What’s the Latest Research in Development Economics? A Round-up from NEUDC 2019

Last weekend was the North East Universities Development Consortium annual conference, held at Northwestern University. Researchers presented more than 150 papers on a wide range of development topics, from agriculture and air pollution to voting and utilities. For those not lucky enough to attend, we’ve prepared bite-sized summaries of each paper available online.

Research is much more than a one-line takeaway (and these takeaways are ours, not necessarily the authors), so if you’re interested in a result, we encourage you to read the linked papers. For many of the papers, we added a brief note to indicate the research method.


Click on any section below to expand the paper summaries.

Households and human capital

Child nutrition and child health

  • “Better water, sanitation and hygiene” increase “the productivity of protein and calories in the formation of child health using as proxies child height and weight,” using data from the Philippines. (Abramovsky et al.)

  • Cash transfers plus communication to change behaviors of pregnant moms reduced stunting in Myanmar. (Field and Maffioli) #RCT

  • In rural Bangladesh, cash transfers or food rations only improved child nutrition when combined with communications intended to change nutrition behavior. (Ahmed, Hoddinott, and Roy) #RCT

  • A training program for parents of young children in Rwanda that included “listening to a radio show and…discussions over the course of seventeen weekly village-level meetings” led to improved child development outcomes nearly three years later. (Justino et al.)

  • Providing community-based parent training and nutrition counseling in Sierra Leone reduced wasting, improved parenting practices, increased fathers’ involvement in parenting, and reduced physical and violent punishments. (Chandra et al.) #RCT

  • A meta-analysis of the impact of mass administration of deworming drugs finds “significant average impacts” in areas with lots of worms. “The weight impact per dollar spent is greater than that from school feeding, a widespread intervention, in settings with over 27.7% prevalence.” (Croke et al.)

  • More air pollution during pregnant mothers’ first trimester reduces children’s subsequent weight-for-age and height-for-age in India, particularly for the poor. (Singh et al.) #IV

  • Being born next to a coal plant in India leaves you shorter. Rich people live closer to coal plants. (Vyas) #FE

  • In India, “a 10% increase in particulate pollution results in more than 90,000 additional infant deaths per year,” with the largest effects for rural people and the urban poor. (Pullabhotla)

  • “Almost half of the global population still relies on solid fuels [e.g., wood, dung, and coal] for cooking.” New analysis from India shows sizeable and adverse impacts on child mortality. (Basu et al.)

  • The use of glyphosate (a pesticide) in agriculture is associated with an increase of 0.93 in the infant mortality rate (per 1,000 births) in Brazil. There are externality effects of glyphosate use on populations distant from the original locations but that receive water from these locations. (Dias, Rochas, and Soares)


  • In Pakistan, female medical graduates marry richer men and are less likely to divorce than women who earn other degrees. The removal of a cap on female medical students led more women to delay marriage and subsequently marry richer men. They also work longer hours if they work. (Aqeel)

  • “How do cultural values regarding gender norms interact with political institutions to determine human capital outcomes?” In Pakistan, an Islamization program beginning in the 1970s had larger adverse impacts on investments in girls’ human capital in areas with weaker existing institutions. (Khan)

  • In Mexico, a major expansion of lower secondary education came through schools with televised lessons (telesecundarias). “Individuals with telesecundaria access earn 14.5 percent higher wages as adults, partly due to being more likely to work and to shifting away from agriculture.” (Navarro-Sola)

  • Providing parents of second graders in Colombia with information about their children’s academic performance improved math and reading scores, but not so for kids in grades three through five. (Barrera-Osorio and Gonzalez)

  • Distributing “tablets preloaded with curriculum aligned learning materials in a group of schools in Kiribati…had a positive impact on English skills of about 27 percent of a standard deviation.” (Ome and Fierros) #RCT

  • “Does gamification in education work?” Math games in low-performing primary schools in Chile lead to sizeable gains in math learning. It also increased “the idea among students that study effort can raise intelligence” (good) and math anxiety (bad). (Araya et al.)

  • In middle schools in Pakistan, in-class technology together with teacher training (to help them integrate the technology effectively) increased both effort and test scores among students. (Beg et al.)

