Pub Date : 2025-12-05DOI: 10.1016/j.jdeveco.2025.103697
Cihan Artunç , Mohamed Saleh
We use a newly assembled dataset covering all Egyptian corporations, their founders, and political officeholders, to demonstrate the differential impact of political connections on firm performance across two distinctive political and economic contexts. Before Egypt’s independence in 1922, political connections reduced firm profitability, as connected firms were perceived to be aligned with the anti-colonial, nationalist movement, unsettling investors. After independence, connections improved firm outcomes by granting preferential access to incorporation and shielding connected companies from competition. These dynamics reflect the shift from a laissez-faire colonial regime to a nationalist industrial policy that selectively favored politically connected firms.
{"title":"Connected national capital: Corporations in colonial and independent Egypt","authors":"Cihan Artunç , Mohamed Saleh","doi":"10.1016/j.jdeveco.2025.103697","DOIUrl":"10.1016/j.jdeveco.2025.103697","url":null,"abstract":"<div><div>We use a newly assembled dataset covering all Egyptian corporations, their founders, and political officeholders, to demonstrate the differential impact of political connections on firm performance across two distinctive political and economic contexts. Before Egypt’s independence in 1922, political connections reduced firm profitability, as connected firms were perceived to be aligned with the anti-colonial, nationalist movement, unsettling investors. After independence, connections improved firm outcomes by granting preferential access to incorporation and shielding connected companies from competition. These dynamics reflect the shift from a laissez-faire colonial regime to a nationalist industrial policy that selectively favored politically connected firms.</div></div>","PeriodicalId":48418,"journal":{"name":"Journal of Development Economics","volume":"180 ","pages":"Article 103697"},"PeriodicalIF":4.6,"publicationDate":"2025-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145840606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-05DOI: 10.1016/j.jdeveco.2025.103694
Mo Alloush , Emily Conover , Susan Godlonton
Across and within countries, there are large differences in how parents discipline their children, and frequently, poverty is associated with higher levels of physical punishment. We leverage the roll-out of a conditional cash transfer program in Peru to test whether its introduction changes parental discipline practices. We find that in districts that begin to receive the program, the average level of reported physical punishment by mothers and fathers among the poor declines by at least 2.7 percentage points (11%) driven by reductions in slapping. Our findings suggest that program participation may have additional second-order benefits through the reduction of harsh physical forms of discipline practices.
{"title":"Poverty and parental discipline","authors":"Mo Alloush , Emily Conover , Susan Godlonton","doi":"10.1016/j.jdeveco.2025.103694","DOIUrl":"10.1016/j.jdeveco.2025.103694","url":null,"abstract":"<div><div>Across and within countries, there are large differences in how parents discipline their children, and frequently, poverty is associated with higher levels of physical punishment. We leverage the roll-out of a conditional cash transfer program in Peru to test whether its introduction changes parental discipline practices. We find that in districts that begin to receive the program, the average level of reported physical punishment by mothers and fathers among the poor declines by at least 2.7 percentage points (11%) driven by reductions in slapping. Our findings suggest that program participation may have additional second-order benefits through the reduction of harsh physical forms of discipline practices.</div></div>","PeriodicalId":48418,"journal":{"name":"Journal of Development Economics","volume":"180 ","pages":"Article 103694"},"PeriodicalIF":4.6,"publicationDate":"2025-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145738161","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-29DOI: 10.1016/j.jdeveco.2025.103692
Caiquan Bai , Jing Cao , Ming Huang , Chen Xi , Peng Zhang
This paper leverages novel high-frequency electricity consumption data to examine whether urban and rural residents in China are adapting unevenly to short-run extreme heat shocks and long-run climate warming. We demonstrate that both urban and rural groups increase their electricity consumption during transient day-to-day extreme heat events, with the former showing significantly stronger responses, thus revealing a notable adaptation inequality. We document the existence of the urban heat island, but its role in explaining the observed urban–rural adaptation disparity is negligible. Instead, the more fundamental urban–rural real income gap, closely tied to adaptive capacity, almost entirely accounts for the observed difference. In the long run, we find that climate warming over the past 5–20 years significantly promotes adaptation strategy in urban and rural groups, but no evident disparity is observed between these groups. Rural residents have experienced higher income growth over the past decade, which enabled them to afford durable adaptive equipment that is on par with that of their urban counterparts.
