Nationally-representative Household Survey Research Articles

A decompositional analysis of social group inequality in India

ABSTRACT Using the national representative Indian household surveys collected by the National Sample Survey Organization, we examine the welfare gaps across social groups for the entire distribution in 1983, 1993–94, 2004–05 and 2011–12. We use spatially adjusted per capita consumption expenditure as the measure of welfare and show that there exists significant welfare gap between Scheduled Tribes/Scheduled Castes (ST/SC) and General Category (GC) households at higher quantiles of the distribution, but the magnitude of the gap has declined over the years for the SC households. Using unconditional quantile regression decomposition, we find that the coefficient effect (unexplained effect) dominates the endowment effect (explained effect) in explaining the gap in all four years. Further decomposition shows that the difference in educational distribution between the SC/ST and GC households drives most of the explained differences between the social groups.

Contraceptive use by number of living children in Ghana: Evidence from the 2017 maternal health survey

Background There is a significant literature describing the link between parity and contraceptive use. However, there is limited knowledge about the disaggregation by parity of the type of contraceptives. In this study, we describe the use of contraceptives by parity among women of reproductive age in Ghana, focusing on use of highly effective methods (injection, pill, intrauterine device, implant, and sterilization). Methods Using the 2017 Ghana Maternal Health Survey, a nationally-representative cross-sectional household survey, we describe contraceptive method use by number of living children among sexually active women of reproductive age. We then estimated predictors of use of highly effective contraception in a multilevel logistic regression model. Results Most women in this survey are not using any method of contraception, although this varies by whether or not they have begun childbearing. Contraceptive method use varies by number of living children. Before having children, natural (periodic abstinence and withdrawal) and episodic (condoms) methods dominate. Once a woman has one living child, method preference changes to injectables and implants. Factors associated with using a highly effective method of contraception are: having >3 children, being in a relationship, having had an abortion, being younger than age 30, and having had sexual intercourse within days of answering the survey (p < .001 for all). Conclusion In this analysis, the number of living children a woman has, her age, and timing of last intercourse are the most significant predictors of using a highly effective method of contraception. However, the majority of participants in this study report not using any method of contraception to avoid unwanted pregnancies. Future research that attempts to unpack the disconnect between not wanting to become pregnant and not using contraception is warranted.

One for all and all for one: Increasing the adaptive capacity of households and communities through a public work programme

This article uncovers the mechanisms shaping the impact of the public work component of Ethiopia’s Productive Safety Net Programme (PSNP) on beneficiaries and communities’ food security and vulnerability to various shocks. Using three waves of a national representative household survey, this study provides quantitative evidence on the pathways through which the social protection intervention affects direct beneficiaries and their community peers. The empirical findings show that the PSNP beneficiaries are less likely to be food insecure and to experience harvest losses in the aftermath of droughts. Notably, the beneficial effects of the programme partially spill over to the direct participants’ community peers. This is likely to be due the nature of the public works implemented through the programme, such as the integrated community-based watershed development, including soil and water conservation measures and rangeland management in pastoral areas. Additionally, no significant impacts have been found when households self-report stresses unrelated to droughts. Our findings are expected to inform the debate on the effectiveness of the PSNP and other adaptive social protection programmes. From a policy perspective, they suggest the explicit integration of environmental and climate considerations into the design of social protection interventions targeting poor agricultural households highly vulnerable and exposed to weather shocks.