  • In Colombia, just rolling out preschool nationwide did not improve children’s development but giving teachers pedagogical training did. (Andrew et al.) #RCT

  • Randomly selected lower secondary school students in Ghana received guidance on how to apply to upper secondary schools and information on which were the best. It changed which schools students applied to, but it didn’t change whether they graduated. (Ajayi, Friedman, and Lucas) #RCT

  • Targeted vouchers for disadvantaged students seemed to improve test scores, but “low-performing students are 14.7 percentage points less likely to take the national exams four years after the introduction of the reform,” so schools may just be gaming the system. (Sánchez)

  • Vouchers for low-income children in Delhi, India to attend private schools had little to no impact on academic or non-cognitive skills. Students did go to schools with better inputs. (Crawfurd, Patel, and Sandefur) #RCT

  • Scorecards, i.e., providing school performance information to parents, improved school management outcomes (parental satisfaction, public access to school information) in the Angolan province of Kwanza Sul. Collective action, especially when combined with information, is a relevant component of these effects. There was no impact on students’ test scores and absenteeism. (Di Maro et al.)

  • How do parents choose to allocate investments across children? Parents place a high value on equality of inputs. They choose exactly equal inputs 35 percent of the time and forego 40–50 percent of their potential experimental earnings due to inequality aversion in inputs. (Berry, Dizon-Ross, and Jagnani) #Lab-in-the-field

  • Children born in the pay period during hyperinflation in Brazil attain up to 20 percent of a standard deviation more years of schooling than children born in the other periods of the month. Mothers who give birth in the same period are less likely to have mental health problems and are more likely to be working in the future. (Menezes-Filho, Goncalves, and Menezes)

Female empowerment

  • In rural Rajasthan, India, “parents have a strong preference for delaying a daughter’s marriage until eighteen but no further. Conditional on a marriage match, parents place little intrinsic value on a daughter’s education. However, they believe the probability of receiving a good marriage offer increases strongly with a daughter’s education but deteriorates quickly with her age on leaving school.” (Adams and Andrew)

  • Women with better job opportunities have more say in who they marry in India. They’re also less likely to move far from home to marry. (Chatterjee and Heath)

  • “Restricting to garment factory neighborhoods” in Myanmar, “we find that women in close proximity to an exporting factory are significantly more likely to be working, more involved in household decision-making, and have lower tolerance and experiences of domestic violence. (Molina and Tanaka)

  • Women commuters in Rio de Janeiro were 40 percent less likely to experience sexual harassment when able to ride in a women-only train car. A third of riders were willing to forego money to ride in the reserved car. Commuters—both male and female—report that they see women in the non-reserved car as more open to sexual advances. (Kondylis et al.) #RCT


  • Richer patients in the Democratic Republic of the Congo receive better health care. More than half of that is explained by the fact that richer areas have better health facilities, but the relationship holds even within facilities. (Fink, Kandpal, and Shapira)

  • After severe shocks on health spending, households in Thai villages draw on their working capital, cutting input spending and reducing labor hiring, hence propagating the shocks to other households. (Kinnan et al.)

  • What is the demand for glasses in a resource-poor setting? In Burkina Faso, willingness to pay for glasses is low, amounting to 20 percent of their market price. A layaway scheme does not affect willingness to pay, while a video intervention raises the willingness to pay by 16 percent without having a lasting influence on the use of corrective glasses. (Grimm and Hartwig)

  • Clean energy access improves women’s lung capacity and increases women’s labor supply in Indonesia, as a result of higher domestic productivity allowing for more working hours. (Priya and Imelda)

Household bargaining and community interactions

  • Teaching adolescent girls life skills enhanced their perceived social and emotional support, encouraged positive gender norms and goal-setting, and reduced school dropout in Rajasthan, India. (Edmonds, Feigenberg, and Leight) #RCT

  • Mexico’s gender-targeted cash transfer program Progresa more than doubled women’s household bargaining power, with subsequent benefits for the household diet. (Klein and Barham)

  • Introducing unilateral divorce legislation (no consent of spouse or cause needed) did not affect married women’s labor market participation in Mexico. (Hoehn-Velasko and Penglase)

  • Does caste identity affect labor supply? Working on a task associated with another caste group leads to a 23 percentage point drop in the job take-up rate, and an additional 24 percentage drop if the associated caste group ranks lower in the hierarchy than the worker’s own. (Oh) #RCT

