Ungated working paper versions of my published papers can be downloaded, where available. If you would like a copy of the paper, please drop Anh an email at email@example.com.
“How does household residential instability influence child health outcomes? A quantile analysis”, with Emma Baker, Lyrian Daniel and Rebecca Bentley, International Journal of Environmental Research and Public Health (special issue), 16(21), 10/2019. Open access.
- Abstract: At the core of housing and welfare research is a premise that stable residential environments are important to children’s health and development. The relationship between housing stability and health outcomes for children is however complex – stable housing situations are sometimes associated with poorer health outcomes, and some children may be more or less resilient to residential instability. The Longitudinal Study of Australian Children (LSAC) dataset enables us to longitudinally follow the housing and health of more than 10,000 children and their families. We employ a quantile analysis technique, a currently underutilized tool for testing associations across the distribution of an outcome, to test whether exposure to housing instability has differential impact on children’s health dependent on their initial health status. Our findings suggest that health outcomes of residential instability are highly dependent on children’s initial health status.
- Abstract: Cold housing is not widely recognised as a problem that occurs in mild-climate countries like Australia. But emerging evidence suggests that it is an important, albeit under-acknowledged, problem that may contribute to high rates of ill health and mortality during the winter months. We bring together two historically important theoretical developments in order to better understand the social and economic distribution of cold housing. Drawing on nationally representative data from the Household, Income and Labour Dynamics in Australia survey between 2001 and 2016, we find that the characteristics of households unable to adequately heat their homes strongly reflects known patterns of inequality across, for example, tenure, employment and health, but that there are also more unexpected trends in age and income. Critically, our analyses demonstrate that individuals’ vulnerability to cold housing risk can be anticipated, which has important implications for public policy and community-based interventions.
“New evidence on mental health and housing affordability in cities: A quantile regression approach”, with Emma Baker, Lyrian Daniel and Rebecca Bentley, Cities, 96, 2020. Free access by October 26, 2019.
- Abstract: Unaffordable housing costs are one of the most pressing issues facing our cities, affecting people’s health in difficult to measure ways. People’s health varies over time and dynamically interacts with experiences of housing. Longitudinal analyses rarely explicitly model these variations. In this paper we apply panel quantile regression to test whether cumulative exposure to unaffordable housing over time has differential impact on mental health, dependent on initial health status. Using an annual longitudinal sample of urban Australians (2001-2016), we model mental health outcomes using quantile regression. Although traditional fixed-effects models find weak evidence of cumulative effect, quantile regression reveals that individuals with low-median initial mental health were more affected by unaffordable housing, and individuals with high initial mental health appeared to be protected. Our findings suggest quantile regression as a promising method for understanding complex human effects of urban problems, and that policies targeted toward people with the poorest mental health may mitigate the consequences of exposure to unaffordable housing.
“Natural Disasters and Mental Health: A Quantile Approach”, with Nadezhda Baryshnikova, Economics Letters, 180, pp 62-66, 2019.
- Abstract: This article investigates heterogeneity in the effect of experiencing natural disasters on mental health. Using Australian longitudinal data, we find that home owners generally show a reduction in mental health score after a disaster. There is a strong negative effect in the lowest two quantiles of the distribution for non-owners.
- Ungated working paper: Here
- Early working paper: Here
Dynamics of household demand for renewable energy infrastructure, with Lyrian Daniel, Research in progress.
“Social Housing exit points, outcomes and future pathways”, with Emma Baker, Chris Leishman, Rebecca Bentley, and Lyrian Daniel. AHURI Final Report 326, Australian Housing and Urban Research Institute Limited, Melbourne, doi: 10.18408/ahuri-3119901, 2020.
- Description: Social housing is increasingly seen by governments and many tenants as just one part of tenants’ broader housing pathways (be it a safety net, stepping-stone, or springboard). This research provides a new national evidence-base on pathways into, within, and out of social housing tenancies to assist understanding of entry and exit patterns, and housing and non-housing outcomes.
International Trade and Development
“Shipping Cost and Development of the LLDCs: Panel Evidence from the Common Correlated Effects Approach”, with Nicholas Sim, World Economy, 00: 1– 29, 2019.
- Abstract: We estimate the impact of shipping cost on development for landlocked developing countries (LLDCs). Since container trade is important to them, we construct a country-specific measure of shipping cost, called HarpexCost, which combines the global cost of container shipping with information on how exposed to container shipping each LLDC is. We employ the Common Correlated Effects (CCE) estimator of Pesaran (2006) to first estimate the impact of HarpexCost on the LLDCs’ development, and then recover the actual impact of shipping cost from these estimates. Overall, we observe that shipping cost has large negative effects on the LLDCs. Building upon these results, we provide new estimates on the cost of landlockedness and how trade benefits their development.
“Do Exports Affect Urbanization in Sub-Saharan Africa? Evidence From The Baltic Dry Index And Panel Regressions With Cross-Sectional Dependence”, with Nicholas Sim, Journal of African Economies, FREE ACESS, 2020.
