You are here

2018 CRDCN Emerging Scholars Grant Recipients

Sarah SIngh

Affiliation: Western University

Project Title: Examining the Multilevel Influence of Social Determinants on Cardiovascular Health

Project Summary: With 1 in every 12 Canadians affected by heart disease, there is an urgent need for novel approaches towards cardiovascular health (CVH) and longevity.(PHAC, 2016) Historically, many studies have investigated factors affecting health, or health determinants, primarily at the individual level.(Bryant, 2011) However, more recent studies have successfully identified other societal determinants of health at the community or national levels that can independently influence health.(Carey, 2015) Still missing in this field of research, are studies that have taken a real-world perspective where these multiple levels of determinants act simultaneously, rather than independently, to influence CVH. Main Objective: The proposed research aims to address the gap in Canadian and international scientific literature by investigating the relative influences of individual and neighborhood social determinants acting simultaneously on the CVH of Canadians. Theoretical Framework: To accomplish the main objective of this study, the Main Determinants of Health model proposed by Dahlgren and Whitehead in 1991, will be used as a guiding framework. Dahlgren and Whitehead, in conjunction with the World Health Organization, proposed the ‘Main Determinants of Health’ model which described four ‘layers’ of social determinants that could be targeted by policies and interventions for addressing inequities in health. (Dahlgren and Whitehead, 1991) The model suggests that health is determined by factors acting in unison at both the individual and ecological levels. Methods: Overall, the study design is cross-sectional and uses secondary data from the Canadian Community Health Survey (CCHS) 2015-2016, provided by Statistics Canada and available at the Research Data Center at Western University. The dependent variable is individual cardiovascular health, as measured by the Canadian Cardiovascular Health Index (CCVHI) which is a composite score that indicates the presence of exercise, healthy diet, no smoking, low BMI, and healthy blood pressure, cholesterol and glucose levels. Higher CCVHI scores indicate better overall CVH. (adapted from Lloyd Jones, 2009) The independent variables are known social determinants of health present at the individual and neighborhood levels: age, sex, race, education, healthcare availability, neighborhood median income and density. Multilevel regression models will be utilized to examine the influence of independent variables– social determinants of health, on the dependent variable– individual CVH. The study sample will include adults aged 20-75 years from the 2015-2016 CCHS two-year file who reside in private dwellings in 10 provinces and 3 territories. Discussion: A main strength of the study is the use of multilevel methodology to address the main objective. The structure of the theoretical model informs a multilevel analysis which allows for the partitioning of individual and neighborhood influences simultaneously in the model. An expected challenge is that the health of neighborhoods is influenced by the surrounding environment. For example, persons living in one neighborhood may seek healthcare from another neighborhood thus, their health is not based solely on factors within their neighborhood of residence. In general, multilevel spatial analysis may be employed in research to address this challenge. However, multilevel spatial analysis can be highly complex and is considered outside the scope of this study. Conclusion: Results from this study will demonstrate that, while individual and neighborhood social determinants both influence CVH, their relative influences on CVH are critical for understanding the real-world setting where determinants act simultaneously, rather than independently, to influence CVH.

 Kanika Samuels-Wortley

 Affiliation: University of Waterloo

 Project Title: Looking beyond the Racial Lens: Combining Critical Race Theory and Mixed-Methods to Examine Racialized and White Youths’ Perceptions of police.

 Project Summary: Numerous policing scholars argue that effective law enforcement is contingent on public support A growing volume of U.S. and British research examining public perceptions of the police suggests that trust and confidence in the police is very low among youth and specific racialized populations. However, there is a gap in Canadian research that examines the complexities of the relations between youth, particularly in relation to racialized youth and the police—which is the focus of my research.