Livelihood Diversification and Household Vulnerability to Climate Shocks in Rural Kenya

This study tests the hypothesis that livelihood diversification mitigates vulnerability to climate shocks among rural households in Kenya. The study is based on two waves of nationally-representative household survey data collected in 2005/06 and 2015/16. A measure of livelihood diversification in the sampled households was computed using Simpson Diversity Index. Comparison of means and correlation analysis was used to test the study’s hypothesis on a sample of 8,483 and 12,217 rural households in 2005/06 and 2015/16 respectively. Consistent with the study’s hypothesis, results indicate lower mean values of livelihood diversification indices for rural households that reported loss in welfare due to climate shocks compared to those that reported no adverse effects in both 2005/06 and 2015/16. Correlation analysis results indicate an inverse relationship between livelihood diversification and vulnerability to climate shocks for all sampled households. Distinctive and nuanced differences in results were found when analysis was disaggregated along income classes and agro-ecological zones. Sensitivity analysis confirmed robustness of the results. Based on these results, it is recommended that rural households be supported to pursue a diverse portfolio of income generating activities and assets in order to build resilience against climate shocks.

Maize price variability, land use change, and forest loss: evidence from Ghana

This paper documents the relationship between maize price variability and forest loss in Ghana. Using the fixed effects regression technique while controlling for weather variables and accounting for spatial dependence, the study finds that an increase in maize price variability is associated with an increase in forest loss. Drawing on nationally-representative panel household survey data, the analysis further reveals that an increase in maize price variability contributes to forest loss through changes in household land use decisions. The finding indicates that an increase in maize price variability in the previous year positively correlates with a household’s total cultivated farm size. This increase in cropped area is primarily due to a rise in the household cultivated area of cocoa – an export crop with a guaranteed price set annually by the government. The study highlights an important linkage between agricultural price stabilization policies and environmental externalities.

Shifting priorities and employment choices among workers in Wales during the COVID-19 pandemic: a longitudinal analysis

BackgroundEmployment is a determinant of health. The COVID-19 pandemic disrupted working lives, forcing individuals to adapt to new ways of working. These shifts might shape people's priorities and their consideration of changes for future work. We examined how these outcomes differed depending on self-reported health status. MethodsIn this longitudinal analysis, we used data from the COVID-19 Employment and Health in Wales Study; a nationally-representative household survey of workers aged 18–64 years. Timepoint 1 (T1) data were collected between May 27, 2020, and June 22, 2020, and timepoint 2 (T2) data between Nov 30, 2020, and Jan 29, 2021. Participants who responded at both timepoints were eligible. Respondents selected five employment priorities at both timepoints, and the employment changes they considered during the COVID-19 pandemic at T2 only. We used multivariable logistic regressions (including sociodemographics, current employment factors, and self-reported health) and examined associations with health firstly for employment priorities, and secondly for the consideration of employment changes. Health measures were self-reported general health, limiting pre-existing health conditions (both using National Survey for Wales validated questions), and mental wellbeing (using the shortened Warwick Edinburgh Mental Well-being Scale). FindingsWe analysed data from 592 respondents (382 [65%] women). 766 (56%) of 1358 T1 respondents were excluded as no T2 responses were provided. Those who self-reported poor general health were consistently more likely to prioritise flexible working arrangements than those rating fair or above (T1 adjusted odds ratio [aOR] 2·06 [95% CI 1·10–3·88], p=0·033; T2 aOR 1·87 [95% CI 1·05–3·33], p=0·034). Those with low (as opposed to average) mental wellbeing were more likely to consider securing a permanent contract (aOR 5·49 [95% CI 1·32–22·81], p=0·023, and those with limiting pre-existing conditions were four times more likely to consider becoming self-employed (aOR 4·00 [95% CI 1·35–11·84], p=0·011) than those without. InterpretationPromoting the adoption of flexible working policies and supporting those in poor health to obtain flexible employment could benefit people in Wales. Those with low mental wellbeing might seek the security of permanent employment, and those with limiting pre-existing conditions might value the autonomy of self-employment. FundingNone.