  • Is following traditions driven by social image concerns? In Malawi, those who plan to marry off their under-age daughters are seen as more pro-social in villages where child marriage prevalence is high, but alternative signals (public donations) change perceptions and decrease favorable attitudes towards harmful traditions by 20–30 percent. (Haenni and Lichand)

  • Descendants of lepers in Colombia “behave more altruistically and display higher in-group favoritism when compared to individuals who grew up in the same environment but have no socially-excluded ancestry.” What’s more, “social exclusion has a lasting adverse effect on attitudes towards the group that is responsible for the ancestral exclusion, which in this context is captured by unfavorable attitudes towards physicians and modern medicine.” (Ramos-Toro)

Intergenerational mobility and poverty traps

  • Adding communication encouraging nutrition behavior change to cash and food transfer programs leads to larger impacts on both consumption and assets in rural Bangladesh, the latter possibly due to improvements in households’ social capital and empowerment. (Ahmed, Hoddinott, and Roy)

  • Taxes slightly increase the income mobility across generations. Compared to a no-tax environment, there is a large increase in mobility due to higher fertility rates (quantity) and lower educational outcomes (quality) when income taxes are progressive. (Kurnaz and Soytas)

  • A child labor elimination program in the Philippines granted a $518 productive asset grant to families with child laborers. This increased household-based economic activity and well-being, but also drew children, who were not in child labor at baseline, into economic activities. (Edmonds and Theoharides) #RCT

  • While overall intergenerational mobility in India has remained constant since liberalization, there is rising mobility among historically marginalized Scheduled Castes and declining intergenerational mobility among Muslims. (Asher, Novosad, and Rafkin)

Intimate partner violence and gender discrimination

  • Providing women in Ethiopia with jobs increases their income but has no impact on physical intimate partner violence. There are reductions in emotional abuse in the short-run, but for women who had little bargaining power in their relationships before the jobs, the job offers may have increased abuse. (Kotsadam and Villanger) #RCT

  • In rural Bangladesh, more than three out of four women have experienced intimate partner violence. Offering their households interest-free loans to allow husbands to migrate reduced violence in those households. (Mobarak and Ramos) #RCT

  • Fewer girls are born relative to boys when droughts hit in India, but a workfare program helps. As back-of-the-envelope calculation suggests that “the program could have saved around 550 girls per district per year if the government implemented it in the years 2001 to 2005.” (Chatterjee and Merfeld)

  • In Bangladesh, news headlines about sexual assault lead women to reduce their work hours the next day (and men to increase their hours). (Chowdhury)


  • In rural Mexico, people migrate temporarily within the country “in anticipation of crop shocks” induced by extreme temperatures. Women and households with less land per working person are the most likely to move. (Quiñones)

  • Vocational training in Bangladesh led to persistent increases in employment when it was complemented with apprenticeships or stipends. (Shonchoy, Fujii, and Raihan) #RCT

  • Increased female migration within South Africa at the end of Apartheid reduced employment and hours of low-skilled male non-migrants. (Sharp)

  • In Indonesia, a “1 percent increase in the proportion of [internal] migrants in the population leads to a 3.9 percent increase in the number of economically-motivated crimes reported by local media,” but—in an exciting twist—when you look at reports from household surveys, more migrants reduce the probability of being a crime victim. (Feld and Kleemans)

Government, institutions, and conflict

Civil service

  • Performance pay attracted more money-oriented teachers in Rwanda, without comprising teacher effectiveness. Overall, the effect of performance pay is at 0.21 standard deviation of pupil learning. One quarter of this impact can be attributed to selection at the recruitment stage, with the remaining three quarters arising from increased effort. (Leaver et al.)

  • Text messages targeted to civil servants in charge of a school maintenance program in Peru reduced the compliance gap by 15 percent (the distance between the current level of compliance and full compliance with national programs). The effect of monitoring reduced the compliance gap by 21 percent. (Dustar, Maldonado, and Hernandez-Agramonte)

  • In Pakistan, public credit is priced lower to the same firm and have much higher default rates compared to credit from privately owned banks. (Faraz)

  • Employees in the Ministry of Education in Afghanistan report colleague’s wrongdoing (absence) less when offered monetary reward for their reports, only if they expect their reports to be consequential (making their choice morally charged). (Fiorin) #RCT


  • Exposure to terrorist attacks in Kenya reduced children’s primary school enrolment attendance. (Alfano and Görlach)

  • In countries close to ISIS headquarters, high unemployment pushes more individuals to join the terror group, while unemployment hampers recruitment in far-away countries, from which migration is costly. (Brockmeyer et al.)

