Paper Abstracts
Santosh Kumar
Preference for Boys, Family Size and Human Capital in India
Using data from the National and District Level Health Surveys, we test whether Indian parents make trade-offs between the number of children and investments in education and health of their children. To address the endogeneity arising from the joint determination of quantity and quality of children by parents, we instrument family size with gender of the first child. Given strong family preferences for boys in India, parents tend to have more kids when the first born is a girl. Our IV results show that children from bigger families have lower educational attainment and are less likely to be literate and to be enrolled in school, even after controlling for parent’s characteristics and birth order of children. The effects are larger for poorer and low-caste families and families with less educated mothers who are unlikely to practice selective abortions, suggesting that selection due to abortions biases the IV results downwards. By contrast, we find little impact of family size on children’s health.
Alfredo Burlando
Power Outages, Power Externalities, and Baby Booms
Determining whether power outages have significant fertility effects is an important policy question in developing countries, where blackouts are common and modern forms of family planning scarce. Using birth records from Zanzibar, this paper shows that a 2008 month-long blackout caused a significant, 15-18\% increase in the number of births eight to ten months later. The increase is similar across villages that had electricity, regardless of the level of electrification, while villages with no electricity connections saw no changes in birth numbers. Household surveys confirm that people were more likely to consume their leisure inside the home during the blackout, but time use shifts are unlikely to be the sole drivers of the fertility change. While it is unclear whether the baby boom is likely to translate to a permanent increase in the population, the paper highlights an important hidden consequence of power instability in developing countries. It also shows evidence that electricity imposes significant externality effects on those populations that have little exposure to it.
Alicia Menendez
The Economic Consequences of Death in South Africa
Abstract: Using a large longitudinal dataset, we quantify the impact of adult deaths on household economic wellbeing. The timing of lower socioeconomic status observed for households in which members die of AIDS suggests that the socioeconomic gradient in AIDS mortality is being driven primarily by poor households being at higher risk for AIDS. Following a death, households that experienced an AIDS death are observed being poorer still. However, the additional socioeconomic loss following death is very similar to the loss observed from deaths from other causes. Funeral expenses can explain some of the impoverishing effects of death in the household.
Shamma Adeeb Alam
Parental Health Shocks, Child Labor and Educational Outcomes: Evidence from Tanzania
This study examines the impact of adults’ health on child labor and schooling outcomes. Despite the effect of numerous fatal diseases on households, there has been no prior research that focuses on the effect of parental illness on child labor and educational outcomes. Employing longitudinal data from the Kagera region in Tanzania, a region severely affected by malaria and HIV/AIDS, this study demonstrates that parental sickness lead to increased child labor. Consequently, the increased child labor leads to lower school attendance and enrollment, where father’s illness having a more harmful effect compared to mother’s illness. Subsequently, it culminates in children failing their grade and in some cases dropping out of school altogether. Moreover, it
also delays children’s entry into the primary school. Furthermore, this study finds longer term effects on education: parental illness significantly decreases the likelihood of finishing primary school and on average decreases a child’s schooling by one-half of a year. Furthermore, using doctor’s diagnosis, this study is able to show that parents suffering from malaria have a very similar effect on child labor and education. Additionally, this study also analyzes the
heterogeneity in the data: older children and girls are more likely to work and less likely to be enrolled in school because of parental illness. The paper concludes with policy implications.
Ben Fitch-Fleischmann
Team Incentives and Health Worker Productivity: Evidence from Mozambique
Can improved treatment of HIV/AIDS in Africa be achieved simply by paying doctors to do more? I consider the effects of piece-rate (per unit output) financial incentives, paid based on team output, on the quantity of health services delivered at public health facilities in Mozambique. The results suggest that these incentives increased the delivery of five out of fourteen health services for which treatment effects can be identified. I find no evidence of a corresponding decrease in the delivery of services that are not nancially incentivized, suggesting there is no "crowding out" of intrinsic motivation. I also find no evidence of a relationship between the size of the incentive and the impact on output. Identication is based on a funding mechanism implemented
at a subset of facilities that created variation at the facility level in how health workers' pay is determined.