{"title":"Feeling the same heat? Urban–rural inequalities in adapting to extreme temperatures via electricity consumption","authors":"Caiquan Bai , Jing Cao , Ming Huang , Chen Xi , Peng Zhang","doi":"10.1016/j.jdeveco.2025.103692","DOIUrl":"10.1016/j.jdeveco.2025.103692","url":null,"abstract":"<div><div>This paper leverages novel high-frequency electricity consumption data to examine whether urban and rural residents in China are adapting unevenly to short-run extreme heat shocks and long-run climate warming. We demonstrate that both urban and rural groups increase their electricity consumption during transient day-to-day extreme heat events, with the former showing significantly stronger responses, thus revealing a notable adaptation inequality. We document the existence of the urban heat island, but its role in explaining the observed urban–rural adaptation disparity is negligible. Instead, the more fundamental urban–rural real income gap, closely tied to adaptive capacity, almost entirely accounts for the observed difference. In the long run, we find that climate warming over the past 5–20 years significantly promotes adaptation strategy in urban and rural groups, but no evident disparity is observed between these groups. Rural residents have experienced higher income growth over the past decade, which enabled them to afford durable adaptive equipment that is on par with that of their urban counterparts.</div></div>","PeriodicalId":48418,"journal":{"name":"Journal of Development Economics","volume":"180 ","pages":"Article 103692"},"PeriodicalIF":4.6,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-29DOI: 10.1016/j.jdeveco.2025.103695
Daniel Gerszon Mahler, Marta Schoch, Christoph Lakner, Minh Cong Nguyen
This paper develops a method to predict comparable income and consumption distributions for all countries in the world from a simple regression with a handful of country-level variables. To fit the model, the analysis uses around 2,000 distributions from household surveys covering 168 countries from the World Bank's Poverty and Inequality Platform. More than 1,000 economic, demographic, and remote sensing predictors from multiple databases are used to test the models. A model is selected that balances out-of-sample accuracy, simplicity, and the share of countries for which it can be applied. The paper finds that a parsimonious model relying on gross domestic product per capita, under-5 mortality rate, life expectancy, and rural population share gives almost the same accuracy as a complex machine learning model using 1,000 indicators jointly. This small set of basic indicators related to human development explains most cross-country variation in income distributions and can facilitate distributional analysis even in countries with extreme data deprivation.
{"title":"A parsimonious approach to predicting income distributions","authors":"Daniel Gerszon Mahler, Marta Schoch, Christoph Lakner, Minh Cong Nguyen","doi":"10.1016/j.jdeveco.2025.103695","DOIUrl":"10.1016/j.jdeveco.2025.103695","url":null,"abstract":"<div><div>This paper develops a method to predict comparable income and consumption distributions for all countries in the world from a simple regression with a handful of country-level variables. To fit the model, the analysis uses around 2,000 distributions from household surveys covering 168 countries from the World Bank's Poverty and Inequality Platform. More than 1,000 economic, demographic, and remote sensing predictors from multiple databases are used to test the models. A model is selected that balances out-of-sample accuracy, simplicity, and the share of countries for which it can be applied. The paper finds that a parsimonious model relying on gross domestic product per capita, under-5 mortality rate, life expectancy, and rural population share gives almost the same accuracy as a complex machine learning model using 1,000 indicators jointly. This small set of basic indicators related to human development explains most cross-country variation in income distributions and can facilitate distributional analysis even in countries with extreme data deprivation.</div></div>","PeriodicalId":48418,"journal":{"name":"Journal of Development Economics","volume":"180 ","pages":"Article 103695"},"PeriodicalIF":4.6,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145665451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-27DOI: 10.1016/j.jdeveco.2025.103688
Justice Tei Mensah , Peter Chacha Wankuru , Benard K. Kirui
How important is government business to the private sector in developing economies? We use administrative tax data on firm-to-firm transactions in Kenya to examine the effects of becoming a government contractor on firm performance. Using an event study design, we document significant gains from becoming a supplier to a government entity. Four years later, beneficiary firms experience a 27% increase in productivity and employ 10% more. These effects are somewhat comparable to the gains from joining a multinational supply chain. Beneficiary firms also expand their trading networks to other private firms. Relaxing credit constraints and improving resilience to shocks are likely operative channels of impact. These findings highlight the potential welfare gains from improving efficiency in public procurement.
