|
on Development |
| By: | de la Fuente Stevens, Diego (University of Sussex); Marchais, Gauthier (Institute of Development Studies); Brandt, Cyril (Institute of Development Studies); Holland-Szyp, Carolina (Institute of Development Studies); Matabishi, Samuel (Institut Supérieur Pédagogique de Bukavu) |
| Abstract: | Why do teachers remain in schools affected by prolonged crisis? Using longitudinal data on 1, 831 teachers across 124 schools in Niger and the Democratic Republic of the Congo, we examine the drivers of school-level teacher retention over three survey rounds. We find that retention is shaped primarily by local embeddedness, pay, and exposure to violence. Higher salaries are associated with retention, but almost entirely among teachers who have not always lived in the village where they teach. Pay and local embeddedness therefore operate as alternative rather than cumulative retention margins. Local embeddedness is among the strongest correlates of retention, but mainly for male teachers. Exposure to violence is the strongest predictor of exit, with effects roughly twice as large for women as for men. These findings highlight the relational, material and gendered dimensions of teacher retention in conflict-affected settings. |
| Keywords: | Sub-Saharan Africa, Job turnover, Teacher retention, Conflict-affected education, Violence exposure, Gender |
| JEL: | I25 J45 D74 |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp18917 |
| By: | Hoy, Christopher Alexander; Strehl-Pessina, Matias; Doino, Ruggero; Naidoo, Darian; Mambon, Kingtau; Kunda, Bobby |
| Abstract: | Tax exemptions for basic items exist worldwide, typically justified by governments as a way to reduce the cost of living for poorer households. This paper studies the unanticipated removal of consumption taxes on selected food items in Papua New Guinea. It conducts a difference-in-differences analysis using administrative data, a supermarket price census, web-scraped online prices, and a nationally representative household phone survey panel. The findings show complete pass-through in formal supermarkets in central urban districts, where retail competition is strong, but no pass-through in informal stores or rural areas, where poorer households mainly shop. Only 16 percent of the foregone revenue accrued to the poorest two quintiles, while the richest two quintiles and stores/wholesalers each captured almost 40 percent of the benefits. This degree of regressivity and variation in pass-through was unanticipated by 237 experts who participated in a prediction survey. Using a static marginal value of public funds–style comparison, the paper shows that universal cash transfers would generate roughly two and a half times as much quintile-weighted value per fiscal dollar as the tax exemption. The findings demonstrate that food tax exemptions are inherently regressive in lower-income settings, due to high informality, substantial market segmentation, and positive income elasticity of demand for basic food. |
| Date: | 2026–09–16 |
| URL: | https://d.repec.org/n?u=RePEc:wbk:wbrwps:11453 |
| By: | Tara Bedi (Trinity College Dublin and TIME); Anu Jose (University of Galway); Michael King (Trinity College Dublin and TIME); Samuel McArdle (Economic and Social Research Institute (ESRI)) |
| Abstract: | Natural disasters impose high and lasting psychological costs, yet the distribution of these costs within households remains poorly understood. Using a unique panel dataset of couples in Malawi, this paper examines the effect of Cyclone Idai damage on couples' mental health. We find that female spouses experience a 5.65% erosion in self‐efficacy five to seven months after the cyclone, while their husbands show no significant change. We find no significant effects on depression or stress‐related disability for either spouse. To explain this gender divergence in self‐efficacy, we investigate differential access to labour market coping strategies. Following severe cyclone‐related losses, husbands are 14.5% more likely to engage in wage employment and report 26.8% higher wage earnings relative to husbands in low‐loss households, whereas wives' participation in income‐generating activities remains unchanged. Together, these findings show that the psychological effects of the disaster are gender‐differentiated within the household, and differential labour market opportunities may explain the divergent psychological outcomes observed between spouses. These patterns have implications for the design of disaster response and social protection policies. |
| Keywords: | Natural Disasters, Mental Health, Intrahousehold, Gender, Labour |
| JEL: | D13 D15 Q54 I10 I30 I32 |
| Date: | 2026–08 |
| URL: | https://d.repec.org/n?u=RePEc:tcd:tcduee:tep1726 |