John Luke Gallup
Consistent Estimation of Income Inequality in Vietnam
Measurement error directly biases income inequality statistics, and practical methods to correct the bias have not been developed. Inequality statistics differ from most statistics in economics, which, as means, are not biased by random measurement error. Income data usually have large measurement errors and often distort not just the estimated level of inequality but its change over time due to sectoral change. Estimating the inequality of expenditure instead may help (if expenditure is measured without error), but cannot be decomposed by income source. I derive a new estimator of inequality that corrects for measurement error using conventional survey data, by combining estimates of income and expenditure. I show that this estimator is the only consistent inequality index in a class of common inequality measures using data subject to measurement error. With data from Vietnam, I show that the bias due to measurement error is very large and, uncorrected, gives a deceptive picture of recent trends in income distribution.
Benjamin Crost
Election Fraud and Post-Election Conflict:Evidence from the Philippines
A number of studies have documented a positive association between election fraud and the intensity of civil conflict. It is not clear, however, whether this association is causal or due to unobserved institutional or cultural factors. This paper examines the relationship between election fraud and post-election violence in the 2007 mayoral elections in the Philippines. Using the density test suggested by McCrary (2008), we find evidence that incumbents were able to win narrow elections through fraud. We further show that narrow incumbent victories led to an additional 3-4 post-election causalities, which is consistent with the hypothesis that election fraud causes conflict. We conduct several robustness tests and find no evidence that incumbent victories
increase violence for reasons unrelated to fraud.
Logan Lee
Pirates at Sea: Estimating the Trade and Welfare Consequences of Maritime Insecurity
We explore how maritime piracy in in the Gulf of Aden and Strait of Malacca has affected the volume of trade between country-pairs that ship through these choke points. By comparing trade volume changes along routes affected by piracy to those that are not, we estimate that piracy in the Gulf of Aden caused an average reduction in trade of 5% per year from 2006 – 2011. We extend this analysis to consider whether piracy affected local market prices for bulk food items in countries surrounding the Gulf of Aden. Our estimates suggest one additional attack attributed to Somali pirates increases local market prices by between 0.5 and 1% in close countries such as Kenya with the effect falling to 0 as the distance increases.
Santosh Kumar
Preference for Boys, Family Size and Human Capital in India
Using data from the National and District Level Health Surveys, we test whether Indian parents make trade-offs between the number of children and investments in education and health of their children. To address the endogeneity arising from the joint determination of quantity and quality of children by parents, we instrument family size with gender of the first child. Given strong family preferences for boys in India, parents tend to have more kids when the first born is a girl. Our IV results show that children from bigger families have lower educational attainment and are less likely to be literate and to be enrolled in school, even after controlling for parent’s characteristics and birth order of children. The effects are larger for poorer and low-caste families and families with less educated mothers who are unlikely to practice selective abortions, suggesting that selection due to abortions biases the IV results downwards. By contrast, we find little impact of family size on children’s health.
Alfredo Burlando
Power Outages, Power Externalities, and Baby Booms
Determining whether power outages have significant fertility effects is an important policy question in developing countries, where blackouts are common and modern forms of family planning scarce. Using birth records from Zanzibar, this paper shows that a 2008 month-long blackout caused a significant, 15-18\% increase in the number of births eight to ten months later. The increase is similar across villages that had electricity, regardless of the level of electrification, while villages with no electricity connections saw no changes in birth numbers. Household surveys confirm that people were more likely to consume their leisure inside the home during the blackout, but time use shifts are unlikely to be the sole drivers of the fertility change. While it is unclear whether the baby boom is likely to translate to a permanent increase in the population, the paper highlights an important hidden consequence of power instability in developing countries. It also shows evidence that electricity imposes significant externality effects on those populations that have little exposure to it.
Alicia Menendez
The Economic Consequences of Death in South Africa
Abstract: Using a large longitudinal dataset, we quantify the impact of adult deaths on household economic wellbeing. The timing of lower socioeconomic status observed for households in which members die of AIDS suggests that the socioeconomic gradient in AIDS mortality is being driven primarily by poor households being at higher risk for AIDS. Following a death, households that experienced an AIDS death are observed being poorer still. However, the additional socioeconomic loss following death is very similar to the loss observed from deaths from other causes. Funeral expenses can explain some of the impoverishing effects of death in the household.
