|
" Three Essays in Health Economics "
Mahmud, Mir N.
Chatterji, Pinka
| Document Type
|
:
|
Latin Dissertation
|
| Language of Document
|
:
|
English
|
| Record Number
|
:
|
1106831
|
| Doc. No
|
:
|
TLpq2416198032
|
| Main Entry
|
:
|
Chatterji, Pinka
|
|
|
:
|
Mahmud, Mir N.
|
| Title & Author
|
:
|
Three Essays in Health Economics\ Mahmud, Mir N.Chatterji, Pinka
|
| College
|
:
|
State University of New York at Albany
|
| Date
|
:
|
2020
|
| student score
|
:
|
2020
|
| Degree
|
:
|
Ph.D.
|
| Page No
|
:
|
218
|
| Abstract
|
:
|
This dissertation comprises three empirical essays. The first essay estimates the long-term health and socioeconomic impact of early life nutritional deficit using the Bangladesh Famine from 1974–1975 as a natural experiment. The second essay investigates if pubic health insurance coverage of “ immigrant children” changed as a result of the lift of the “five-year-ban” in the Children Health Insurance Reauthorization Act-2009. The final chapter estimates the relationship between child health conditions and maternal employment using conventional econometric models. To expand the analysis and contribute to the literature, I estimate the relationship by implementing a Multiple Indicator Multiple Causes Model (MIMIC). Fetal origins hypothesis proposed by David J. Barker (1990) suggests that intrauterine undernutrition may have long-lasting health impacts. Although early epidemiological studies provide evidence in support of the hypothesis using correlational analysis, economists deployed identification strategies using catastrophic events like natural disasters, famines, pandemic episodes or war as natural experiments to establish a causal link between fetal nutritional shock and adult health and non-health outcomes. In the first chapter, I investigate whether exposure to nutritional shock in early childhood adversely impacts health and socioeconomic outcomes at older ages using the Bangladesh Famine from 1974–1975 as a natural experiment. The Bangladesh Famine of 1974–1975 was primarily caused by food shortage from successive natural disasters such as floods and droughts and the inability of the government to import food or ensure food aid in a timely manner. Rice market performed poorly as massive hoarding took place, resulting in a sharp increase in rice prices, massive spike in unemployment, and a drop in purchasing power for the rural population. Early literature examined short-term effects on mortality and birth outcomes. Evidence of long-term impact is available only using regional data. This is the first study to analyze the long-term effects of the Bangladesh Famine using a nationally representative sample from two separate sources: the Bangladesh Sample Census 2011 and the Bangladesh Demographic and Health Survey. Exploiting the temporal and geographical variation in famine exposure, I estimate a difference-in-differences model to test the fetal origins hypothesis for Bangladesh. Famine exposure is determined by reported place of birth in the Bangladesh Sample Census 2011, with those living in the rural parts of Rangpur, Mymensingh and Sylhet defined as exposed and those living in the rural areas in the other parts of Bangladesh defined as non-exposed. Using the month and year of birth from the Bangladesh Demographic and Health Survey, I examine if the effects of famine exposure different for those who are famine-born from those who are famine-conceived. I find that fetal exposure to acute undernutrition has a differential impact on males compared to females. There is evidence of an adverse impact on labor market participation for males with early childhood famine exposure. Educational attainment of the famine survivor male cohort is better than the neighboring cohorts, suggesting selection effects. Female cohort exposed to famine has lower educational attainment, although the effect is not statistically significant. Both famine exposed cohorts–famine-born and famine-conceived–in general, have worse outcomes compared to non-exposed cohorts, however, the effect is larger for famine-conceived cohort providing support for the fetal origins hypothesis. This study highlights the importance of policies and programs focusing nutritional intake of pregnant women to yield short-term objectives in the form of improved birth outcomes and long-term benefits in the form of better health and economic status. I also find that famine exposure did not result in long-term migration to big cities. In the second chapter, I examine the impact of the Children Health Insurance Reauthorization Act-2009 on public health coverage of immigrant children. Although Immigrant children represent approximately 3 percent of the total U.S. child population, they remain the most vulnerable group in terms of access to public health insurance since the enactment of the “five-year-ban” for legal immigrants in the Personal Responsibility and Work Opportunity