shock to overall GDP. (Fujiy, Ghose, and Khanna) #FE
- "Starting in 1997, India dismantled its policy of product reservation whereby hundreds of products had been reserved for exclusive production by small firms." The effect? "entry in the downstream product market increases with no observable decline in quality of entrants." (Rastogi) #Other
- Inviting Zambian farmers to participate in a simple budgeting exercise (i.e., think through their budget and formulate a spending plan) increased how much they expected to spend for the coming year by 20-60 percent and lowered their willingness to pay for a nonessential item of clothing by 34 percent. By the end of the year, farmers decreased their expenditures by 15 percent and ended up with one additional month of savings. (Augenblick et al.) #RCT
- A new way to measure productivity of retailers in low- or middle-income countries captures their three-fold need to attract customers, manage a storefront, and maintain inventory across many products. In Malawi, "the three dimensions of productivity are correlated with one another" but not perfectly, so that a training that focuses on just one may fail to boost overall productivity. (Huntington) #Other
- In Mexico, a rise in gas prices led to an increase in mom-and-pop shops but "their average size and quality fell." (Ramos-Menchelli and Sverdlin-Lisker) #IV
- Waiving competitive bidding for small-value purchases in Brazil led to 23 percent more expensive purchased products. At least half of this overpricing is explained by discretion allowing agencies to purchase higher-quality products. (Fazio) #FE
- “A program targeting ultra-poor women in Uganda” paired “business and entrepreneurship skills development with psychological empowerment.” It increased profits by 105 percent. (Lang and Seither) #RCT *
Labor (including child labor)
- A late 1990s labor market reform in China led to tens of millions of layoffs in a short period. That led to a drop in employment for workers who did not finish high school—by 20 percent in the industrial sector—and a 5 percent increase in the high school completion rate. (Zhao) #DID
- “A 2014 Bolivian law that recognized the work of children as young as 10 years old, whose age placed them below the minimum working age of 14 years old, enabling them to legally work (subject to a work permit) while simultaneously extending benefits and protections to child workers” (such as adult minimum wages) actually decreased work for children under 14. (Lakdawala, Martínez Heredia, and Vera-Cossio) #Other
- A federal policy that set minimum fares for drivers of motorcycle taxis on a ridesharing app in Indonesia led to higher trip prices but not driver earnings, both because more drivers signed on for any given day AND drivers logged onto the app for more hours, meaning that each driver got fewer rides. (Nakamura and Siregar) #DID #SC
- Providing mentorship to vocational students in Uganda to help with their training-to-work transition increased their likelihood of working a few months later by more than a quarter and also boosted their incomes after a year. Why? It's mostly through info about how the entry-level job market works (not through referrals): As a result, mentored youth "turn down fewer job offers." (Alfonsi, Namubiru, and Spaziani) #RCT
- College students in Mumbai, India, were less likely to share information about jobs if they knew they'd have to compete for them, and the men in particular tended not to share the information with the peers they viewed as having high abilities. (Chiplunkar, Kelley, and Lane) #RCT
- Why do "workers in richer countries experience faster rates of wage growth over their lifetimes than workers in poorer countries"? Cross-country data suggest that workers in rich countries received more training from the firms they work for, and that this is a major component of workers' skills. "Firm-provided training accounts for 38% of cross-country wage growth differences." (Ma, Nakab, Vidart) #Other
- Many interventions help workers with job searches. Doing that without increasing the number of jobs could limit the effectiveness of those interventions. On the other hand, "making it easier for firms to find qualified workers could reduce the cost of hiring" and generate more jobs. With an intervention to subsidize job searches for people in Ethiopia, the lack of jobs ends up limiting the effectiveness. (Van Vuren) #RCT
- A survey in Accra, Ghana, showed that lots of job vacancies were not widely circulated, and---as a result---many employers are unable to find qualified workers during six months. But publishing detailed advertisements on a state-operated online portal increases both the likelihood of finding workers and of those workers being suitable for the jobs. (Lambon-Quayefio et al.) #RCT
- The timing of when auctions for public procurement contracts end in Brazil is random, which permits comparison of winners and runners-up. "Winning a government contract increases wages." (Carvalho, Galindo da Fonseca, and Santarrosa) #IV