{"title":"Public procurement and firms: Evidence from Kenya","authors":"Justice Tei Mensah , Peter Chacha Wankuru , Benard K. Kirui","doi":"10.1016/j.jdeveco.2025.103688","DOIUrl":"10.1016/j.jdeveco.2025.103688","url":null,"abstract":"<div><div>How important is government business to the private sector in developing economies? We use administrative tax data on firm-to-firm transactions in Kenya to examine the effects of becoming a government contractor on firm performance. Using an event study design, we document significant gains from becoming a supplier to a government entity. Four years later, beneficiary firms experience a 27% increase in productivity and employ 10% more. These effects are somewhat comparable to the gains from joining a multinational supply chain. Beneficiary firms also expand their trading networks to other private firms. Relaxing credit constraints and improving resilience to shocks are likely operative channels of impact. These findings highlight the potential welfare gains from improving efficiency in public procurement.</div></div>","PeriodicalId":48418,"journal":{"name":"Journal of Development Economics","volume":"180 ","pages":"Article 103688"},"PeriodicalIF":4.6,"publicationDate":"2025-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145665450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-25DOI: 10.1016/j.jdeveco.2025.103693
Jaideep Roy , Bibhas Saha
When automation in a developing economy displaces low-skilled workers in the advanced sector, backward sector wages may fall due to in-migration of the ‘newly’ unemployed. Fear of job and income loss may then induce office-seeking political parties to announce regulatory policies on automation for electoral success. We show that absent sectoral spillover, democratic adoption of automation is relatively higher and protects only high-skilled jobs in the advanced sector. However, the possibility of spillover limits this adoption. More specifically, if the backward sector is large, automation faces full resistance. In contrast, if the advanced sector is large, automation is moderate, making only the low-skilled jobs vulnerable. But these vulnerable workers, unlike their counterparts in the backward sector, may prefer automation because their advanced-sector wages fall below the severance pay plus backward-sector opportunities. When neither sector is large, the size of automation becomes uncertain, pushing similar economies into different growth paths.
{"title":"Electoral politics over automation in a dual economy","authors":"Jaideep Roy , Bibhas Saha","doi":"10.1016/j.jdeveco.2025.103693","DOIUrl":"10.1016/j.jdeveco.2025.103693","url":null,"abstract":"<div><div>When automation in a developing economy displaces low-skilled workers in the advanced sector, backward sector wages may fall due to in-migration of the ‘newly’ unemployed. Fear of job and income loss may then induce office-seeking political parties to announce regulatory policies on automation for electoral success. We show that absent sectoral spillover, democratic adoption of automation is relatively higher and protects only high-skilled jobs in the advanced sector. However, the possibility of spillover limits this adoption. More specifically, if the backward sector is large, automation faces full resistance. In contrast, if the advanced sector is large, automation is moderate, making only the low-skilled jobs vulnerable. But these vulnerable workers, unlike their counterparts in the backward sector, may prefer automation because their advanced-sector wages fall below the severance pay plus backward-sector opportunities. When neither sector is large, the size of automation becomes uncertain, pushing similar economies into different growth paths.</div></div>","PeriodicalId":48418,"journal":{"name":"Journal of Development Economics","volume":"179 ","pages":"Article 103693"},"PeriodicalIF":4.6,"publicationDate":"2025-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145623672","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-24DOI: 10.1016/j.jdeveco.2025.103689
Ömer F. Sözbir
Analyzing children’s agency is crucial for understanding the inner workings of households. This issue is especially salient for developing countries, as childless families constitute a small portion of the population. This study tests the intra-household decision-making power of work-eligible children (age 12–17) in rural Bangladesh using a restriction implied by the collective household model. By comparing households with working and non-working children, the study also tests whether working children have a greater say in their households than non-working ones. The findings suggest some evidence for children’s decision-making power. The unitary household model is strongly rejected, and there is no evidence against Pareto efficiency. The results are particularly useful for the growing body of research that estimates intra-household inequality and individual-level poverty in developing countries using the collective model with certain assumptions on children’s decision-making power. In addition to investigating children’s say in the household, the study addresses several methodological issues in the literature regarding the price-based test of alternative collective models and Pareto efficiency.