  • Mobile Victims Unit, a Colombian government intervention that provided access to justice for victims of the armed conflict, increased victims’ reception of reparations as well as their perception of justice, but it also increased the salience of the conflict. (Vargas et al.)

  • The 2010 Haiti earthquake had strong negative impacts on child stunting and wasting as well as on school attendance and attainment, even after accounting for the influx of $2 billion of aid. (Dodlova, Escobar, and Grimm)


  • In India, billed electricity consumption is lower for constituencies of the winning party after an election, however, actual consumption, as measured by satellite nighttime lights, is higher. Politicians illicitly subsidize their constituents by systematically allowing the manipulation of electricity bills. The net deadweight loss is large enough to power 3.7 million rural households over an electoral term. (Mahadevan) #RDD

  • In India, winners of close elections receive three times as many paid work days (through the National Rural Employment Guarantee Scheme) as losers in the year after the election. (Jeong, Shenoy and Zimmermann)

  • Politically connected firms in Ecuador enjoy a higher probability of winning discretionary contracts and charge higher prices for homogeneous goods and services than unconnected firms. Also, politically connected firms are less efficient, translating into a 1.5 to 3.5 percent excess cost of provision. (Brugués, Brugués, and Giambra)

  • Is access to corruption a major reason why workers join the public sector? In Brazil, a 20 percentage point increase in the likelihood of audit causes 3.6 percent of high-level bureaucrats to leave their jobs, suggesting that a government job is no longer optimal for them. (Poulsen)

  • In China, local newspapers underreport corruption scandals involving high-level officials from their own province, especially when a newspaper does not rely on advertising revenue and a corrupt official is well connected. (Zhuang)

  • Regions receive more health aid when a region-born health minister is in office. Newborns from the same region as the health minister are less likely to die, as data from 45 African countries shows. (Widmer and Zurlinden)

Infrastructure and property

  • Two historically marginalized groups India, Scheduled Castes and Scheduled Tribes and Muslims, are concentrated in poorer cities, but Muslims much more so. Cities with more Muslims have worse access to schools and health services, while cities with more Scheduled Castes and Scheduled Tribes have better access. (Adukia et al.)

  • Even though the Bus Rapid Transit system reduced commuting time by about 18 percent in Accra, Ghana, it did not dramatically change how people commute. (Abeka-Nkrumah, Opoku Asuming, and Telli)

  • How spatially misallocated is public infrastructure investment, and why? Misallocation in healthcare infrastructure is lower after Indonesia’s democratization in 1999, because (i) of a reduced bias toward previously Suharto-favored villages, and (ii) spillover effects are less internalized as districts become more focused on their own constituents. (Hsiao)

Political economy, institutions, and voting

  • Mobile phone and internet access reduced violent collective action by 21 percent during the Libyan revolution. (Absher and Grier)

  • Economic distress in Brazil in the 1990s increased adherence to the Pentecostal church, a more fundamentalist religious group, alongside a decrease in adherents to other Christian denominations. (Costa, Marcantonio, and Rocha)

  • Does revolution work? Following the people’s war in Nepal, the caste distribution of elected politicians largely resembles the population they represent, without a representation-ability trade-off. (Bhusal et al.)