Intersection of alcohol use, HIV infection, and the HIV care continuum in Zambia: nationally representative survey

ABSTRACT Through a nationally-representative household survey, we measured the prevalence and correlates of unhealthy alcohol use (UAU) in Zambia and its association with the HIV care continuum. Adolescent and adult (ages 15–59 years) data, including the Alcohol Use Disorders Identification Test-Consumption (AUDIT-C), from the 2016 Zambia Population-based HIV Impact Assessment, were analyzed. UAU was defined as AUDIT-C of 3 + points for women and 4 + for men. Among 20,923 participants, 15.3% had UAU; this was 21.6% among people living with HIV (PLWH). Male sex, increasing age, being employed, urban residence, and having HIV were independent correlates of UAU (all P < 0.05). Among PLWH, UAU was associated with reduced HIV diagnosis (adjusted odds ratio [AOR]: 0.66, 95% CI 0.50–0.88) and non-significant trends toward reduced ART use if diagnosed (AOR: 0.73, 95% CI 0.73–1.10) and reduced viral suppression (VS) if on ART (AOR: 0.91, 95% CI 0.57–1.44). Overall, UAU was linked to 25% lower odds of VS compared to abstinence. UAU in Zambia disproportionately affects certain groups including PLWH. Achieving and sustaining HIV epidemic control in Zambia will require evidence-based approaches to screen and treat UAU.

Trends and country-level variation in age at first sex in sub-Saharan Africa among birth cohorts entering adulthood between 1985 and 2020

BackgroundDebuting sexual intercourse marks exposure to pregnancy or fatherhood and sexually transmitted infections (STIs), including HIV. In sub-Saharan Africa (SSA), sexual debut varies according to cultural, religious, and economic factors, and encouraging delay has been a longstanding component of behavioural HIV prevention strategies. Age at first sex (AFS) is routinely collected in national household surveys, but data are affected by reporting biases, limiting utility to monitor trends and guide sexual health interventions.MethodsWe collated individual-level data from nationally-representative household surveys to analyse timing and national trends in AFS in 42 SSA countries. We used a log-skew-logistic distribution to characterize the time to AFS in a Bayesian spatio-temporal model, providing estimates of the sexual debut rate by sex, age, time, and country. We statistically adjusted for reporting biases by comparing AFS reported by the same birth cohorts in multiple survey rounds, allowing different reporting biases by sex and country.ResultsMedian AFS in 2015 ranged from 15.8 among Angolan women to 25.3 among men in Niger. AFS was younger for women than men in 37/40 countries. The gap was largest for Sahel region countries and minimal in southern African countries. The distribution of female AFS was asymmetric with half debuting sex in an age range of 3.9 years [IQR 3.4–5.0 across countries]. Median AFS increased slightly between 1985 and 2020, ranging 0.84 years [IQR 0.11–1.55] and 0.79 [IQR -0.23–1.98] for females and males, respectively. The gender gap changed little over time in most countries. Female teens often reported higher AFS compared to when asked in their late twenties while male teens reported lower AFS; both sexes recalled a higher AFS in older ages compared to their thirties.ConclusionsAFS increased slightly in most SSA countries, but changes were modest relative to large and persistent variation between countries and sexes, indicating relatively entrenched socio-cultural practices around sexual debut. Sexual health, family planning, and HIV/STI prevention services should adapt to local practices rather than focusing interventions to change AFS. These estimates for rates of sexual debut provide data to guide programmatic prioritization and implementation of sexual health services.

Delays in Children's Preventive Health Services During the COVID-19 Pandemic.

Stay-at-home orders, social isolation recommendations, and fear of COVID-19 exposure have led to delays in children's preventive health services during the pandemic. Delays can lead to missed opportunities for early screening and detection of health problems, and increased risks for outbreaks of vaccine-preventable diseases. Understanding prevalence of and reasons for missed, delayed, or skipped preventive health services is important for developing strategies to achieve rapid catch-up of essential health services. Using the Household Pulse Survey (n=37,064), a large, nationally-representative household survey fielded from April 14 to May 10, 2021, we examined prevalence of households with children who have missed, delayed, or skipped preventive health services, and factors associated with and reasons contributing to missed, delayed, or skipped preventive health services. About one-quarter of parents had children who missed, delayed, or skipped preventive check-ups in the past year. Delays in children's preventive health services were more common among respondents with higher education, households with greater numbers of children, and children who learned remotely or did not participate in formal education. Main reasons attributed to delayed preventive health services were limited appointments at health providers' offices (42.9%), concern about COVID-19 exposure at health providers' offices (42.2%), and closed health providers' offices due to the pandemic (29.0%). Physician office closures and concern about COVID-19 exposure resulted in over one-quarter of parents delaying preventive services for their children since the pandemic began. Coordinated efforts are needed to achieve rapid catch-up of preventive services and routine vaccines.