{"title":"Children’s say in the household","authors":"Ömer F. Sözbir","doi":"10.1016/j.jdeveco.2025.103689","DOIUrl":"10.1016/j.jdeveco.2025.103689","url":null,"abstract":"<div><div>Analyzing children’s agency is crucial for understanding the inner workings of households. This issue is especially salient for developing countries, as childless families constitute a small portion of the population. This study tests the intra-household decision-making power of work-eligible children (age 12–17) in rural Bangladesh using a restriction implied by the collective household model. By comparing households with working and non-working children, the study also tests whether working children have a greater say in their households than non-working ones. The findings suggest some evidence for children’s decision-making power. The unitary household model is strongly rejected, and there is no evidence against Pareto efficiency. The results are particularly useful for the growing body of research that estimates intra-household inequality and individual-level poverty in developing countries using the collective model with certain assumptions on children’s decision-making power. In addition to investigating children’s say in the household, the study addresses several methodological issues in the literature regarding the price-based test of alternative collective models and Pareto efficiency.</div></div>","PeriodicalId":48418,"journal":{"name":"Journal of Development Economics","volume":"179 ","pages":"Article 103689"},"PeriodicalIF":4.6,"publicationDate":"2025-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145693010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-24DOI: 10.1016/j.jdeveco.2025.103691
Zhuo Zheng , Timothy Wu , Richard Lee , David Newhouse , Talip Kilic , Marshall Burke , Stefano Ermon , David B. Lobell
Accurate and comprehensive measurement of household livelihoods is critical for monitoring progress towards poverty alleviation and targeting social assistance programs for those who most need it. However, the high cost of traditional data collection has historically made comprehensive measurement a difficult task in many locations. This paper evaluates alternative satellite-based deep learning approaches to local-level livelihoods measurement, using detailed and multi-year household census extracts from four African countries as training and evaluation data. We show that new machine-learning architectures based on transformers solve multiple open measurement problems, including high performance on measuring changes in livelihoods over time and accurate measurement of local-level variation in household asset wealth within cities, offering general improvement over previous benchmarks especially when training datasets are large. Experiments that artificially restrict data availability show that satellite-based models can make accurate predictions with limited training data. The proposed approach demonstrates the promise of combining satellite imagery and new deep learning architectures for hyperlocal and dynamic measurement of poverty in data-scarce environments.
{"title":"Dynamic, high-resolution poverty measurement in data-scarce environments","authors":"Zhuo Zheng , Timothy Wu , Richard Lee , David Newhouse , Talip Kilic , Marshall Burke , Stefano Ermon , David B. Lobell","doi":"10.1016/j.jdeveco.2025.103691","DOIUrl":"10.1016/j.jdeveco.2025.103691","url":null,"abstract":"<div><div>Accurate and comprehensive measurement of household livelihoods is critical for monitoring progress towards poverty alleviation and targeting social assistance programs for those who most need it. However, the high cost of traditional data collection has historically made comprehensive measurement a difficult task in many locations. This paper evaluates alternative satellite-based deep learning approaches to local-level livelihoods measurement, using detailed and multi-year household census extracts from four African countries as training and evaluation data. We show that new machine-learning architectures based on transformers solve multiple open measurement problems, including high performance on measuring changes in livelihoods over time and accurate measurement of local-level variation in household asset wealth within cities, offering general improvement over previous benchmarks especially when training datasets are large. Experiments that artificially restrict data availability show that satellite-based models can make accurate predictions with limited training data. The proposed approach demonstrates the promise of combining satellite imagery and new deep learning architectures for hyperlocal and dynamic measurement of poverty in data-scarce environments.</div></div>","PeriodicalId":48418,"journal":{"name":"Journal of Development Economics","volume":"179 ","pages":"Article 103691"},"PeriodicalIF":4.6,"publicationDate":"2025-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145623669","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-21DOI: 10.1016/j.jdeveco.2025.103690
Mahesh Karra , Kexin Zhang
We test how two user-centered approaches to counseling shape women’s contraceptive preferences and behavior: (1) tailored counseling that presents contraceptive methods based on women’s stated preferences; and (2) prompting women with the choice to invite male partners to counseling. A total of 782 women were randomized to receive tailored or standard counseling, cross-randomized with the prompt to invite their partners. Following counseling, women were offered free transport and access to a family planning clinic for one month. Women who received tailored counseling were less likely to be concordant between their stated preferred method and method use, both intertemporally and contemporaneously. Women who were prompted with partner invitations were less likely to change their stated preferred method but more likely to be concordant between their stated preferred method following counseling and method use at follow-up. While both approaches aimed to facilitate user-centered contraceptive decision-making, neither necessarily yielded strictly preferred outcomes for women.