  • Do anti-bribery laws affect international trade and investment? Yes, but not because of the laws themselves, but due to an underlying trend of increased political cooperation among OECD countries, leading to more business among OECD countries, which tend to be less corrupt. (Firth)

  • In two-round election systems (ensuring the winner obtains at least 50 percent of the vote) of Brazilian municipal mayoral races, candidates’ voter bases are more geographically diverse, more resources are allocated to public schools, and there is less variance in resources allocated to public schools across the municipality. (Chin) #RDD

  • In India, setting up a grievance redressal system for representatives to issue formal complaints to the higher bureaucracy and signal breakdowns improves public good provision and has significant spillovers onto neighboring jurisdictions. (Sharan and Kumar) #RCT

  • In Pakistan, politicians who receive information about citizen preferences make recommendations to their party leadership that are closer to what citizens prefer. (Liaqat)

  • Are the state and traditional leaders (village chiefs) substitutes or complements? If chiefs are integrated into the institutional structure, chiefs become complements (state presence will increase service provision by the chief). If they are not integrated, they are substitutes (service provision by the chief will decrease with greater state presence). (Henn)

  • How effective are monetary incentives to vote? In Peru, non-monetary incentives provided by compulsory voting, which include the expressive value of the law, social image concerns, and the non-monetary burden of the sanction, vastly outweigh the monetary incentive provided by a fine. (Gonzalez, León-Ciliotta, and Martínez)

  • Do voters care about (education) service delivery in Liberia? Yes! The presidential candidate who had set up private-public partnerships to improve school quality was rewarded by voters in places where the program was successful, and punished where the program diminished school quality. (Sandholtz)

  • An anti-vote-buying campaign in Uganda struggled to instill norms of refusing gifts from politicians in exchange for votes, but it leveled the electoral playing field by convincing some voters to abandon norms of reciprocity—thus accepting gifts from politicians but still voting for their preferred candidate. (Blattman et al.)

  • Providing information about the level of inequality increased the likelihood that respondents would vote against the president in Indonesia. Providing information about a respondent‘s position in the distribution resulted in richer Indonesians becoming less supportive of redistribution. (Hoy, Toth, and Meredikawati)

  • Higher court performance positively affects lending behavior in local credit markets, because timely resolution supports lenders and acts as a deterrent against non-repayment in India. This, in turn, relaxes the credit constraints firms face, expanding production and improving profits. (Rao)

  • A joint licensing platform for drug bundling increases the diffusion of affordable generic versions of drugs as well as the number of new clinical trials. (Wang)

  • In Pakistan, a property rights reform which (i) computerized rural land records and (ii) provided access to digitized records and automated transactions to agricultural landowners and cultivators, improved tenure security and partially relieved land markets frictions. (Beg)

Public finance

  • How to raise property tax revenues? Lowering late fees to the risk-adjusted rate of return, reducing enforcement and increasing the tax rate can raise both revenue and welfare in Mexico City, although the latter may result in non-compliance. (Brockmeyer et al.)

  • An index of the marginal utility expenditure (IMUE) may better capture changes in households’ welfare from receiving transfers: In over 25 percent of the Indonesian villages the IMUE measure rankings had a higher correlation with the community ranking than proxy means tests or total consumption rankings did. (Trachtman)

  • Retirement plays a significant role in explaining cognitive decline. In China, a pension program reduced episodic memory and intact mental status among the elderly, especially among women. (Nikolov and Adelman)

  • Firms in China increase their stock of plant and machinery in response to the investment tax credit offered through the value-added tax framework, especially firms in industries with a high dependence on external finance and firms which have a higher likelihood of being financially constrained. (Aneja, Kulkarni and Ritadhi)

  • Are tax rates too high in developing countries? in Kananga, Democratic Republic of the Congo, when tax liability decreases by half, tax compliance increases by 94 percent, bribes decrease by 75 percent, and there is no change in contributions to informal taxes, particularly among the poorest and least politically connected citizens. (Bergeron, Tourek, and Weigel) #RCT

  • Decreasing tax rates is not always optimal to reduce the informal sector since there is a trade-off with enforcement, creating as a result a range within which (income) tax rates reductions are optimal to minimize informality and increase output. (Henao)

Utilities and energy

  • In India, the “demand for electricity in the wholesale market is downward sloping and substantially more elastic than in Western markets.” Also, “power plant outages are an important contributor to out-of-merit generation.” (Burlig, Jha, and Preonas)

  • The rapid rise of fracking in the United States led to a commodity boom in northwestern India. At the same time, India was rolling out access to electricity. When those two overlapped, non-agricultural employment shot up as firms related to the boom were able to grow. (Fetter and Usmani)

  • When thermal power plants in India are “faced with uncertainty in supply of coal,” they “respond by decreasing consumption of input and decreasing production by effecting partial shutdown.” (Pathak)