 Jill Furzer

 Affiliation: University of Toronto

 Project Title: Mental Health and Human Capital: Understanding Diagnostic Decisions and Long-term Effects

 Project Summary: Previous work in special education found that identification of cognitive or non-cognitive disabilities in school should improve a child's academic achievement (Hanushek, 2002a) and that of their peers (Aizer, 2009; Figlio, 2007; Fletcher, 2010). However, identification and treatment may also compromise later life academic attainment and employment outcomes (Currie, Stabile and Jones, 2015). Taken together, assessing mental illness in children necessitates teachers, parents and physicians balancing appropriate identification to improve a child's wellbeing with mitigating undue harm. Nevertheless, the process for identification of mental health disorders and their interactive effects with the education system remain largely understudied. The limited research in mental health identification processes leaves several key questions unanswered. These include the role of the teacher and family preferences in decision making, along with the potential for heterogeneous treatment effects on long-term outcomes. My dissertation work addresses these gaps in the literature, evaluating questions of mental health diagnosis and misdiagnosis and their long-term effects. I address these questions over three papers utilizing the National Longitudinal Survey of Child and Youth (NLSCY).

I first evaluate the role of teachers in known instances of mental health misdiagnosis based on relative age in a classroom setting. I test whether this known diagnostic jump for younger children in a grade is due to over-diagnosis of younger students or the under-identification of older students. I further assess the role of symptom type and student gender, along with teacher training and peer behaviour factors on teacher perception of student mental health risk. Teacher assessments in the first five cycles of the NLSCY provide the basis for this analysis. Secondly, I investigate which students are diagnosed or not based on data-driven mental health risk prediction. Using discord between machine-predicted underlying disorder risk and physician diagnosis, I test how missed identification or low-risk identification might impact later educational outcomes. I employ simulation modelling to re-evaluate previous adverse average treatment effects of identification on long-term socioeconomic issues. Use of the longitudinal NLSCY cohort linked to the T1 tax files provides an extensive data contribution to the literature and allows for robust evaluation of heterogeneous outcomes. In my final paper, I probe further into teacher effects on diagnosis, assessing how stereotyping impacts in-school mental health assessments for students. This work builds on theories of education production and behavioural economics, relying again on the teacher component of the NLSCY.

Given the facilitating role of schools in mental illness identification for children and the significant impact of mental health on long-term life outcomes, it is essential to investigate the role of educational settings on mental health identification. Moreover, evaluating why some children might be missed is integral for identifying long term health inequities. As mental health incidence in both child and adult populations continue to rise, understanding these mental health decisions making processes will only grow in importance.

 Peri Abdullah

 Affiliation: York University

 Project Title: Factors associated with prenatal ultrasound utilization

 Project Summary: Despite its abundant use, the factors surrounding prenatal ultrasound utilization are still relatively underexplored. The aim of this dissertation is to investigate three different factors in association with prenatal ultrasound utilization with the following objectives: 1)    investigate the factors associated with the timing of first prenatal ultrasound. It is important to investigate this area because having the first prenatal ultrasound too early may harm the developing fetus and can lead to misinterpretation and unnecessary medical interventions. In addition, it can set a trend of overutilization as women who receive an early first ultrasound tend to receive more ultrasounds during their pregnancy. On the other hand, having a late first ultrasound may lead to missed opportunities of screening. 2)    investigate the relationship between the number of prenatal ultrasounds and caesarean deliveries in Canada and the USA. Caesarean deliveries are on the rise globally and efforts are being made to decrease this trend. It is important to identify the factors associated with caesarean delivery in order to reduce the risks associated with such a procedure (e.g. admission of the baby to the NICU, complications in future pregnancies and maternal and neonatal mortality). One of the factors that could contribute to the rise in caesarean deliveries is the number of ultrasounds received during pregnancy. Very few small-scale studies have looked into this relationship in different parts of the world and have generally found an association between prenatal ultrasound utilization and caesarean deliveries. 3) Investigate the relationship between having a single versus multiple providers with the number of prenatal ultrasounds. Having a single healthcare provider caring for a pregnant patients can lead to a lower likelihood of ordering tests. However, no research has looked into the specific association between having a single/multiple providers during pregnancy and the number of ultrasounds per pregnancy. Having too many ultrasounds during pregnancy can harm the developing fetus and can burden the healthcare system. Having too few ultrasounds during pregnancy can lead to missed diagnoses that might need intervention.