- In rural Kenya, a “future orientation” workshop that teaches participants techniques to imagine a positive future, lay out concrete short-term steps to achieve their vision, and plan for obstacles, lifted aspirations and expectations. It led to increased labour supply and spending on productive inputs. The “intervention is at least twice as cost-effective as an (unconditional) cash transfer.” (Orkin et al.) #RCT
- Peru shut down a bunch of low-quality universities in 2015. Graduates from surviving universities experienced an increase in wages and higher employment rates. (Vivar, Flor-Toro, and Magnaricotte) #DID *
- Does free childcare improve mothers’ careers? Yes. In Sao Paulo, mothers in sub-districts with above median childcare availability have a persistent increase of 8 percent in earnings, driven by 1 percentage point higher labor force participation and 4 percent longer hours. (Garcia, Latham-Proença, and Mello) #FE *
- Can temporary foreign work permits “throttle human smugglers’ businesses? “Combining internal and external controls with a regulated market for temporary visas alleviates the policy trade-off between migration control and ending human smuggling.” Data from migration between Senegal & Spain and the Democratic Republic of the Congo & South Africa. (Auriol, Mesnard, and Perrault) #Other *
- Pairing employers in Uganda with a refugee and providing an incentive to offer a free internship to that refugee "improves employers’ beliefs about refugees’ skills, but it does not change their willingness to hire new refugees," but certain types of matches (depending on employer and refugee characteristics) do result in more refugee hires. (Loiacono and Silva-Vargas) #RCT *
- Expansion of the coffee mills in Rwanda led to increased “women’s paid employment, women’s and their husbands’ earnings and decreases domestic violence.” Decline in violence is driven by women’s increased bargaining power and their contribution to household earnings, not exposure reduction between couples. (Sanin) #DID *
- Sharing a hyperlocal digital job search platform with couples as well as the wives' social networks in Delhi, India, increased husband's labor market outcomes (including working hours and total earnings), but only home-based self-employment among the women, potentially due to social norms. (Afridi et al.) #RCT *
- Giving a widely known award "to top performers on a mandatory nationwide exam in Colombia" boosts their earnings by between 7 and 12 percent, and the effect endures for 5 years after graduation. It helps students graduating from low-reputation colleges the most. (Busso, Montaño, Muñoz-Morales) #RD *
- New data from more than 90 countries demonstrates three things: (1) the shift out of agriculture that happens as countries grow richer is driven by whole households (not just individuals within households), (2) "in the poorest countries, the gap between female and male market employment is only large for married urban women," and (3) "countries where employment rates of urban married women are low relative to their rural counterparts also see low urbanization rates of married men." (Doss et al.) #Other *
Poverty Measurement
- Poverty is often measured using annual income. But using monthly data from India shows that “experiences of poverty are substantially more common than annual measures suggest; entry into and exit from poverty are much less clear than typically assumed;” and the use of monthly poverty measures “is a stronger predictor of development outcomes – child weight and height – than conventional” annual measures. (Merfeld and Morduch) #Other
- Are poverty lines real? This study articulates social and theoretical underpinnings for such a distinction between the poor and the nonpoor. (Dutta et al.)
- Many social programs identify their beneficiaries collaboratively with multiple community members together. Comparing the judgments that the same individuals make about who to target when they’re deciding collaboratively versus individually in Indonesia suggests that gains from collaborative targeting are negligible. “These results suggest that policymakers should think carefully before asking community members to invest valuable time in participating in [community-based targeting] exercises.” (Trachtman) #Other
Governments, institutions, and conflict
Conflict and crime
- A national-level electoral reform in Mexico that increased politicians’ cost of accepting bribes decreased the number of suspicious financial reports by around 4 percent (∼650 fewer reports), while the number of attacks by violent groups increased by approximately 2 percent (∼44 more attacks), in places with the presence of criminal organizations. (Rámon Enríquez) #DID
- How does ethnic violence and subsequent segregation shape children’s lives? In India, Muslims perform better in cities that were more susceptible to (Hindu) communal violence in terms of early education outcomes. (Kalra) #PSM
- In India, judges are more likely to convict offenders in cases of sexual crimes (excluding rape) if they are exposed to more media coverage about sexual crimes that are unrelated to the case on trial. A central mechanism behind this result is heightened judicial scrutiny in response to greater media coverage. (Vasishth) #DID