Person-centered care in Uganda: analysis of responsiveness, patient satisfaction, patient-reported health outcomes, and trust among adults

Abstract Objectives Person-centeredness and patient experience are key components of high-quality health systems. The World Health Organization defines seven domains of health system responsiveness, and there is a growing body of research that uses these domains as a proxy for experiential quality. Methods We examined overall levels of responsiveness across the domains and across demographic and visit characteristics in Uganda. We also explored the relationship between responsiveness and patient satisfaction, self-rated health outcomes, and trust. This study uses data from a nationally-representative cross-sectional household survey conducted in Uganda in 2019. Respondents who had sought care in the last 6 months were asked questions about responsiveness of care during their most recent visit. We compared demographic and visit characteristics in total and between respondents within the highest and lowest responsiveness and also explored the distribution of the responsiveness index across all responsiveness domains. Finally, we assessed the relationship between responsiveness quintiles and the highest rating of our outcomes using Poisson regression. Our six outcomes fell within three categories: patient satisfaction, patient-reported health outcomes, and trust. Results Of the 4823 men and women surveyed, 2924 (60.6%) visited a healthcare facility in the last 6 months, and the total weighted study sample of those who visited a healthcare facility was 2827. Respondents who reported high responsiveness were more likely to be wealthier, urban dwellings, and with higher levels of education, and more likely to have greater access to care, to see the same provider at each visit, and to have gone to a private facility. We found that respondents reporting higher responsiveness were significantly more likely to report the highest rating on all six outcomes. Conclusions Our findings suggest that responsiveness is a central component of high-quality health systems and that efforts to improve health outcomes must look beyond technical quality.

The comparison of catastrophic health expenditure and its inequality between urban and rural households in China

BackgroundIn recent years, the goal of universal coverage of the basic medical insurance schemes has been basically achieved in China, but the heavy economic burden of diseases is still the main cause of poverty in many households. Exploring catastrophic health expenditure (CHE) and its inequality are highly important for forward-looking policymaking. This study aims to compare the incidence, intensity and inequality of CHE between urban and rural households in China.MethodsThis study was based on a national representative household survey—the China Family Panel Studies (CFPS)—that was conducted from 2012 to 2018. Concentration index (CI) was employed to measure the inequality of CHE incidence and overshoot, while the decomposition method of the CI was used to estimate the main influencing factors affecting inequality of CHE incidence.ResultsFrom 2012 to 2018, the CHE incidence of urban households increased from 11.01 to 11.88%, while the CHE incidence of rural households decreased from 18.42 to 18.31%. During the same period, the CI of CHE incidence for urban households decreased from − 0.1480 to − 0.1693, while that for rural households declined from − 0.1062 to − 0.1501. The major contribution to the pro-poor inequality in CHE incidence was associated with socioeconomic status, lagged CHE, receiving inpatient services, having elderly members, education of household head, and self-assessed health status of household head.ConclusionsRural households had higher risk of incurring CHE than urban households. The strong pro-poor inequality for CHE incidence and overshoot could be found in both two groups. The problem of poverty due to illness was more severe among low-income groups in rural areas than in urban areas. The relevant policy interventions should further focus on encouraging the development of supplementary medical insurance and increasing the reimbursement rate for hospitalization expenses in the medical assistance system.

Exploring the health impacts and inequalities of the new way of working: findings from a cross-sectional study.