 Laura Warren

 Affiliation: University of Toronto

 Project Title: Aging among community-dwelling, off-reserve Indigenous populations in Canada

 Project Summary: We will engage with local communities and the broader Indigenous population through meetings and conferences to increase awareness of frailty and healthy aging practices in Indigenous populations across Canada. This study will be the first study focused on aging among the community-dwelling, off-reserve Indigenous population in Canada. By characterizing the epidemiology of frailty and age-related conditions in Indigenous populations we hope to identify the prevalence of and factors associated with frailty.  A strengths-based approach to aging will be used to look at qualities that help Indigenous people overcome challenges and remain healthy as they age despite multiple marginalization and colonial factors. We propose to characterize the epidemiology of frailty in Canada’s community-dwelling, off-reserve Indigenous populations using data from the Canadian Community Health Survey. The objectives of the study are to: 1) Estimate the prevalence of frailty using Frailty Index criteria among off-reserve Indigenous populations in comparison to non-Indigenous populations; 2) Compare risk factors associated with frailty between Indigenous and non-Indigenous populations; 3) Identify protective factors associated with frailty among Indigenous populations despite marginalization and colonial factors.

There are approximately 130,000 records for each CCHS survey for cycles 1.1 (200-2001) to 4.1 (2007) and 65,000 records for survey years 2008-2014 yielding a total sample size of ~975,000. For cycles 1.1 and 3.1 (2005-2006) the average response rate was 90%, with 56% of respondents 45 years of age or older and 3.8% of respondents self-identifying as Indigenous. Given these figures the expected sample size available for analysis is ~17,400 Indigenous respondents and ~440,000 non-Indigenous respondents over the age of 45.

We will use a definition of frailty which has been previously validated using data from the CCHS. Thirty age-related measures from the CCHS including chronic diseases, physical and psychological health, falls, and the need for assistance with activities of daily living (ADL) were used to create a FI. In addition to looking at factors typically associated with age-related health conditions (e.g. level of education, smoking status) we will use data from the CCHS to identify qualities associated with successful aging in the Indigenous population. Variables of interest will include: fruit and vegetable consumption, feeling of belonging to local community, self-help group attendance, use of alternative medicine, leisure activities (e.g. reading, walking, fishing), physical, but not leisure activities (e.g. bicycling to work), self-esteem, mastery, social supports, spirituality, speaking traditional language, pride and pleasure from accomplishment, mood, and absence of distress and depression.

 Karen Ugarte Bravo

 Affiliation: McMaster University

 Project Title: Essays in Applied Econometrics

 Project Summary: Over the past two decades Canada has experienced a dramatic rise in the dispensing and consumption of prescription opioid medications (INCB, 2015). The rapid growth in prescription opioid consumption has brought pain relief to many Canadians who suffer from chronic and/or acute pain (Jovey, 2003) but has also led to increased rates of prescription opioid abuse, dependence and mortality due to overdose (Fischer et  al., 2008). In order to develop policies that reduce prescription opioid related issues without restricting access to necessary pain treatment, it is imperative to identify the factors that influence prescription opioid consumption. Pharmaceutical insurance status is one factor that could influence prescription opioid use. By exploring the impact of Quebec’s mandatory universal pharmaceutical drug insurance program on prescription opioid use in Quebec (relative to other provinces), valuable insight can be gained into the relationship between pharmaceutical insurance status and prescription opioid use. This pharmaceutical insurance reform serves as a source of exogenous variation in pharmaceutical insurance status, which when coupled with longitudinal data allows us to extract a causal relationship between pharmaceutical insurance status and prescription opioid use, as opposed to a simple association between the two variables. Evaluating Quebec’s pharmaceutical insurance policy reform will also help inform the policy debate regarding the expansion of public drug
coverage to a universal level in Canada. Thus, the findings that will arise from analyzing the aforementioned policy will help inform both prescription opioid and pharmaceutical insurance related policy topics.

