Economic Development Midterm Study Papers

“The Short-Term Impact of Unconditional Cash Transfers to the Poor: Experimental Evidence from Kenya” (Haushofer and Shapiro, 2016)

We use a randomized controlled trial to study the response of poor households in rural Kenya to unconditional cash transfers from the NGO GiveDirectly. The transfers differ from other programs in that they are explicitly unconditional, large, and concentrated in time. We randomized at both the village and household levels; furthermore, within the treatment group, we randomized recipient gender (wife versus husband), transfer timing (lump-sum transfer versus monthly installments), and transfer magnitude (US$404 PPP versus US$1,525 PPP). We find a strong consumption response to transfers, with an increase in household monthly consumption from $158 PPP to $193 PPP nine months after the transfer began. Transfer recipients experience large increases in psychological well-being. We find no overall effect on levels of the stress hormone cortisol, although there are differences across some subgroups.
Monthly transfers are more likely than lump-sum transfers to improve food security, whereas lump-sum transfers are more likely to be spent on durables,
suggesting that households face savings and credit constraints. Together, these results suggest that unconditional cash transfers have significant impacts on
economic outcomes and psychological well-being.

• RQ: What is the impact of UCTs on economic and psychological outcomes?

• Methods: RCT. Treatment included changes in gender, transfer timing, and transfer magnitude.

• Findings: Treatment households increased in consumption, savings, investment, psychological well-being, and female empowerment.

• Validity:

  • Follow: 18 of the 503 households failed to receive VCTs.

  • Randomize: Households receive VCTs differently than intended.

  • Attrition: Low level of attrition and little difference between attrition and non-attrition households.

“Schooling and Labor Market Consequences of School Construction in Indonesia: Evidence from an Unusual Policy Experiment” (Duflo, 2001),

Between 1973 and 1978, the Indonesian government engaged in one of the largest school construction programs on record. Combining differences across regions in the number of schools constructed with differences across cohorts induced by the timing of the program suggests that each primary school constructed per 1,000 children led to an average increase of 0.12 to 0.19 years of education, as well as a 1.5 to 2.7 percent increase in wages. This implies estimates of economic returns to education ranging from 6.8 to 10.6 percent.

• RQ: How does building primary schools affect educational attainment and wages?

• Method: Quasi-experiment. Increased access to primary schools and increased intensity of schools built.

• Findings: Increased access to primary education increases educational attainment, which in turn increases future wages.

• Validity:

  • Follow: Families move in response to the program.

  • Randomize: Placebo check and control for variables that change over time.

  • Attrition: Participants who emigrated outside of Indonesia and/or refused to respond to the survey.

“The Impact of AIDS Treatment on Savings and Human Capital Investment in Malawi (Baranov, Kohler, 2018)

Antiretroviral therapy (ART), a treatment for AIDS, is rapidly increasing life expectancy throughout sub-Saharan African countries affected by the AIDS epidemic. This change in life expectancy has potentially profound influences on life-cycle decisions. A longer life expectancy increases the value of human capital investment, while the effect on savings is theoretically ambiguous and life-cycle saving could increase or decrease. This paper uses spatial and temporal variation in ART availability to evaluate the impact of ART provision on savings and investment. We find that ART availability significantly increases savings, expenditures on education, and children’s schooling, including among HIV-negative individuals who do not directly benefit from ART. These results are not driven by the direct health effects of treatment or reductions in caretaking responsibilities, but rather by reduced perceptions of mortality risk after ART has become available.

• RQ: How does the availability of ART affect savings and education expenditures?

• Method: Distance was measured from households to the ART facility.

• Findings: Decreased distance from the facilities increases savings by 12% and education by 43% on average.

• Validity:

  • Follow: Households are near an ART facility, but don’t have transportation access.

  • Randomize: Similar program begins in the same area at the same time.

  • Attrition: Sickest households drop out of the study.

“Free Distribution or Cost Sharing Presentation” (Cohen, Dupas, 2010)

It is often argued that cost-sharing - charging a subsidized, positive price for a health product is necessary to avoid wasting resources on those who will not use or do not need the product. We explore this argument through a field experiment in Kenya, in which we randomized the price at which prenatal clinics could sell long-lasting antimalarial insecticide-treated bed nets (ITNs) to pregnant women. We find no evidence that cost-sharing reduces wastage on those who will not use the product: women who received free ITNs are not less likely to use them than those who paid subsidized positive prices. We also find no evidence that cost-sharing induces selection of women who need the net more: those who pay higher prices appear no sicker than the average prenatal client in the area in terms measured anemia (an important indicator of malaria). Cost-sharing does, however, considerably dampen demand. We find that uptake drops by sixty percentage points when the price of ITNs increases from zero to $0.60 (i.e., from 100% to 90% subsidy), a price still $0.15 below the price at which ITNs are currently sold to pregnant women in Kenya. We combine our estimates in a cost-effectiveness analysis of the impact of ITN prices on child mortality that incorporates both private and social returns to ITN usage. Overall, our results suggest that free distribution of ITNs could save many more lives than cost-sharing programs have achieved far, and, given the large positive externality associated with widespread usage ITNs, would likely do so at a lesser cost p

• RQ: How does cost-sharing impact the demand for ITNs? Does cost-sharing or free distribution have a more positive impact on health issues?