| By: | Gasior, Katrin; Tasseva, Iva; Wright, Gemma |
| Abstract: | The effectiveness of social protection during economic shocks depends on two types of benefit: non-shock-responsive benefits, fixed prior to crisis, and automatic stabilisers, which adjust with income or employment losses. We analyse effectiveness in seven Sub-Saharan African countries using tax-benefit microsimulation models and household survey data. Simulating employment losses, we apply a decomposition framework to isolate the role of each benefit type. We find that high pre-crisis coverage of non-shock-responsive benefits in Ghana and Zambia and income related automatic stabilisers in South Africa provide an income floor for poor households. South Africa’s unemployment insurance further mitigates losses among better-off households. In con trast, limited protection is available in Mozambique, Rwanda, Tanzania, and Uganda, where benefits are modest, coverage gaps substantial, and automatic stabilisers ineffective due to ab sent unemployment insurance, reliance on proxy means-tests, and narrow eligibility. |
| Keywords: | social protection benefits;non-shock-responsive benefits;automatic stabilisers;economic shocks;poverty |
| JEL: | H55 D31 |
| Date: | 2025–10–01 |
| URL: | https://d.repec.org/n?u=RePEc:ehl:lserod:140838 |
| By: | Akren, Mehmet B.; Ertan, Arhan S.; Putterman, Louis G. |
| Abstract: | The regional distribution of world population underwent profound changes during the age of exploration and colonization, in which a handful of European powers colonized the better part of the remaining continents. One consequence is that many countries in the Americas, Australia and New Zealand, after sharp contractions of indigenous populations following exposure to Old World diseases, today have populations in which European or African descendants form majorities and speak predominantly European languages. Migration and in some cases colonization also altered population compositions in countries including South Africa, Cape Verde, Fiji and Singapore. And descendants of 'emigr'es came to outnumber those of "stayers" even for non-colonizing countries, including, e.g., Ireland and Lebanon. We identify major determinants of who settled where, during this era of colonialism and migration, demonstrating that both "natural" factors like availability and fertility of agricultural land and climate and disease prevalence, on the one hand, and power inequalities and cultural variables, on the other, explain the broad patterns of migration. |
| Abstract: | Die regionale Verteilung der Weltbevölkerung erfuhr im Zeitalter der Entdeckungen und der Kolonialisierung tiefgreifende Veränderungen, in dem eine Handvoll europäischer Mächte den größten Teil der übrigen Kontinente kolonisierten. Eine Folge davon ist, dass in vielen Ländern Amerikas, Australiens und Neuseelands - nach einem drastischen Rückgang der indigenen Bevölkerung infolge der Ansteckung mit Krankheiten aus der Alten Welt - heute die Nachfahren europäischer oder afrikanischer Einwanderer die Mehrheit bilden und überwiegend europäische Sprachen sprechen. Migration und in einigen Fällen auch Kolonisierung veränderten zudem die Bevölkerungszusammensetzung in Ländern wie Südafrika, Kap Verde, Fidschi und Singapur. Und selbst in Ländern, die keine Kolonialmächte waren - darunter beispielsweise Irland und der Libanon -, überstieg die Zahl der Nachkommen von Auswanderern schließlich die der "Daheimgebliebenen". Wir identifizieren die wichtigsten Determinanten dafür, wer sich in dieser Ära des Kolonialismus und der Migration wo niedergelassen hat, und zeigen, dass sowohl "natürliche" Faktoren wie die Verfügbarkeit und Fruchtbarkeit von Ackerland sowie das Klima und die Krankheitsprävalenz einerseits als auch Machtungleichheiten und kulturelle Variablen andererseits die allgemeinen Migrationsmuster erklären. |
| Keywords: | International Migration, Columbian Exchange, Historical Demography, Power Asymmetries, Colonialism, Disease, Land Endowments |
| JEL: | F22 N30 N50 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:rwirep:343572 |
| By: | Makaaru Arinaitwe, Jacklyn (Ace Policy Research Institute); Bogere, George (Ace Policy Research Institute); Fiore, Anna (World Bank); Herskowitz, Sylvan (World Bank); Lang, Megan (World Bank); Mwesigye, Francis (Uganda Development Bank); Nalumansi, Tracy (Nova SBE) |
| Abstract: | Childcare burdens limit productivity among women in the informal sector, but we know little about the optimal design of childcare support. We randomize 883 women market vendors in Kampala, Uganda to receive free workplace- or community-based childcare. Free childcare redistributes caregiving away from household members, raises women’s hourly earnings, and improves child development and women’s well-being. Community centers achieve higher utilization and larger gains in development and well-being, while workplace centers generate larger economic impacts. Both models have a high marginal value of public funds. Community centers currently dominate, but workplace-based centers could prove superior if operated at scale. |