 Xian Zhang

 Affiliation: McGill University

 Project Title: Essays on School Duration and Cohort Size: Implications from the 1999 Ontario High School Reform

 Project Summary: This dissertation is composed of two essays which provide a comprehensive analysis of the educational reform policy implemented in Ontario, Canada in 1999. This reform eliminated the fifth year of high school and established a new four-year curriculum, enabling students to obtain high school graduation certificates after a total of twelve rather than thirteen years of schooling. This quasi-experimental feature of the policy, together with rich data sources, enables me to investigate the policy effect along multiple dimensions and understand its underlying mechanisms.

The first chapter of my thesis investigates the policy effect on students’ academic performance during high school, their graduation decision, students’ postsecondary education choice and earnings. The second chapter focuses on a special subgroup of students: the Ontario double cohorts who have university degrees. It aims to examine the impact of cohort size on labor market outcomes.

 James Stutely

 Affiliation: McMaster University

 Project Title:  Essays on labor supply and public pensions

 Project Summary: My dissertation will consist of four chapters. The first two will draw on the Longitudinal Administrative Database (LAD) to study the influence of public pension program rules on the labor supply and wage earnings of older workers. The other two will use Census 2016 data linked with the LAD files to investigate retirement and related issues of older immigrants who face a set of public pension rules that differ from the Canadian born.

The first two chapters will exploit the 2008 policy reform that increased from $500 to $3,500 the amount that could be earned (the employment earnings exemption level) before being subject to a reduction in the Guaranteed Income Supplement (GIS). GIS benefits are available to low-income Old Age Security pensioners who are without other sources of retirement income. Conditional on working, their employment income faces a GIS claw-back for earnings above the exemption level, implicitly taxing employment income at a 50 percent rate. These implicit taxes impose disincentives to work by creating a budget constraint kink at the exemption level.

The first chapter will look at the intensive margin: the influence of the increase in the earnings exemption level on the conditional distribution of employment income of older workers. The work thus far suggests a significant “bunching” mass that develops at the exemption level with a corresponding missing mass both below and above that level. Interestingly, most of the excess mass is due to aggregate bunching. That suggests that the bunching may be the result of employer responses, with some firms offering wage packages accommodating the GIS exemption level. I propose to estimate the wage loss/gain that results from aggregate bunching by comparing the observed missing mass to a counterfactual distribution of older worker’s earnings if there were no earnings exemption.

The second chapter explores the response at the extensive margin: the decision to work or not to work. I plan to rank individuals who were in their early 50s before the 2008 reform in to groups based on their predicted probabilities of GIS eligibility at age 65. The ranking will be based on their initial position in the net-income distribution; I will follow the groups longitudinally before and after the reform. Preliminary results suggest that the post-age-65 employment rate among those with the highest probabilities of future GIS eligibility increased following the reform.

The last two chapters will exploit the Social Security Agreements (SSA) that Canada has with various countries that make it possible for immigrants from those countries to substitute years spent outside of Canada for Canadian residency requirements in assessing their eligibility for public pension benefits. Such agreements allow for quasi-experimental variation in the retirement decisions of immigrants based on their pre-migration country of last permanent residency. Furthermore, Canadian public pensions can be collected in SSA countries based on contributory years within Canada. I plan to investigate whether SSAs influence immigrant retirement timing within Canada and whether they influence public pension take-up outside of Canada in chapters three and four, respectively.

 Xavier St. Denis

 Affiliation: McGill University

 Project Title: Job instability and its consequences in OECD countries

 Project Summary: The main contribution of my dissertation is measuring and explaining the increase in instability in the career trajectories of Canadian, British and German workers. The motivation behind this project is solving the contradiction between two bodies of literature. On one hand, a large body of scholarship has focused on the shift away from the standard employment relationship and towards precarious work since the 1980s. This literature studies how the adoption of flexible employment practices by employers has led to decreased job and employment security, alongside with other forms of social protection. Recently, scholars have emphasized the rise of the gig economy, including platform work, as part of that trend. On the other hand, many empirical studies in Canada, the US and the UK show that jobs have on average remained relatively stable over the past four decades, as measured by the probability of job separation or average job tenure duration. This would seem to contradict the claims made in the precarious work literature.