- During WWII, nine ethnic groups were entirely deported from the Soviet Union to Central Asia. In the 50s, five returned to their former homeland, while the other four remained marginalized in internal exile. Locals in host regions had significantly higher levels of education two generations later. “A strong positive effect on higher education is found among returnees to origin regions, suggesting that these ethnic groups hedged against further negative shocks.” (Zimmermann) #IV *
Corruption
- In Colombia, close family connections in the public administration are pervasive and they weaken performance. “Connected bureaucrats receive higher salaries and are more likely to be hierarchically promoted but are negatively selected in terms of public sector experience, education, and records of misconduct.” A 2015 anti-nepotism legislation had limited effectiveness. (Riaño) #RCT
- In China, after a high-profile corruption inspection, labor “strikes experienced a twofold increase within six months and a threefold increase in two years.” (Chen) #Other
- Reducing corruption in China "induces positive selection for integrity and public-mindedness into the state sector, which remains present even when conditioning on ability and family background." (Hong) #DID
- Do autocrats favor their supporters during economic shocks? The Maduro regime was more likely to exempt regime-supporting regions affected by the Venezuelan blackout from later power rationing. In Sub-Saharan Africa, “droughts magnify differences in development, protests and state-coercion outcomes in favor of leaders’ home regions.” (Morales-Arilla) #FE
Elections
- In Brazil, a 1 percentage point increase in the share of Pentecostals increased Evangelical and far-right candidates’ vote share by 18 percent and 16 percent, respectively. These effects are larger in municipalities with less educated, poorer, and more rural populations. (Solá Cámpora) #DID
- During a social media ban at the climax of the Uganda 2021 election campaign, those maintaining access through the use of virtual private networks (VPNs)—who are more likely to be opposition partisans—came to view the dominant National Resistance Mmovement party relatively positively. This is driven by an increase in pro-NRM social media content during the ban. (Bowles, Marshall, and Raffler) #DID
Taxes and subsidies
- Simulations suggest that removing subsidies for electric vehicles in China would raise the effective marginal costs of vehicle production, reducing total welfare by 7.4 billion yuan (RMB) per quarter, amounting to 13.9 percent of the total subsidy expenditure. (Kwon) #IV
- In South Africa, "firms with paid tax practitioners exhibit sharper bunching, driven primarily by a lower lumpiness parameter rather than by a different income elasticity." (Anagol et al.) #Other
- In Liberia, including identifying information (the name of the owner and a property photograph) in tax bills more than quadruples the tax payment rate, from 2 percent to almost 10 percent, when the notice also includes details on the penalties for noncompliance. Compliance goes up even more when the tax bill warns delinquent property owners that they’re in the “next batch” of properties designated for “intensive follow-up.” (Okunogbe) #RCT
- An increase in the subsidy for liquefied natural gas (LNG) in India leads to a surprising decrease in LNG purchases by poor households. Why? The subsidy goes up when the market price rises—i.e., the government keeps the price to consumers stable—but consumers only receive the subsidy as a refund a few days after they purchase the LNG, and “poor households may find it difficult to pay the higher unsubsidized price upfront.” (Afridi, Barnwal, and Sarkar) #FE
Regulation and government
- In Argentina, “serving in the military leads to stronger national identity and openness to fellow countrymen several decades after serving.” (Ronconi and Ramos-Toro) #FE
- In Pakistan, policymakers who received a special training in econometrics are twice as likely to actually choose policies for which there is evidence from randomized controlled trials. They triple the funding for those same policies. (Mehmood, Naseer, and Chen) #RCT
- In Brazil, state judges with higher grades on admission examinations perform better than their lower-ranked peers. (Dahis, Schiavon, and Scot) #FE
- In India, improving the maintenance of fee-funded community toilets improved delivery and reduced free riding, but excludes a share of residents from using the service. (Armand et al.) #RCT
- In, Kenya’s nationwide electrification program, imposing audits improved network size, voltage, household connectivity, and electricity usage at African Development Bank (AfDB) sites. (World Bank sites already had lots of inspections, so random audits didn’t affect those.) The World Bank’s procedures delayed construction at the average site by 9.6 months relative to AfDB sites but improved construction quality by 0.6 standard deviations. (Wolfram et al.) #RCT