| Keywords: | labor, childcare, gender, time use, early child development |
| JEL: | O12 O18 J13 J22 |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp18910 |
| By: | Vadlamannati, Krishna; Iacono, Roberto (Norwegian University of Science and Technology) |
| Abstract: | The question of whether aid supplements or substitutes government spending has long been debated in public finance literature. Building on Vadlamannati and Iacono (2026), who examine how development aid affects discretionary fiscal transfers to Indian states, we address a distinct fiscal question: how does aid exposure affect the composition of subnational government expenditure? Using panel data for 30 Indian states over 1979-2019 (41 years) and an instrumental variable strategy addressing potential endogeneity, we find that greater exposure to aid is associated with a lower share of capital expenditure and a higher share of revenue expenditure. This pattern is consistent with substitution of externally aided projects for domestically financed spending. The effect however varies systematically with electoral incentives. In states with greater electoral competition, aid is more strongly associated with capital expenditure, suggesting that electoral incentives can encourage state governments to channel external resources towards visible capital investments. Our findings extend the evidence on fiscal consequences of aid in India from the allocation of intergovernmental discretionary transfers to the composition of state government expenditure. Our findings reveal that the developmental effects of aid depend not only on how resources reach subnational governments but also on how governments subsequently allocate them. |
| Date: | 2026–09–23 |
| URL: | https://d.repec.org/n?u=RePEc:osf:socarx:dahwz_v1 |
| By: | Pechar, Stefanie; Hartwig, Renate; Priebe, Jan |
| Abstract: | We study how women's political representation affects fertility in sub-Saharan Africa. Leveraging the staggered adoption of reserved-seat gender quotas across the region and a heterogeneity-robust difference-in-differences estimator, we find that quotas increased women's parliamentary representation by 13.3 percentage points and reduced the total fertility rate by 0.23 births per woman. The effect follows a policy-to-behavior pathway: resources shift toward family planning and reproductive health, contraceptive use increases, and unmet need declines; individual-level data further show delayed marriage and later first births. Women's political representation thus translates into greater policy emphasis on reproductive health, illustrating how legislative representation shapes policy outcomes. |
| Abstract: | Wir untersuchen, wie sich die politische Vertretung von Frauen auf die Fertilität in Subsahara-Afrika auswirkt. Unter Berücksichtigung der schrittweisen Einführung von Geschlechterquoten mit reservierten Sitzen in der gesamten Region und unter Verwendung eines gegenüber Heterogenität robusten Differenz-Differenzen-Schätzers stellen wir fest, dass die Quoten die parlamentarische Vertretung von Frauen um 13, 3 Prozentpunkte erhöht und die Gesamtfruchtbarkeitsrate um 0, 23 Geburten pro Frau gesenkt haben. Der Effekt folgt einem "Policy-to-Behavior"-Pfad: Ressourcen werden in Richtung Familienplanung und reproduktive Gesundheit umgeschichtet, die Verwendung von Verhütungsmitteln nimmt zu und der ungedeckte Bedarf sinkt; Daten auf individueller Ebene zeigen zudem eine spätere Heirat und spätere Erstgeburten. Die politische Vertretung von Frauen führt somit zu einer stärkeren politischen Fokussierung auf reproduktive Gesundheit und veranschaulicht, wie die Vertretung im Parlament politische Ergebnisse prägt. |
| Keywords: | Family Planning, Female Political Representation, Fertility, Gender Quotas, Reserved Seats, Sub-Saharan Africa |
| JEL: | J13 J16 D72 I18 O15 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:rwirep:343542 |
| By: | Fall, Fatou |