My results using labour market data show important shifts in job instability patterns in Canada, the UK and Germany. My research is the first to identify a clear increase in job instability since the 1980s in Canada and in the UK for men in the private sector. In Germany, I instead find evidence of a polarization of job instability, consistent with the dualization literature in comparative political economy. This trend is partly driven by differences in trade union strength across sectors of the economy. It is also driven by gendered dynamics: family policies lead to high rates of part-time work among mothers under employment contracts that offer little job security. In addition, my results point to several methodological limitations in previous studies that lead to inaccurate conclusions about trends in job instability in Canada and elsewhere.

Finally, in a single-authored paper, I developed a new method to obtain reliable estimates of the changing importance of lifetime jobs. Lifetime jobs are generally identified as a central feature of the postwar labour market, in contrast with flexible employment and unstable jobs predominant in the new economy. This paper argues that the literature would gain from more precise empirical evidence of its changing importance. It develops and evaluates a new cohort-based measure using UK cross-sectional data. The results show that lifetime jobs were rare even in the postwar economy, in contrast with the descriptions provided by some of the sociological literature.

Overall, my work emphasizes the need to consider transformations in employer practices beyond the use of non-standard work arrangements as drivers of the flexibilization of the worker-firm relationship.

  Diego Capurro Fernandez

 Affiliation: McGill University

 Project Title: Educational inequality in cancer mortality in the Canadian adult population

 Project Summary: As the leading cause of death in Canada, cancer is a major public health concern. It accounted for nearly a third of all deaths in 2012, equaling 80,000 deaths in 2017. Nearly half of all Canadians are expected to develop cancer during their lifetime, with a quarter dying as a result. Beyond population average burdens, a consistent body of evidence demonstrates that many cancer-related deaths show clear socioeconomic patterns in their distribution: individuals at the lower end of the socioeconomic ladder have higher mortality burden compared to those who are better off. Education is frequently used as a proxy indicator of socioeconomic position (SEP) since it captures the knowledge-related assets and qualifications of individuals which strongly determine occupation, income and many socioeconomic aspects of life. Thus, education is widely used socioeconomic stratifier to monitor progress to or diversion from health inequality reduction.
Educational inequality in cancer mortality is substantial but context-dependent, with important variations between countries in terms of its magnitude, change over time, and potential for reductions. This heterogeneity in inequality highlights the importance of national estimations to characterize cancer beyond population averages and inform policy makers on: a) the extent to which public health efforts to reduce health inequalities have been successful; and b) the extent to which inequalities can be further reduced. In Canada, our knowledge of educational inequalities in cancer mortality is limited, in part due to the historical lack of individual-level indicators of SEP in mortality databases. Reducing socioeconomic inequalities in health has become a priority goal of Canadian health policy and cancer mortality may represent a considerable part of the excess mortality carried by lower educated groups. Thus, necessary steps towards health inequality reduction involve the assessment of recent changes in inequalities, the estimation of main contributors to these changes as well as quantitative estimates of the potential for reductions.
My thesis is organized in two components aimed at advancing our knowledge on inequalities in cancer mortality in Canada. In Component 1, I focus on the measurement of cancer mortality inequality across education groups as well as the estimation of changes in the gradient over a period of two decades. I will estimate the contributions of different cancer types to the educational gradient and assess whether these changed over time. Furthermore, I will decompose the changes in educational inequality into the proportion explained by changes in education-specific mortality and the proportion explained by shifts in the distribution of education in the population. Component 2 is a natural extension of inequality description and centers on the potential for reduction of cancer mortality inequality through differential changes in the distribution of a major risk factor: smoking. For both components, I am taking advantage of census data and national surveys linked to mortality datasets that have been recently created and which provides a unique opportunity to analyze mortality inequality at the national level.