• Methods: RCT/Field experiment. Clinics distributed ITNs at different prices: free, 10k, 20k, and 40k. 

• Findings: Increased prices did not improve waste prevention. Cost-sharing resulted in a 60% decrease in demand. Demand for ITNs dropped by 75% when sold at market costs compared to being given free. Providing ITNs for free was most effective in reducing malaria and child-mortality cases.

• Validity:

  • Follow: Clinics sell ITNs at a price different than agreed to.

  • Randomize: Other clinics in the area start distribution/selling ITNs that aren’t part of the experiment.

  • Attrition: Participants refusing to report their usage of ITNs.

“The (Perceived) Returns to Education and Demand for Schooling” (Jensen, 2010)

Economists emphasize the link between market returns to education and investments in schooling. Though many studies estimate these returns with earnings data, it is the perceived returns that affect schooling decisions, and these perceptions may be inaccurate. Using survey data for eighth-grade boys in the Dominican Republic, we find that the perceived returns to secondary school are extremely low, despite high measured returns. Students at randomly selected schools given information on the higher measured returns completed on average 0.20–0.35 more years of school over the next four years than those who were not.

• RQ: What is the effect of correcting perceived returns to schooling on students’ decisions to continue schooling?

• Methods: Measured individual and household characteristics as well as expected returns from education.

• Findings: The treatment is expected to increase expected earnings and the chance of students deciding to finish school.

• Validity: 

  • Follow: Statistically significant evidence that provided information increases returns according to the school performance and income between the treatment and control.

  • Randomize: Differences in control and treatment are not statistically significant according to the means, standard deviations, and test of treatment-control covariate balance.

  • Attrition: Family members and the school was contacted if the students didn’t follow up. Response rate was 92%.

“Banking the Unbanked? Evidence from Three Countries” (Dupas et al., 2018)

We experimentally test the impact of expanding access to basic bank accounts in Uganda, Malawi, and Chile. Over two years, 17, 10, and 3 percent of treatment individuals made five or more deposits, respectively. Average monthly deposits in treatment accounts were sizable among users, corresponding to the seventy-ninth, ninety-first, and ninety-sixth percentiles of baseline savings. Survey data show no
discernible intention-to-treat effects on savings or any downstream outcomes, though we cannot reject large effect sizes for active users. Results suggest that policies merely focused on expanding access to basic accounts are unlikely to improve welfare noticeably on average.

• RQ: Does providing free bank accounts help people open accounts, start bank savings, and create improved outcomes in education, health, and business?

• Methods: RCT. Provided 2,160 Malawians, 2,107 Ugandans, and 2,000 Chileans who were unbanked with free bank accounts. From there, 50% of the subjects were randomized to receive free bank accounts, and their savings, income, expenditures, assets, business activity, and demographics were analyzed at 6, 12, and 18 months later for percentage increases.

• Findings: 54% of Ugandans and 69% of Malawians opened an account, but only 17% of Ugandans and 10% of Malawians remained active users with low average deposits in value and usage. Only 17% of Chileans opened an account and only made an average of 0.49 deposits and 0.56 withdrawls.

• Validity:

  • Follow: Anyone who did not open a bank account technically failed to follow the treatment protocol. In recognition of this, the authors experimented on active users, specifically the treatment-on-treatment group.

  • Randomize: Treatment and control groups are different in terms of bank account usage, deposits, and downstream impacts.

  • Attrition: Not a concern because 97% of respondents continued to follow-up. No significantly different amounts of attrition between treatment and control.

“Impact of Saving Groups on the Lives of the Poor” (Karlan et al, 2017)

Savings-led microfinance programs operate in poor rural communities in developing countries to establish groups that save and then lend out the accumulated savings to each other. Nonprofit organizations train villagers to create and lead these groups. In a clustered randomized evaluation spanning three African countries (Ghana, Malawi, and Uganda), we find that the promotion of these community-based microfinance groups leads to an improvement in household business outcomes and women’s empowerment. However, we do not find evidence of impacts on average consumption or other livelihoods.
or other livelihoods.