BackgroundEmployment is a wider determinant of health, and one of the most striking societal transformations witnessed during the SARS-CoV-2 (COVID-19) pandemic was the sudden shift to working from home (WFH). Within the COVID-19 recovery and efforts to tackle climate change, having large proportions of the workforce WFH has clear benefits. However, the wider adoption of such measures should account for inequalities in accessing home working, the workforce’s preferences for the future, and crucially, the impact that WFH might have on the workforce’s health.&#x0D; MethodsA nationally-representative household survey was undertaken across Wales (Public Health Wales’ COVID-19, Employment and Health in Wales study), with cross-sectional data on home-working being collected between November 2020 and January 2021 from 615 employed working-aged adults in Wales (63.7% female, 32.7% aged 50-59). Respondents were asked about their ability to WFH, their perceptions of its impact on their health and their preferences for time spent home-working in future.&#x0D; FindingsOver 50% were able to WFH, and showed a preference towards home-working to some capacity, with over a third wishing to WFH at least half the time. However, those living in the most deprived areas, in atypical employment, with high wage precarity or with limiting pre-existing conditions were less likely to report being able to WFH. Of those that could WFH, over 40% reported that it worsened their mental well-being and loneliness, and for people in poorer health, home-working negatively impacted their diet, physical activity, smoking and alcohol use. People aged 30 to 39 and those who lived alone were more likely to report wanting to spend some time working in an office/base instead of at home.&#x0D; ConclusionsThe new way of working introduces new challenges to preserving workforce mental well-being, regulating health behaviours and tackling inequalities. Our findings shine a light on the negative health impacts associated with WFH (particularly for those already in poorer health), highlight how disadvantaged population sub-groups face further inequalities through being less able to WFH, and provide novel observations of the preferred working models of different sub-groups within the population. Hybrid approaches, combining home and work-based employment are preferred and could alleviate the negative health impacts associated with WFH. Addressing inequalities across the workforce to ensure that everyone is as able to participate in this transformative shift in ways of working is key.

Financial vulnerability and the impact of COVID-19 on American households.

In May 2020, Westat, in partnership with Stanford University School of Medicine, conducted a nationally-representative household survey of American attitudes and behaviors regarding COVID-19. In this article, we examine what the Coronavirus Attitudes and Behaviors Survey tells us about the impact of COVID-19 on financial status and how this impact varies by demographic characteristics, the presence of health risk factors, and financial status (including employment factors). The survey reveals significant inequality in financial impact, as those who were most financially vulnerable prior to the pandemic found themselves under greater financial strain, while those who were more financially secure have experienced a neutral or even positive impact of the pandemic on household finances. These findings have important implications for public policy as policymakers seek to target aid to those who need it most.

Annual prevalence of unmet healthcare need in Thailand: evidence from national household surveys between 2011 and 2019

BackgroundExtending Universal Health Coverage (UHC) requires identifying and addressing unmet healthcare need and its causes to improve access to essential health services. Unmet need is a useful monitoring indicator to verify if low incidence of catastrophic health spending is not a result of foregone services due to unmet needs. This study assesses the trend, between 2011 and 2019, of prevalence and reasons of unmet healthcare need and identifies population groups who had unmet needs.MethodThe unmet healthcare need module in the Health and Welfare Survey (HWS) 2011–2019 was used for analysis. HWS is a nationally representative household survey conducted by the National Statistical Office biennially. There are more than 60,000 respondents in each round of survey. The Organisation for Economic Co-operation and Development (OECD) standard questions on unmet need and reasons behind were applied for outpatient (OP), inpatient (IP) and dental services in the past 12 months. Data from samples were weighted to represent the Thai population. Univariate analysis was applied to assess unmet need across socioeconomic profiles.ResultsThe annual prevalence of unmet need between 2011 and 2019 was lower than 3%. The prevalence was 1.3–1.6% for outpatient services, 0.9% - 1.1% for dental services, and lower than 0.2% for inpatient care. A small increasing trend was observed on dental service unmet need, from 0.9% in 2011 to 1.1% in 2019. The poor, the elderly and people living in urban areas had higher unmet needs than their counterparts. Long waiting times was the main reason for unmet need, while cost of treatment was not an issue.ConclusionThe low level of unmet need at less than 3% was lower than OECD average (28%), and was the result of UHC since 2002. Regular monitoring using the national representative household survey to estimate annual prevalence and reasons for unmet need can guide policy to sustain and improve access by certain population groups.