 Ailin He

 Affiliation: McGill University

 Project Title: Three Empirical Essays on Program Evaluation Focus on Childcare Policy and Immigration Law

 Project Summary: My thesis is comprised of three empirical essays, focusing on evaluating two programs of interest. The first program is a before- and after-school care policy, implemented in the province of Quebec in 1998. This program provided childcare services to primary-school children on school premises before school time, during lunch and after school. This unique program not only reduced the cost of after-school care to $5 per day, but also increased the provision of care substantially. Using this policy as a natural experiment, we study the causal effects of after-school care’s expansion and subsidization on child’s development as well as maternal labor market outcomes.

Chapter 1 focuses on studying the causal impact of this program on child’s cognitive, non-cognitive skills, health outcomes as well as habit formation. Using the Canadian National Longitudinal Survey of Child and Youth (NLSCY), we adopt a difference-in-differences methodology to compare primary school children in Quebec before and after the reform, to the same school-grade cohorts in the rest of Canada. The policy effectively increases the use of after-school care by 8 percentage points. The intent-to-treat estimates further show a deterioration in child’s overall non-cognitive development but an improvement on their health outcomes. However, these effects do not persist in the medium run.

Chapter 2 examines the effect of the same program on maternal labor market outcomes, since childcare options are closely connected to mothers’ labor supply decisions. We define mothers in Quebec with the youngest child aged 6 to 11 as the treatment group, and mothers with the youngest child in the same age cohort in the rest of Canada as the control group. Using the Survey of Labor Income Dynamics (SLID), we analyze mother’s labor market outcomes. Our results show that the after-school care reform increases maternal employment on the extensive margins by 2-3 percentage points, but has no significant effect on employment intensity. Further examination of heterogeneous samples reveal that the policy effect on maternal labor supply is driven exclusively by highly educated mothers from lower non-maternal income households.

Chapter 3 turns to another policy targeting immigrants to Canada. This chapter attempts to uncover citizenship premiums on labor market outcomes. To identify the causal effect of citizenship, we make use of changes in the Canadian Citizenship Act of 2014, which extended the physical presence requirement for citizenship from 3 to 4 years. After addressing selection issues, a difference-in-differences methodology is employed to compare changes in labor market outcomes of equivalent immigrants, who only differ from each other with respect to their eligibility for citizenship due to the revamped residency requirement. Using the Canadian Labor Force Survey (LFS) along with the Permanent Resident Landing File (PRLF), our results suggest that delaying citizenship eligibility by one year imposes significant impacts on both the extensive and the intensive margins of labor supply. Even though affected immigrants tend to participate more actively in the labor market during the selected periods after the new law has been implemented, their wage earnings are negatively affected.

 Boriana Miloucheva

 Affiliation: University of Toronto

 Project Title: Essays in Health Economics

 Project Summary: The feeling of falling behind has been identified in recent research as a potential factor behind the rise in morbidity and mortality  among certain population groups in developed economies, such as the US and Canada. Improving our understanding of such dynamics is important, but identifying the health effects of changes to people's relative economic situation separately from the health impact of changes to their absolute level of income is not straightforward. First, both absolute and relative income are likely endogenous determinants of health. Second, changes to individuals' absolute income levels often simultaneously affect their position within the income distribution.

To address these challenges, this paper proposes an empirical strategy that draws on the importance and geographic concentration of the extractive industry in Canada. To deal with the potential reverse causality characterizing the relationship between health and income, we exploit exogenous movements in the price of oil, which predominantly affect the earnings of workers in the extractive industry. Oil price variations further induce different combinations of changes to absolute and relative income across individuals, based on their own labour market activity and on the share of their neighbours employed in the extractive industry. Using hospitalization records linked to census data, we capitalize on these combinations to investigate the extent to which people's absolute and relative income trajectories separately contribute to the development of severe health conditions and to the utilization of inpatient care. Our results shed new light on mechanisms through which income inequality might affect people's well-being.