Reconciliation Act of 1996. Children's Health Insurance Reauthorization Act (CHIPRA) of 2009 provided states an option to receive federal funds to expand eligibility for immigrant children regardless of their length of residency by lifting the “five-year-ban”. In this paper, I utilize the cross-state variation in policy environment before and after the adoption of CHIPRA to compare the differences in access to public health insurance among low-income immigrant children. I find that adoption immigrant child option of CHIPRA has resulted in 8 percentage points increase in health coverage for the target group, almost entirely contributed by an equal increase in coverage through public health insurance. I explicitly check if there is a differential outcome for immigrant children CHIPRA states with some form of the state-funded program compared to other CHIPRA states without such arrangements before CHIPRA. The estimated effects on the coverage on the immigrant children show a large and statistically significant impact in the states with state-funded substitute programs before CHIPRA. This suggests that there might have been unmet demand for insurance in these states and provision of CHPRA helped to increase the scale of the programs resulting in greater coverage in these states following CHIPRA. An increase in coverage entirely comes from the ranks of previously uninsured children; no evidence of crowding out from the private insurance was found. I also verify the lack of crowding out by estimating the labor market response among mothers of immigrant children. In addition, I examine the effect of CHIPRA-2009 on interstate migration and population growth into CHIPRA states as the mechanism of the increase in coverage. Although CHIPRA states show an increase in population growth, evidence of interstate migration from the immigrant population is not found. In the third chapter, I estimate the relationship between child disability and maternal employment. Child health is an important determinant of maternal employment. Children with disabilities are both time and resource-intensive, generating opposing effects for maternal work decisions. Existing empirical literature on the relationship between child health and maternal employment outcome is diverse in terms of disability definitions, chiefly due to the limitation imposed by the survey used for the study: for married mothers of disabled children, the estimated employment effect is negative in almost all the studies, but for single mothers of disabled children result remain mixed. In this chapter, I estimate the relationship between child disability and maternal employment decisions using the Child Development Supplement of the Panel Study of Income Dynamics (PSID-CDS). The PSID-CDS contains comprehensive child's health, cognitive and behavioral development measures which will allow us to define child disability using the conceptual framework provided by the International Classification of Functioning, Disability, and Health, version for children and Youth (ICF-CY, 2007). My study contributes to the literature by implementing a Multiple Indicator Multiple Causes Model (MIMIC), where I assume that Child Disability is a latent variable that is linked to several observable health-related indicators and incorporates other exogenous factors in the determination of child disability. To my knowledge, this is the first paper that uses the MIMIC model to define the Child Disability Index in the context of estimating the impact on maternal employment. It appears from the regression analysis that aggregate measure of disability masks the impact of child health on maternal employment.
|
|
|
:
|
When I estimate the regression with different disability types, I can detect the opposing forces that are at play: across various specifications, I have found a consistent negative, but statistically insignificant effect on most cases, the relationship between the presence of a child with a physical disability on maternal labor market outcome, while the presence of children with a mental disability is mixed and needs further investigation to capture the mixed results. I find that the presence of a child with severe mental disability generates a negative impact on a mother’s likelihood of working or weekly working hours. For married mothers, the presence of a child with a less severe mental disability has a positive impact on the likelihood of working, and when employed they tend to work longer hours. To verify my result from the standard models, I extend my study to construct a latent child health index using the Multiple Indicator Multiple Causes Model (MIMIC) and estimate the impact of child health index on maternal employment using the structural equation model. The result of this modeling framework is consistent with the results found in traditional models used in the literature.
|
| Subject
|
:
|
Economics
|
|
|
:
|
Public health
|
| |