Predicting Livelihood Indicators from Community-Generated Street-Level Imagery

Major decisions from governments and other large organizations rely on measurements of the populace's well-being, but making such measurements at a broad scale is expensive and thus infrequent in much of the developing world. We propose an inexpensive, scalable, and interpretable approach to predict key livelihood indicators from public crowd-sourced street-level imagery. Such imagery can be cheaply collected and more frequently updated compared to traditional surveying methods, while containing plausibly relevant information for a range of livelihood indicators. We propose two approaches to learn from the street-level imagery: (1) a method that creates multi-household cluster representations by detecting informative objects and (2) a graph-based approach that captures the relationships between images. By visualizing what features are important to a model and how they are used, we can help end-user organizations understand the models and offer an alternate approach for index estimation that uses cheaply obtained roadway features. By comparing our results against ground data collected in nationally-representative household surveys, we demonstrate the performance of our approach in accurately predicting indicators of poverty, population, and health and its scalability by testing in two different countries, India and Kenya. Our code is available at https://github.com/sustainlab-group/mapillarygcn.

Relative importance of quantitative and qualitative aspects of urban green spaces in promoting health

The health benefits of urban green spaces (UGS) are important considerations in the design of urban environments. Researchers have analysed how the quantity of UGS impacts health outcomes, but less work has quantified the relative health benefits of different components of UGS provision. We examined the associations of different indicators of UGS provision with self-reported mental and general health based on a national representative household survey (n = 1000) in Singapore. We quantified three quantitative attributes of UGS (vegetation cover, canopy cover, park area) using Geographic Information System (GIS) data and measured visual greenness using the Green View Index (GVI) based on Google Street View panoramic images. We measured the subjective attributes of UGS, namely, usage quality and quantity of UGS within a 400 m circular buffer zone of residences using the survey. Among all the UGS indicators at the 400 m scale, only canopy cover and perceived usage quality were significantly associated with positive mental health, whereas none of the UGS indicators were related to general health. For visual greenness at different distance scales, there was a significant relationship between the indicator calculated by the sum of GVI at a 100 m scale and mental health. Canopy cover at a 400 m scale had a stronger correlation with mental health than perceived usage quality of nearby UGS and visual greenness at a 100 m scale. The findings highlight the importance of measuring both objective and subjective attributes of UGS provision and provide useful information for urban greening.

How Prepared Were Households for COVID-19? Implications for Young Children in 45 Low- and Middle-Income Countries