• RQ: How does access to saving groups affect financial inclusion? How do saving groups affect women’s empowerment, household business outcomes, consumption, food security, and households’ ability to cope with shocks?

• Methods: data from three countries: Malawi, Ghana, and Uganda, and a random trial with villages. Some villages were selected randomly to get help starting VSLAs (Village Savings and Loan Associations), a type of savings group where members gather at regular meetings
  to contribute a fixed amount of money, and the total pot is assigned
  in full to each member in turn, and others not. The ones who got help are called the treatment group, and the other is the control group. There were 282 treatment and 279 control villages, and around 15,000 households in total. Each savings group had about 19 to 30 people. The study used "intent to treat", so even if someone didn’t join VSLA, they still count in the data if they are in the treatment group. Outcomes were measured in percentages of improvement in the areas of interest.

• Findings: Better financial access, business activities, and more power for women. But there was no real change in how people live, like income, food, or spending.

• Validity:

  • Follow: Participant didn’t join the VSLA or report their results.

  • Randomize: If NGOs chose villages that already had better access to markets or already saved more, then those villages might have done better even without the VSLA program.

  • Attrition: High follow-up rate (91%).

“The Miracle of Microfinance? Evidence from a Randomized Evaluation” (Banerjee et al, 2013)

This paper reports results from the randomized evaluation of a group-lending microcredit program in Hyderabad, India. A lender worked in 52 randomly selected neighborhoods, leading to an 8.4 percentage point increase in takeup of microcredit. Small business investment and profits of preexisting businesses increased, but consumption did not significantly increase. Durable goods expenditure increased, while “temptation goods” expenditure declined. We found no significant changes in health, education, or women’s empowerment. Two years later, after control areas had gained access to microcredit but households in treatment area had borrowed for longer and in larger amounts, very few significant differences persist.

• RQ: How does increased access to MFIs impact business outcomes, consumption, credit access, and decrease cost of lending to the poor?

• Methods: RCT. 52 out of 104 randomly selected poor neighborhoods for the opening of a Spandana branch, India’s fastest-growing MFI. Surveys were conducted in the neighborhood over the course of 2 years that measured the impact of MFIs on different areas of life for the treatment and control groups.

• Findings: MFIs don’t impact some business outcomes, such as self-employment, but increase others, such as investment and profit. Also results in increased consumption of durables but no results in other types of consumer products. Also no results were measured in human development outcomes. 

• Validity:

  • Follow: Control areas also get access to the program or when the program is not applied correctly.

  • Randomize: The treatment and control areas are different before the project starts, so results might not come from the program itself.

  • Attrition: Families with too much debt moved away. But the authors found that the dropout rate was about the same (25%) in both treatment and control groups, and most families had lived in the area for at least three years before.

“Credit, Attention, and Externalities” (Berkouwer, Dean, 2022)

We study an energy efficient charcoal cookstove in an experiment with 1,000 households in Nairobi. We estimate a 39 percent reduction in charcoal spending, which matches engineering estimates, generating a 295 percent annual return. Despite fuel savings of $237 over the stove’s two-year lifespan—and $295 in emissions reductions—households are only willing to pay $12. Drawing attention to energy savings does not increase demand. However, a loan more than doubles willingness to pay: credit constraints prevent adoption of privately optimal technologies. Energy efficient technologies could drive sustainable development by slowing greenhouse emissions while saving households money.

• RQ: Does relaxing credit constraints and/or increasing the salience of future energy savings increase the WTP for a cookstove?

• Methods: RCT. The x-variable is the stove prices and credit options. The measured outcomes include stove, charcoal, and household energy usage, which is measured in the household’s amount of money spent on charcoal in addition to its weight. Households were randomly assigned to treatment groups.

• Findings: credit access doubles a household's WTP and the average WTP is $12.

• Validity:

  • Follow: Household not using the credit as promised.

  • Randomize: Potential area of concern due to households potentially being motivated by other factors besides being told the benefits of long-term savings to use energy-efficient technologies.

  • Attrition: Households refuse the credit provided.

“Pay Me Later” (Brune et al, 2021)

We study a simple savings scheme that allows workers to defer receipt of part of their wages for three months at zero interest. The scheme significantly increases savings during the deferral period, leading to higher postdisbursement spending on lumpy goods. Two years later, after two additional rounds of the savings scheme, we find that treated workers have made permanent improvements to their homes. The popularity of the scheme implies a lack of good alternative savings options. The results of a follow-up experiment suggest that demand for the scheme is partly due to its ability to
address self-control issues.