- In Bihar, India, instituting a complaint filing system for politicians who run into bureaucratic obstacles in the implementation of their projects led to a 26 percent rise in the implementation of public projects in constituencies run by low-caste local politicians. (Kumar and Sharan) #RCT
- Informing government agents about illegal (gold) mining in Colombia, reduced illegal mining by 11 percent in the disclosed locations and surrounding areas. (Saavedra) #ML
- In India, switching the approving authority of economic development projects that require forest diversion from central to state government “leads to a modestly adverse impact on forest conservation while approving lower quality development work.” (Chiplunkar and Das) #DID
- When two districts in India share groundwater resources, extraction is more likely to be unsustainable and districts are more likely to have defunct wells. (Bhogale and Khedgikar) #DID
Agriculture, infrastructure, and the environment
Agriculture and land
- In Mozambique, contract farming has spillover effects: “both contracted farmers and non-contracted farmers from villages with contracted farmers earn approximately 11 percent more in price per kilogram of maize” than farmers in areas without contract farming. (Ingram) #FE
- The vanilla price boom in Madagascar led to increased asset accumulation and higher informal savings, improved performance on cognitive tests, well-being, and optimism. There were positive effects for smallholder vanilla farmers, but without spillover benefits on non-producing households. (Boone, Kaila, and Sahn) #FE
- In Rwanda, using text messages to remind agricultural extension workers of their self-set goals increased their performance by 0.08 SD. (Abel et al.) #RCT
- How do rural marketplaces shape local development? In Kenya, “while rural population quadrupled, two thirds of weekly markets operating in 1970 no longer do so today.” Population concentrated around markets that were active in 1970, and markets further from large cities saw the most population concentration. (von Carnap) #FE
- If you rely on farmer reports on what seeds they’re using in Ethiopia, you’ll see apparently small, negative returns to using new seed varieties. But that’s because farmers are including “older and genetically diluted varieties, for which” they “may be paying a premium.” If you just look at seed varieties with “higher-purity germplasm, drought-tolerant maize, and newly released varieties,” you do see positive returns. (Michuda et al.) #Other
- How much is one farmer in Uganda willing to pay for their neighbor to use a pest-control technology? A “novel incentive-compatible elicitation mechanism” finds that “a farmer’s willingness-to-pay for one other farmer [to use the technology] is equal to two days’ wage on average, about half the willingness-to-pay for self.” (Lerva) #RCT
- Maybe farmers don’t train their workers in modern planting technologies because they know they won’t get all the benefits of that training: the workers may go and use it somewhere else. In Burundi, providing a contract that guaranteed farmers the benefit from training their workers increased farmers’ willingness to train their workers by about 90 percentage points! (Cefala et al.) #RCT
- In Odisha, India, villages with more variation in castes have lower adoption of flood-resistant seeds. Adoption is more likely to spread within castes and less likely to spread to castes with lower status. (de Janvry, Rao, and Sadoulet) #RCT
- In India, bureaucrats who are assigned to their home states decrease protests or riots in opposition to land acquisition projects by 9-12 percent. (Tóth) #DID #RD
- Lake Chad shrunk by 90 percent between the 1960s and the late 1980s: the reduced water supply had negative effects for neighboring communities in Cameroon, Chad, Nigeria and Niger—25 percent of sub-Saharan Africa’s population—on fishing, in addition to farming and herding. (Jedwab et al.) #DID *
Climate and pollution
- Lake Chad shrunk by 90% between the 1960s and the late 1980s: the reduced water supply had negative effects for neighboring communities in Cameroon, Chad, Nigeria and Niger – 25% of sub-Saharan Africa’s population – on fishing, in addition to farming and herding. (Jedwab et al.) #DID
- Households from heavily damaged communities in Indonesia after the 2004 Indian Ocean Tsunami saw a 75 percent decline in real wealth immediately after the tsunami. The large adverse effects persisted 10 years after the tsunami. (Lombardo et al.) #FE
- In India and Pakistan, the incidence of fires from crop burning drops by 10 percent when air pollution is likely to be borne by the bureaucrats’ own constituents. “Reduction in fires is present or larger when bureaucrats can better monitor (due to road proximity) or manage fires (due to changes in experience during a turnover), and when they have incentives to act (e.g., when pollution is most visible).” (Dipoppa and Gulzar) #DID
- In utero exposure to high ocean salinity levels (induced by climate change) reduces a child’s height-for-age z-score in Bangladesh, and increased prevalence of stunting and severe stunting due to nutritional deficiencies by age five. (Guimbeau et al.) #FE *