{"title":"User-centered approaches to contraceptive counseling: Experimental evidence from urban Malawi","authors":"Mahesh Karra , Kexin Zhang","doi":"10.1016/j.jdeveco.2025.103690","DOIUrl":"10.1016/j.jdeveco.2025.103690","url":null,"abstract":"<div><div>We test how two user-centered approaches to counseling shape women’s contraceptive preferences and behavior: (1) tailored counseling that presents contraceptive methods based on women’s stated preferences; and (2) prompting women with the choice to invite male partners to counseling. A total of 782 women were randomized to receive tailored or standard counseling, cross-randomized with the prompt to invite their partners. Following counseling, women were offered free transport and access to a family planning clinic for one month. Women who received tailored counseling were less likely to be concordant between their stated preferred method and method use, both intertemporally and contemporaneously. Women who were prompted with partner invitations were less likely to change their stated preferred method but more likely to be concordant between their stated preferred method following counseling and method use at follow-up. While both approaches aimed to facilitate user-centered contraceptive decision-making, neither necessarily yielded strictly preferred outcomes for women.</div></div>","PeriodicalId":48418,"journal":{"name":"Journal of Development Economics","volume":"179 ","pages":"Article 103690"},"PeriodicalIF":4.6,"publicationDate":"2025-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145623673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-21DOI: 10.1016/j.jdeveco.2025.103655
Johanne Pelletier , Mira Korb , Solomon Alemu , Manex B. Yonis , Travis J. Lybbert , Matthieu Stigler
Recent advances in Earth observation and machine learning open new frontiers in impact evaluation that appear well-suited for agricultural settings. We probe the nuances and limitations of these promising methods using satellite-predicted maize yields to measure the impact of Ethiopia’s Direct Seed Marketing (DSM) program, which was rolled out after 2011 to enhance farmer access to improved seed varieties. Machine learning-generated outcomes inevitably introduce prediction error, which attenuates coefficients and understates standard errors in downstream causal analysis. We find positive effects of DSM on crop cut maize yields, but weaker effects when using (naive) predicted yields. We demonstrate how ground-truth data can be leveraged to diagnose and correct this bias using tools from the prediction-powered inference literature — albeit with additional assumptions about unobservable prediction errors. Our corrected estimates suggest substantial DSM impacts on satellite-predicted maize yields in all producing areas and years.
{"title":"Causal Inference with Predicted Outcomes: Correcting prediction error bias in satellite-based impact evaluation","authors":"Johanne Pelletier , Mira Korb , Solomon Alemu , Manex B. Yonis , Travis J. Lybbert , Matthieu Stigler","doi":"10.1016/j.jdeveco.2025.103655","DOIUrl":"10.1016/j.jdeveco.2025.103655","url":null,"abstract":"<div><div>Recent advances in Earth observation and machine learning open new frontiers in impact evaluation that appear well-suited for agricultural settings. We probe the nuances and limitations of these promising methods using satellite-predicted maize yields to measure the impact of Ethiopia’s Direct Seed Marketing (DSM) program, which was rolled out after 2011 to enhance farmer access to improved seed varieties. Machine learning-generated outcomes inevitably introduce prediction error, which attenuates coefficients and understates standard errors in downstream causal analysis. We find positive effects of DSM on crop cut maize yields, but weaker effects when using (naive) predicted yields. We demonstrate how ground-truth data can be leveraged to diagnose and correct this bias using tools from the prediction-powered inference literature — albeit with additional assumptions about unobservable prediction errors. Our corrected estimates suggest substantial DSM impacts on satellite-predicted maize yields in all producing areas and years.</div></div>","PeriodicalId":48418,"journal":{"name":"Journal of Development Economics","volume":"179 ","pages":"Article 103655"},"PeriodicalIF":4.6,"publicationDate":"2025-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145623671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号