• RQ: How does access to deferred wages affect workers’ spending behavior in the short run, asset accumulation and investments in the medium-term, as well as savings and financial stability in the long-term?

• Methods: RCT. 870 randomly selected workers at an information session at the Lujeri Tea Estates in rural Malawi, whereby the treatment group had the option to defer part of their wages for 3 months, while the control group received only regular wages.

• Findings: average 0.82 percentage point increase in the probability that a worker makes a deposit, notable investments were made in durable home assets in the medium-term, and a 7.7 percentage point increase in reported improved roofing compared to the control group in the long-term.

• Validity:

  • Follow: Researchers fail to contact participants about their expenditures following the payout.

  • Randomize: Subjects who were assigned to the treatment group, not receiving the payout or subjects from the control group receiving the payout.

  • Attrition: Participants failing to or refusing to report their expenditures to the researchers following the payout.

“Do Financial Concerns Make Workers Less Productive” (Kaur et al, 2025)

Workers who are worried about their personal finances may find it hard to focus at work. If so, reducing financial concerns could increase productivity. We test this hypothesis in a sample of low-income Indian piece-rate manufacturing workers. We stagger when wages are paid out: some workers are paid earlier and receive a cash infusion while others remain liquidity constrained. The cash infusion leads workers to reduce their financial concerns by immediately paying off debts and buying household essentials. Subsequently, they become more productive at work: their output increases by 7% (0.11 std. dev.), and they make fewer costly, unintentional mistakes. Workers with more cash on hand thus not only work faster but also more attentively, suggesting improved cognition. These effects are concentrated among more financially constrained workers. We argue that mechanisms such as gift exchange or nutrition cannot account for our results. Instead, our findings suggest that financial strain, at least partly through psychological channels, has the potential to reduce earnings exactly when money is most needed.

• RQ: How does lowering financial strain affect worker attentiveness? Does the effect of lowering financial strain on productivity & attentiveness depend on baseline poverty? How does lowering financial strain affect worker productivity?

• Methods: Experiment in India with factory workers who are paid by how much they produce. Randomly changed when workers got their pay. Some workers received money earlier, and others had to wait. Because it was random, the only real difference was how much cash they had that day. X (measured): when the payment was given (randomly decided). X (interpreted): how workers felt financially — either under pressure or more relaxed — depending on if they had cash that day. Y was productivity and attention

• Findings: Workers produced more. They made fewer errors. Poorer workers showed stronger improvement.

• Validity:

  • Follow: Not really a concern, considering a large portion of the workers spend the money on paying off their loans and household expenses.

  • Randomize: The authors made a difference between the assigned payment day and how much relief workers actually felt. They measured both and studied how real financial relief influenced productivity. This helped reduce the bias from people reacting differently.

  • Attrition: Low attrition wasn’t an issue, as the workers needed the money, which strongly discouraged leaving. This included a completion bonus, which incentivized attendance.

“Why do People Stay Poor” (Balboni et al, 2020)

There are two broad views as to why people stay poor. One emphasizes differences in fundamentals, such as ability, talent, or motivation. The poverty traps view emphasizes differences in opportunities that stem from access to wealth. To test these views, we exploit a large-scale, randomized asset transfer and an 11-year panel of 6,000 households who begin in extreme poverty. The setting is rural Bangladesh, and the assets are cows. The data support the poverty traps view—we identify a threshold level of initial assets above which households accumulate assets, take on better occupations (from casual labor in agriculture or domestic services to running small livestock businesses), and grow out of poverty. The reverse happens for those below the threshold. Structural estimation of an occupational choice model reveals that almost all beneficiaries are misallocated in
the work they do at baseline and that the gains arising from eliminating misallocation would far exceed the program costs. Our findings imply that large transfers, which create better jobs for the poor, are an effective means of getting people out of
poverty traps and reducing global poverty.

• RQ: are people poor because they have traits unsuitable for higher paying jobs or because they are born poor and can’t escape poverty traps and how does transferring large assets (cows) to poor households impact poverty traps.

• Methods: Measured how transferring cows as an asset to poor households, ranging from poor, middle, and upper class, would impact economic position with a specific focus on poverty traps. From there they tracked occupation and assets with surveys over an 11 year long period to measure differences in labor outcomes as well as human and physical capital such as poultry, livestock, tools, machines, vehicles, and land.

• Findings:  Poverty traps do exist as well as that people stay poor due to a lack of opportunities.

• Validity:

  • Follow: Researchers fail to provide the assets to certain subjects.

  • Randomize: Some subjects receive asset transfers similar in value not from the experiment around the same time.

  • Attrition: Participants refuse to report their labor and human outcomes.