Shamma Adeeb Alam
Parental Health Shocks, Child Labor and Educational Outcomes: Evidence from Tanzania
This study examines the impact of adults’ health on child labor and schooling outcomes. Despite the effect of numerous fatal diseases on households, there has been no prior research that focuses on the effect of parental illness on child labor and educational outcomes. Employing longitudinal data from the Kagera region in Tanzania, a region severely affected by malaria and HIV/AIDS, this study demonstrates that parental sickness lead to increased child labor. Consequently, the increased child labor leads to lower school attendance and enrollment, where father’s illness having a more harmful effect compared to mother’s illness. Subsequently, it culminates in children failing their grade and in some cases dropping out of school altogether. Moreover, it
also delays children’s entry into the primary school. Furthermore, this study finds longer term effects on education: parental illness significantly decreases the likelihood of finishing primary school and on average decreases a child’s schooling by one-half of a year. Furthermore, using doctor’s diagnosis, this study is able to show that parents suffering from malaria have a very similar effect on child labor and education. Additionally, this study also analyzes the
heterogeneity in the data: older children and girls are more likely to work and less likely to be enrolled in school because of parental illness. The paper concludes with policy implications.
Ben Fitch-Fleischmann
Team Incentives and Health Worker Productivity: Evidence from Mozambique
Can improved treatment of HIV/AIDS in Africa be achieved simply by paying doctors to do more? I consider the effects of piece-rate (per unit output) financial incentives, paid based on team output, on the quantity of health services delivered at public health facilities in Mozambique. The results suggest that these incentives increased the delivery of five out of fourteen health services for which treatment effects can be identified. I find no evidence of a corresponding decrease in the delivery of services that are not nancially incentivized, suggesting there is no "crowding out" of intrinsic motivation. I also find no evidence of a relationship between the size of the incentive and the impact on output. Identication is based on a funding mechanism implemented
at a subset of facilities that created variation at the facility level in how health workers' pay is determined.
John Luke Gallup
Consistent Estimation of Income Inequality in Vietnam
Measurement error directly biases income inequality statistics, and practical methods to correct the bias have not been developed. Inequality statistics differ from most statistics in economics, which, as means, are not biased by random measurement error. Income data usually have large measurement errors and often distort not just the estimated level of inequality but its change over time due to sectoral change. Estimating the inequality of expenditure instead may help (if expenditure is measured without error), but cannot be decomposed by income source. I derive a new estimator of inequality that corrects for measurement error using conventional survey data, by combining estimates of income and expenditure. I show that this estimator is the only consistent inequality index in a class of common inequality measures using data subject to measurement error. With data from Vietnam, I show that the bias due to measurement error is very large and, uncorrected, gives a deceptive picture of recent trends in income distribution.
Benjamin Crost
Election Fraud and Post-Election Conflict:Evidence from the Philippines
A number of studies have documented a positive association between election fraud and the intensity of civil conflict. It is not clear, however, whether this association is causal or due to unobserved institutional or cultural factors. This paper examines the relationship between election fraud and post-election violence in the 2007 mayoral elections in the Philippines. Using the density test suggested by McCrary (2008), we find evidence that incumbents were able to win narrow elections through fraud. We further show that narrow incumbent victories led to an additional 3-4 post-election causalities, which is consistent with the hypothesis that election fraud causes conflict. We conduct several robustness tests and find no evidence that incumbent victories
increase violence for reasons unrelated to fraud.
Logan Lee
Pirates at Sea: Estimating the Trade and Welfare Consequences of Maritime Insecurity
We explore how maritime piracy in in the Gulf of Aden and Strait of Malacca has affected the volume of trade between country-pairs that ship through these choke points. By comparing trade volume changes along routes affected by piracy to those that are not, we estimate that piracy in the Gulf of Aden caused an average reduction in trade of 5% per year from 2006 – 2011. We extend this analysis to consider whether piracy affected local market prices for bulk food items in countries surrounding the Gulf of Aden. Our estimates suggest one additional attack attributed to Somali pirates increases local market prices by between 0.5 and 1% in close countries such as Kenya with the effect falling to 0 as the distance increases.