  • A field experiment of energy-efficient cookstoves in Nairobi, Kenya, shows “an average rate of return of 300% and savings of $120 per year in fuel costs—around one month of income… Factoring in financial savings and avoided environmental damages we estimate that a subsidy on the energy efficient technology would have a marginal value of public funds of $20 per $1 spent.” (Berkouwer & Dean)



  • India’s 2016 demonetization—which “made 86% of currency in circulation illegal overnight” – increase unemployment among the casual laborers and hurt firms that rely more on informal workers. (Subramaniam)

  • What else did demonetization do in India? Households consumed less in the first months (especially richer households) and borrowed more (especially poorer households). (Wadhwa)

  • Yes, but what else did demonetization do? Using variation in the number of banks (areas with fewer banks had higher cash shortages), “in elections following demonetization, the ruling party did relatively worse in regions with discontinuously fewer banks, receiving a 4.7 percentage point lower fraction of votes.” (Khanna and Mukherjee)

Growth, trade, and cycles

  • “Remittances are dollars wrapped with care”: A 1 percent increase in remittances raises real GDP by approximately 0.07 percent in the long run, based on data from 80 countries. (Ahmad et al.)

  • A joint increase in household and firm domestic financial participation can be an important factor behind several, but not all, defining differences in labor-market and business cycle dynamics between emerging and advanced economies. (Epstein and Finkelstein Shapiro)

  • In Colombia, imports decrease employment and increase income inequality in the short run, while increasing efficiency and decreasing misallocation due to low-productivity firms exiting the market. (Bonilla and Munoz)

  • Dynamic scale economies are substantial in high-skilled production but negligible in low-skilled production. As a result, inter-industry productivity spillovers are larger in richer countries, and play a substantial role in accounting for slow cross-country convergence. (Bolhuis)

The long run

  • Colombian “counties with coffee cultivation in 1925 had around 46% lower employment in manufacturing in 1945 than counties not exposed to coffee cultivation… People born in coffee counties, exposed to high prices during school age, have lower levels of education.” (Uribe-Castro)

  • “Districts more exposed to the trade” collapse in India during World War I “experienced substantially faster industrial growth, both during and after the war… The impact of the trade shock was entirely due to an increase in the number of Indian-born (as opposed to British-born) administrative staff and workers, and in the number of Indian-owned (as opposed to British-owned) firms.” (Bonfatti and Brey)

Working and saving


  • Agricultural workers in rural India who switch to non-agricultural jobs in the same areas earn 21 percent higher wages. (Baysan et al.)

  • What drives subsistence farmers to start growing cash crops in Uganda? Agricultural extension did, especially for farmers who started with poor info on the price of the crops. (Bonan, Kazianga, and Mendola) #RCT

  • Providing subsidized watchmen to farmers in Kenya increased agricultural production and dramatically reduced disputes among farmers. (Dyer) #RCT

  • Price supports for wheat and rice in India lead farmers both to lower agricultural productivity (because farmers are putting too many laborers to work on the price-supported crop) and lower productivity in manufacturing (since farmers are hiring workers away from manufacturing to work on the price-supported crop). (Krishnaswamy)

  • Accrediting sellers (retailers) based on buyer (farmer) satisfaction leads to higher agricultural input quality and more repeat purchases when combined with loyalty rewards, but it does not enhance quality signals and farmer welfare. (de Brauw and Kramer) #Lab-in-the-field

  • Village level inequity aversion and fairness norms protect tenants by lowering the rent landowners can charge in rural Malawi. (Krah et al.)

  • Training farmers on aflatoxin (a food safety hazard) and its prevention substantially improves post-harvest practices in Northern Ghana. (Magnan et al.) #RCT

  • In the Indian State of Bihar, soil testing and targeted fertilizer recommendations had no effect on fertilizer use nor on farmers’ willingness to pay for micronutrients. (Fishman et al.) #RCT

  • In Burkina Faso, households with access to warrantage substantially increased the take-up of storage  (94 percent), while credit take-up was moderate (38 percent). (Delavallade and Godlonton)

  • Two different interventions in Kenya—including a game that helped farmers understand how weather index insurance works—both led more farmers to indicate they’d buy it in an auction. Two months later, very few bought it regardless. (Janzen et al.)