- Abstract: Unlike other developing countries, urbanisation in sub-Saharan Africa appears to be unaccompanied by an improvement in economic fundamentals. This paper provides new evidence that exports may increase urbanisation in sub-Saharan Africa. To address the issue of reverse causality, we instrument exports with information linked to the Baltic Dry Index, which reflects the shipping cost of primary commodities that the sub-Saharan Africa countries mainly export. To handle a large class of confounding variables and cross-sectional dependence, we employ panel regressions with interactive fixed effects. We find that exports have a sizable positive effect on urbanisation. Interestingly, we also find that exports will lose their statistical significance if cross-sectional dependence is overlooked, suggesting that the true effect of an economic fundamental on urbanisation could be obscured by cross-sectional dependence.
“Child Mortality in the LDCs: The Role of Trade, Institutions and Environmental Quality”, with Faqin Lin and Nicholas Sim, School of Economics Working Papers 2015-15, University of Adelaide, School of Economics.
- Abstract: Child mortality is a persistent problem for the world’s least developed countries (LDCs). Given that trade fosters economic development, one plausible solution is to raise the low levels oftrade in the LDCs, but how effective this approach might be could depend on the quality of institutions. In this paper, we use a novel instrumental variable approach to estimate the effect that trade might have on child mortality in the LDCs. We find that trade does not lead to lower levels of child mortality. In fact, in autocratic LDCs, trade could even cause child mortality to increase as we find that pollution, which adversely affects health, may rise with trade.
- Ungated working paper: Here
Panel Data Applications
“Does Political and Economic Inequality affect Institutional Quality?”, with Nadezhda Baryshnikova and Maria Wihardja, The Economic Record, 92(297), pp 190-208, 2016.
- Abstract: We study how inequality, democracy and economic development affect institutions in a dynamic panel model using data for 78 countries from 1984 to 2006. We suggest a model that assumes the sluggish adjustment of institutional quality and the interplay between the key explanatory variables. We find that democracy has a non‐linear effect on government stability and military involvement in politics. This effect depends on inequality and GDP per capita. Inequality matters for law and order and has a non‐linear effect as well. However, the results for wealth effect are mixed.
“Panel Data Quantile Regression with Two-Way Fixed Effects”, with Nicholas Sim.
- Abstract: We construct a two-way fixed effect quantile regression estimator by applying the one-way fixed effect quantile regression technique of Canay (2011). We explore the Monte Carlo properties of this estimator under the assumption that the data generating process may contain a one- or two-way fixed effect. We apply this estimator to a panel data with country and year fixed effects to see if trade cost, as proxied by the Baltic Dry Index, affects the trade and income of least developed countries heterogeneously.
- Abstract: Reducing child marriage is seen as one of the essentials to women’s empowerment and wellbeing, ending the intergenerational cycle of poverty and rights violation. In this paper, we use a longitudinal household survey from Indonesia to study whether a food subsidy (Raskin) reduces child marriage. Modelling treatment assignment with Coarsened Exact Matching and Differences-in-Differences, we show that the unconditional rice subsidy significantly reduces the likelihood of marrying as a child.
- Preliminary Draft: Here
- Ungated working paper: Here
Income-based mandates on the demand for private hospital insurance and its dynamics, with Terence Cheng, Thomas Buchmueller and Kevin Staub. Revised and Resubmitted.
- Abstract: We examine the effect of an income-based mandate on the demand for private hospital insurance and its dynamics in Australia. The mandate, known as the Medicare Levy Surcharge (MLS), is a levy on taxable income that applies to high income individuals who choose not to buy private hospital insurance. Our identification strategy exploits changes in MLS liability arising from both year-to-year income fluctuations, and a reform where income thresholds were increased significantly. Using data from the Household, Income and Labour Dynamics in Australia longitudinal survey, we estimate dynamic panel data models that account for persistence in the decision to purchase insurance stemming from unobserved heterogeneity and state dependence. Our results indicate that being subject to the MLS penalty in a given year increases the probability of purchasing private hospital insurance by between 2 to 3 percent in that year. If subject to the penalty permanently, this probability grows further over the following years, reaching 13 percent after a decade. We also find evidence of a marked asymmetric effect of the MLS, where the effect of the penalty is about twice as large for individuals becoming liable compared with those going from being liable to not being liable. Our results further show that the mandate has a larger effect on individuals who are younger.
- Ungated working paper: Here
“Population Diversity, Trust, and the Transmission of HIV/AIDS in Sub-Saharan Africa”, with Pide Lun and Nicholas Sim
- Abstract: We provide new evidence that population genetic diversity can mitigate the transmission of HIV in sub-Saharan Africa. Using a difference-in-differences strategy, we exploit the fact that trade may increase HIV incidence, and demonstrate that HIV transmits more easily in countries that are more genetically homogeneous.