| Abstract: | Instability and ongoing conflict across the Sahel have led to conflict-induced migration and the displacement of populations across borders. As host countries face the challenges associated with refugee inflows, it becomes increasingly important to understand how these movements affect the host communities that are indirectly impacted by the conflict. By combining geocoded household panel data from Niger with administrative data on refugee settlements, I exploit spatial variation in exposure to refugee camps to assess how refugee exposure affects labor market outcomes, labor on household plots, and household consumption among households in areas hosting refugees. A higher exposure to refugees increases the likelihood that households diversify into secondary and non-farm activities, decreases family labor on plot, and improves the Food Consumption Score during the harvest season. Collectively, these results suggest that refugee inflows can generate positive local economic effects in rural settings, potentially through increased demand and humanitarian aid. |
| Abstract: | Die Instabilität und die anhaltenden Konflikte in der Sahelzone haben zu konfliktbedingter Migration und zur grenzüberschreitenden Vertreibung von Bevölkerungsgruppen geführt. Angesichts der Herausforderungen, denen sich die Aufnahmeländer im Zusammenhang mit dem Zustrom von Flüchtlingen gegenübersehen, wird es immer wichtiger zu verstehen, wie sich diese Migrationsbewegungen auf die Aufnahmegemeinschaften auswirken, die indirekt vom Konflikt betroffen sind. Durch die Kombination von geokodierten Haushaltspaneldaten aus dem Niger mit Verwaltungsdaten zu Flüchtlingssiedlungen nutze ich räumliche Unterschiede in der Nähe zu Flüchtlingslagern, um zu bewerten, wie sich die Nähe zu Flüchtlingen auf die Arbeitsmarktergebnisse, die Arbeit auf den Hausgrundstücken und den Konsum der Haushalte in Gebieten auswirkt, die Flüchtlinge aufnehmen. Eine größere Nähe zu Flüchtlingen erhöht die Wahrscheinlichkeit, dass Haushalte ihre Tätigkeit auf sekundäre und nichtlandwirtschaftliche Aktivitäten ausweiten, verringert die Familienarbeit auf dem Grundstück und verbessert den Nahrungsmittelverbrauchsindex während der Erntezeit. Insgesamt deuten diese Ergebnisse darauf hin, dass Flüchtlingszuströme positive lokale wirtschaftliche Effekte in ländlichen Gebieten hervorrufen können, möglicherweise durch eine gestiegene Nachfrage und humanitäre Hilfe. |
| Keywords: | Refugee impacts, Labor markets, Consumption, Rural households, Niger |
| JEL: | J43 O12 O15 R23 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:rwirep:343539 |
| By: | Mogge, Lukas; Roeckert, Julian; Rukundo, Emmanuel Nshakira; Krähnert, Kati |
| Abstract: | We study stated and realized demand for a new parametric insurance product against excessive rainfall among rice farmers in Vietnam's Mekong Delta, a region highly exposed to climate change. Using a household survey of 3, 000 rice-farming households, we elicit willingness to pay (WTP) using a multiple-bounded dichotomous choice approach. Stated demand is substantial: 33% of respondents report a WTP of at least half the market price, and 18% report a WTP equal to or higher than the market price. Stated demand is significantly higher among men, younger respondents, and respondents reporting better self-assessed product understanding. Yet when insurance agents offered the product at a subsidized price below respondents' stated WTP, no respondent purchased it. We discuss mechanisms that may explain this discrepancy between stated and realized demand, including limited incentive compatibility of the WTP elicitation mechanism, changes in valuations, and contextual barriers to insurance adoption. |
| Abstract: | Wir untersuchen die geäußerte und die tatsächliche Nachfrage nach einem neuen parametrischen Versicherungsprodukt gegen übermäßige Niederschläge bei reisanbauenden Haushalten im vietnamesischen Mekong-Delta - einer Region, die stark vom Klimawandel betroffen ist. Auf der Grundlage einer Befragung von 3.000 reisanbauenden Haushalten ermitteln wir die Zahlungsbereitschaft mittels eines "Multiple-Bounded Dichotomous Choice"-Verfahrens. Die geäußerte Nachfrage ist substanziell: 33 % der Befragten geben eine Zahlungsbereitschaft von mindestens der Hälfte des Marktpreises an, und 18 % nennen einen Betrag, der dem Marktpreis entspricht oder darüber liegt. Die geäußerte Nachfrage ist bei männlichen Befragten, jüngeren Befragten sowie Befragten, die ihr eigenes Produktverständnis höher einschätzen, signifikant höher. Als Versicherungsvertreter das Produkt jedoch zu einem subventionierten Preis anboten, der unter der angegebenen Zahlungsbereitschaft der Befragten lag, erwarb kein einziger Teilnehmer das Versicherungsprodukt. Wir erörtern Mechanismen, die diese Diskrepanz zwischen geäußerter und tatsächlicher Nachfrage erklären könnten, darunter die eingeschränkte Anreizkompatibilität des Verfahrens zur Ermittlung der Zahlungsbereitschaft, Veränderungen der Bewertung sowie kontextuelle Hindernisse für den Abschluss der Versicherung. |
| Keywords: | Agriculture, parametric weather insurance, Vietnam, willingness-to-pay |
| JEL: | G52 Q14 Q12 O12 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:rwirep:343540 |