Infrastructure
- In Kenya and Ethiopia, the “impact of bundled road and electricity investments on reducing the sectoral employment share in agriculture is … 2.5 times larger than the impact of roads alone.” (Dappe and Lebrand) #FE #IV
- Aerial bundled cables (an infrastructure improvement) in Karachi, Pakistan, reduced utility losses and increased revenue recovery, with the greatest gains in the worst-performing areas pre-intervention. Gains come via two channels: the formalization of customers previously informally (illegally) connected and the improvement in payment behaviors among existing, formal consumers. (Ahmad et al.) #DID
- A large-scale roll-out of electric transmission infrastructure in Nigeria from 2009 to 2015 increased individuals’ likelihood of migrating by 6 percent and reduced household size by 0.8 individuals, mainly driven by young adults and older teenagers. (Budjan) #DID #IV
- How does transportation infrastructure investment affect spatial inequality of opportunity in Benin, Cameroon, and Mali? “On average only 6 of the top 10 aggregate opportunity-increasing roads also decrease inequality of opportunity.” (Milsom) #IV
- In Dakar, Senegal, it can be hard to find someone to desludge your septic pit. Providing subsidies to use a government run call center to connect households with desludgers increases use, and that use continues for a while after the subsidies end. Later, a city-wide subsidy increased adoption most in those communities that had received subsidies earlier. (Deutschmann) #RCT *
Macroeconomics
Growth and inequality
- Gaining subnational capital status leads to an influx of public investments, an increased population, skilled migration and foreign investments, with positive spillovers to nearby cities. (Bluhm, Lessmann, and Schaudt) #FE #DID
- In China, state-level special economic zones (SEZs)—"geographically delimited areas targeted by governments to implement” policies like tax incentives, government innovation grants, and policies that favor human capital mobility—"have a positive and significant impact on patent output,” but SEZs established at geographic areas smaller than the state don’t have significant impacts. (Wu, Lu, and Zhao) #DID
- Initial foreign direct investment into China “was mainly driven by the Chinese diaspora,” particularly to prefectures in China that members of the diaspora came from. Later, foreign investment that didn’t come from the diaspora was more likely to enter those same prefectures. (Chen, Xiong, and Zhang) #DID
- “Countries are more likely to enter ‘nearby’ industries, i.e., industries that require fewer new occupations.” Also, “countries are more likely to diversify into products that require fewer new inputs,” which means that countries can get stuck on particular paths in their quest toward industrial development, “with certain routes leading to stagnation and others on a pathway to prosperity.” (Diodato, Hausmann, and Schetter) #Other
- “The construction process of many residential buildings in African cities proceeds very slowly and may take over a decade.” Data from Nairobi plus a new theoretical model suggest that “improvements in credit provision can (a) substantially speed up the expansion of the aggregate housing stock which facilitates rural-urban migration, and (b) increase the city’s density by enabling the construction of taller buildings.” (Gomtsyan) #Other
Trade
- Giving information about market prices and official border costs to traders in Kenya increases switches across markets and routes, leading to a large improvement in traders’ profits and significant formalization of trade. (Wiseman) #RCT
- Can temporary trade disruptions lead to a persistent change in domestic trade? Yes. In India, COVID-19 induced lockdowns led to a collapse in trade across states, driven by plants reorienting “trade towards their home states to ... insure against any future disruptions.” (Chakrabati, Mahajan, and Tomar) #DID
- The 2001 US-Vietnam Bilateral Trade Agreement reduced US import tariffs from Vietnam, leading to rapid Vietnamese export growth with “entry responses, driven by foreign and Vietnamese private firms. Entrants—rather than incumbents—drive the tariff-induced employment growth, particularly foreign entrants.” (McCaig, Pavcnik, and Wong) #FE
- The tariff reduction from the U.S-Vietnam Bilateral Trade Agreement decreased school attendance and increased children’s work, mainly in non-wage and household business jobs. Effects were stronger for boys, older children and households where the head had little education. (Nguyen) #DID *
- In India, “larger cultural proximity [by way of caste and religion] between a pair of firms reduces prices and fosters trade at both the intensive and extensive margins.” (Fujiy, Khanna, and Toma) #FE *
The order of authors on this blog was determined by a virtual coin flip. This blog post benefited from research assistance from Amina Mendez Acosta. It also appears on the Development Impact blog and the Yale Economic Growth Center site.
Disclaimer
CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.