  • In Rwanda, irrigation enables dry season horticultural production, which boosts farm cash profits by 70 percent. However, adoption is constrained: Access to irrigation causes farmers to substitute labor and inputs away from their other plots. Eliminating this substitution would increase adoption by at least 21 percent. Substitution is largest for smaller households and wealthier households. (Jones et al.) #RDD

  • “Experience significantly increases willingness to pay for” a new grain storage technology in Bihar, India. (Shukla and Baylis)


  • Chinese FDI in district-sector induces competing domestic Ethiopian firms to shrink, as output prices drop, while firms in up/downstream sectors expand. There is a zero effect on the local economy. (Crescenzi and Limodio)

  • Free riding in loan approvals: Among borrowers without a credit history, once they receive a bank loan approval from one bank, many end up receiving a loan originating from competing financial institutions. (Arraiz et al.)

  • A large-scale credit expansion program in Thailand showed no effect in income or business profits among low pre-program productivity households, while the higher-productivity households show a large increase in profits, with a doubling effect when restricting to high-productivity households with a non-agricultural business at baseline. (Banerjee et al.)

  • Effects of interest rates caps on loans for large and medium-sized manufacturing businesses in Bangladesh. (Miyauchi)

  • Effects of interest rate caps in Bangladesh: A 1 percentage point increase in the pre-regulation interest rate induced a 17 percent increase in outstanding loan amounts and an 11 percent increase of the number of outstanding loans 1 year after the onset of regulation. This increase persisted after the regulation was lifted. (Miyauchi)

  • Placing “young professionals for one month in established firms to shadow middle managers” in Addis Ababa, Ethiopia, increases their likelihood of subsequent wage employment (but not self-employment). (Abebe et al.)

  • In India, larger manufacturing firms charge higher prices, but among firms that sell to both rich and poor, prices fall when poor households’ incomes fall due to bad weather. (Gupta)

  • As palm oil factories proliferated in Indonesia, areas around those factories had “more non-agricultural employment, higher incomes, and more people.” (Edwards)

  • A seven-day training for firms in Liberia on how to bid on “contracts from large buyers that are awarded through a formal bidding process” led firms to bid on more contracts and to win both more and better contracts. The biggest effects were for firms that had access to the internet. (Hjort, Iyer, and Rochambeau) #RCT


  • “Low-income adults in Chennai sleep 5.6 hours of objectively-measured sleep per night, despite 8 hours in bed.” Increasing night sleep had no impact on productivity, cognition, or psychological well-being. Offering afternoon naps at work did all three. (Bessone et al.)

  • During Ramadan, Muslim salespeople in Indonesia are nearly one-third less productive in the two hours before sunset. After sunset, productivity shoots back up. (Hu and Wang)

  • Across Chinese cities, “college-educated workers respond more strongly to pollution than the unskilled by emigrating.” Reducing pollution is good for growth both because of direct effects on health and because it lets people settle where they’re most productive. (Khanna et al.)

  • Among women school-feeding workers in South Africa, “private feedback on performance is more effective at boosting effort than competing for public recognition,” and “image motivation crowds out intrinsic motivation.” (Delavallade and Burns) #RCT

  • Certifying the skills of youth looking for jobs in urban South Africa “and allowing them to share the certification with firms substantially increases employment and earnings.” (Carranza et al.)

  • Randomly providing women in India with job opportunities resulted in them having “greater autonomy to make independent decisions but” not greater “bargaining power in collective household decisions.” (McKelway)

  • A policy in Argentina that “increased sanctions and the probability of detection for employers of informal workers led to a 36% increase in registration rates of domestic workers, a 4% increase in labor income and a 4.5% in total earnings per month.” But none of those earnings gains went to those workers earning the least. (Feld)

  • Providing transfers of food in India led to people working a bit less, but this means their wages rose, which was good for the poor. (Shrinivas, Baylis, and Crost)

  • Sometimes people worry that providing unemployment insurance will lead people to not look as hard for a job, and people often use that “unemployment outflow” (imagine people flowing back into jobs) as an indicator of this problem. But what about when unemployment insurance also make people more likely to leave their existing jobs (i.e., “unemployment inflow”)? New analysis from Brazil shows that the highest costs are from the latter effect. (Ferreira and Machado)

  • When the government monitors firms in Mexico, workers are more likely to transition from informal to formal jobs, and those workers’ subsequent jobs are more likely to be formal. (Bujanda and Parra)

  • In Colombia, getting laid off increases the chances of getting arrested, both for the person laid off and for young people in their family. Access to credit reduces the effect. (Khanna et al.)