Background: COVID-19 and governments&rsquo; attempts to contain the pandemic are having many negative impacts on young children&rsquo;s health and development, including confining them and their families to their homes. However, little is known about levels of household preparedness for protecting young children from the direct and indirect effects of the pandemic in low- and middle-income countries&rsquo; (LMICs). Methods: Using data from nationally-representative household surveys (Demographic and Health Surveys, Multiple Indicator Cluster Surveys) in 45 LMICs since 2016, we estimated the percentages of young children living in households prepared for the pandemic and associated socioeconomic disparities. Preparedness was defined on the basis of space for quarantine, adequacy of toilet facilities and hand hygiene, mass media exposure, and phone ownership. The assessment was made at the country and aggregate levels. The final data set included 681,682 children under age five. Findings: Of all young children in the sample, only 19&middot;6% (95% CI, 16&middot;8%-22&middot;4%) lived in households prepared for COVID-19, ranging from 0&middot;6% in Ethiopia (95% CI, 0&middot;3-0&middot;9%) in 2016 to 70&middot;9% (95% CI, 68&middot;4-73&middot;4%) in Tunisia in 2018. In close to 90% of countries (40), fewer than 50% of young children lived in prepared households and fewer than 10% in 23 countries. Young children in rural areas or in the poorest households were less likely to live in prepared households than their counterparts, with an average gap of 11&middot;2 (95% CI, 9&middot;2-13&middot;3) and 34&middot;3 (95% CI, 26&middot;9-41&middot;7) percentage points, respectively. In 28 countries, fewer than 1% of young children in the poorest wealth quintiles lived in households prepared for the COVID-19, mostly concentrated in sub-Saharan Africa. Interpretation: A large portion of young children in 45 LMICs were not protected from COVID-19 in their homes and were at risk of worsening health and development. Global and national investments are urgently needed to improve housing conditions needed to prevent spread of COVID and other communicable diseases. Funding Statement: UKRI Collective Fund Award (Grant Ref: ES/T003936/1) to the University of Oxford, &ldquo;UKRI GCRF Harnessing the power of global data to support young children's learning and development: Analyses, dissemination and implementation&rdquo; Declaration of Interests: The authors declare no competing interests.

Assessment of Decontamination and Reuse of Disposable Filter Funnels Used in Microbiological Water Quality Tests.

To monitor safely managed drinking water services, an increasing number of countries have integrated water quality testing for Escherichia coli into nationally-representative household surveys such as the Multiple Indicator Cluster Surveys (MICS). However, plastic waste generated during such water quality testing programs, mostly through the use of pre-sterilized disposable materials, is non-negligible. The objective of this study was to evaluate several re-use protocols for disposable filter funnels used by the MICS water quality test kits. Decontamination and re-use protocols were assessed in centralized laboratory and decentralized field settings and neither yielded positive results. Re-use of 100 mL sterile funnels decontaminated with an alcohol wipe resulted in a higher incidence of false positive results (i.e., positive contamination when processing sterile water), both in the laboratory and field; therefore, a higher proportion of positives tests can be expected if these components are re-used. Further improvements to the decontamination technique and training are needed before material re-use can be reliably adopted. Autoclaving the funnels for re-use is feasible, provided that there is capacity to re-package and distribute funnels in a sterile manner.

Internet and private insurance participation

AbstractSince the Internet is an important information intermediary, this paper examines whether and how the Internet affects private insurance participation. Using national representative household survey data from China, we show that the Internet has a significant positive effect on households' private insurance participation. Then, we use both an instrumental variable analysis and propensity score matching to identify this causal effect. Furthermore, we find that this positive effect of the Internet is mainly driven by positive information on News Portals. In contrast, information regarding insurance on BBSs is usually negative and, thus, has a deterrent effect on insurance participation. Additionally, we examine the role of trust in these impacts. The results show that netizens with a high level of trust are more likely to purchase insurance, indicating that trusting the information received online is an essential factor for participating in private insurance.

Reducing poverty by ‘ignoring the experts’? Evidence on input subsidies in Zambia

Since the early 2000s, agricultural input subsidy programs (ISPs) have again become a popular agricultural development policy tool in sub-Saharan Africa. Although many of the programs include as objectives raising incomes and reducing poverty among smallholder farmers, there is limited empirical evidence on whether ISPs are achieving these goals. Focusing on the case of Zambia, we use nationally-representative household panel survey data to estimate the effects of ISP fertilizer and hybrid maize seed on smallholder farm household welfare. Results suggest that during the years of analysis, Zambia’s ISP reduced the poverty headcount ratio and poverty severity among these farm households by 1.5 and 9.0 percentage points, respectively. We found little evidence of spillover effects on households not directly participating in the program. However, we found some evidence of heterogeneous direct program effects: welfare improvements resulting from the ISP are greatest for households with farm sizes of 1–5 ha, female-headed households, and households with relatively older heads.