  • Rice import restrictions in Indonesia benefitted villages more suited for rice production in terms of aggregate income and nutrition. Local governments responded by directing more resources toward the more adversely affected villages: They were more likely to receive a health facility. (Sim)

  • How does risk affect technology adoption by farmers? In Malawi, a one standard deviation increase in the coefficient of variation of predicted yields reduces fertilizer application (use of improved seeds) by between 12.2–18.8 percent. Sensitivity to price risk is highest early in the season, when reliable information on output prices is still many months away. (Soumaila and Dillon)

  • Access to India’s National Rural Employment Guarantee Act (NREGA) slightly increases yields in years with average rainfall, but it also makes yields more sensitive to low rainfall shocks. Two channels are at play: (i) NREGA income allows farmers to make high-risk, higher-return decisions; (ii) NREGA increases farm wages, especially in low rainfall years, leading to reductions in crop yields. (Taraz) #DID

  • In Mexico, expected income shocks have no effect on household behavior, while unexpected income shocks decrease labor supply, food security, mental health, and healthful behavior, but do not affect cognition, risk and time preferences, expectations and aspiration about children’s educational attainment. (Angelucci et al.)

Savings and credit

  • Microfinance institutions usually don’t build flexibility for seasonal income shocks into their repayment plans, in part for fear of interrupting the payment discipline of borrowers. In an experiment in Bangladesh, “seasonality adjusted flexible microcredit” led to no increase in default and no decrease in repayment frequency, but after a year, it did translate to higher food intake during the lean season. (Shonchoy and Kurosaki)

  • According to a new model, “when financiers are impact investors rather than donors, organizations may operate at a larger scale.” Other implications are outlined. (Roth)

  • A large expansion of microcredit in Rwanda led more people to take loans not only from microlenders but also from commercial banks. (Agarwal et al.)

  • In Bangladesh, after two years of entrepreneurial training and receiving livestock, women had the option of accessing microcredit. Those who applied for credit were those who were already performing relatively well with their assets and income, and further screening by loan officers also predicted better loan repayment. (Hossain and Mullally)

  • Offering a savings account that automatically deducts from the paycheck and then pays out after three months (with zero interest) led agricultural workers in Malawi to save more and then make more large purchases after the payout. They also work more. (Brune, Chyn, and Kerwin)

  • Microfinance programs reduce the effect of informal financial networks on consumption smoothing among Chinese households. (Cai) #RCT

  • Amongst marginally eligible clients for a loan in Paraguay, those who were marginal because of their financial behavior worsened their credit situation after receiving a loan, while for those who were unknown by financial entities, obtaining access had no negative impacts on credit scores and default. (Azevedo et al.) #RDD

  • Excessive top-down pressure in group lending microfinance programs may result in households inefficiently investing the loan in safer projects rather than high-risk and high-reward projects. (Ananth)

  • Microcredit that is labelled—linked with the investment by name only—is a viable strategy to improve uptake of lumpy preventive health investments in rural India. (Augsburg et al.)



  • “Using nationally representative data from four sub-Saharan African countries, we find strong evidence that measurement error in plot size reflects a mixture of farmer misreporting and misperceptions.” (Abay, Bevis, and Barrett)

  • If you think that diaries are a better way to measure consumption than asking people to recall their consumption, think again! New data from Iraq show comparable outcomes, with particular dangers from diary data in certain areas. (Battistin, Nadai, and Krishnan)

Innovations in policy choice and behavioral economics

  • Rather than a simple randomized controlled trial to help policymakers choose among policy options, this work suggests assigning different treatments in waves and updating them over time. (Kasy and Sautmann)

  • You might already be familiar with the behavioral concept of “present bias.” Let us introduce you to “parent bias,” which can explain the difference between what parents plan to do and the share of their future consumption that actually gets allocated to their children. (Lichand and Thibaud)


This blog post is also available on the Development Impact blog.


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.