  Helen Cerigo

  Affiliation: McGill University

 Project Title: An exploration of the pathways from early-life adversity to adult mental health in Canada

 Project Summary: Major depression is a commonly occurring, seriously impairing, and often recurrent mental disorder.
There is a significant body of evidence that demonstrates that adverse early life conditions, such as abuse, neglect and exposure to violence, are associated with depression in adulthood. Adversity in childhood is common, with upwards of a third of the population in Canada reporting these experiences. The knowledge about the processes through which childhood adversity impacts depression in adulthood is emerging. There is converging neurobiological and epidemiological evidence that chronic stress in childhood impacts brain development, specifically the stress-response system, increasing vulnerability to future stressful life events (stress-sensitization). The relationship may also work through an accumulation of stress and a reduction in opportunities through the life course whereby adversity in childhood sets off a cascade of experiences that increase adult life stressors, such as divorce, unemployment and financial crisis, which often precede depression.
Using two representative Canadian datasets, I propose to explore the association between childhood adversity and depression, and its associated pathways, with an emphasis on the broader social context. I will explore these pathways in Canada, both in the general population and the Indigenous population, a population that warrants special consideration due to a high risk of depression and a history of systematic childhood adversity caused by government-mandated residential schooling. Guiding this thesis is the integration of structural determinants and life course frameworks to understand the formation and maintenance of social gradients and health distributions. Specifically, in Part 1 of this thesis I will explore the stress-sensitivity pathway to depression (Objective 1.1) and the mediating potential of education, income and use of income protection programs in the development of the childhood adversity-depression association (Objective 1.2) in the general Canadian population. In Part 2, I will conduct a literature review on the impact of childhood adversity on health among Indigenous Peoples in Canada (Objective 2.1) and test whether residential schooling constitutes an instance of early life adversity generating a stress-sensitization response (Objective 2.2). An understanding of the pathways leading to depression can help to identify intervention points to prevent the potential long-run harmful health and social effects of childhood adversity.

  Brittany Etmanski

  Affiliation: University of Waterloo

  Project Title: Beyond Academia: Exploring the Career Pathways of Social Science PhDs

  Project Summary: Although most Ph.D. graduates pursue non-academic employment we know little about the non-academic pathways other  disciplines pursue or whether the hard and soft skills required in non-academic careers differ substantially from those within academia. Drawing on the 2013 National Graduates Survey (NGS) as well as primary qualitative and quantitative data, this research intends to broaden our understanding of the career pathways available for Ph.D. graduates. This examination will draw on theories of New Institutional Theory, the education-economy nexus, credentialism, forms of capital, and rational choice theories, and will demystify the specific career pathways of recent graduates, and outline what skills and experiences aided (or hindered) their career pathwaysAlthough most Ph.D. graduates pursue non-academic employment we know little about the non-academic pathways other disciplines pursue or whether the hard and soft skills required in non-academic careers differ substantially from those within academia. Drawing on the 2013 National Graduates Survey (NGS) as well as primary qualitative and quantitative data, this research intends to broaden our understanding of the career pathways available for Ph.D. graduates. This examination will draw on theories of New Institutional Theory, the education-economy nexus, credentialism, forms of capital, and rational choice theories, and will demystify the specific career pathways of recent graduates, and outline what skills and experiences aided (or hindered) their career pathways

Batholomew Chireh

Affiliation: University of Saskatchewan, SKY RDC

Project title: Dimensions of aging: Chronic conditions, Multi-morbidity and Self-rated health in the context of an aging population in Canada

Project summary: Mental illness and cognitive impairment are diseases that are of substantial burden to Canada and the world at large. They are chronic and persistent conditions with considerable associated disability. The links between mental and physical health are well-established. The primary goal of this thesis is to use nationally representative population-based survey data sets to establish the distinctive risk factors and trends of chronic conditions (both physical and mental) and multi-morbidity among middle aged and older adults in Canada. Understanding the evidence surrounding chronic diseases and their determinants is key to interpreting trends and crucial to developing public health interventions that can effectively reduce rates of chronic diseases and improve the population's health and quality of life. 

 Elham Adibnia

 Affiliation: University of Calgary

 Project Title: The Causal Effect of Environmental Regulations on Health, Evidence from Ontario Coal Phase out

 Project Summary: Ontario had five coal-fired generating station (GS) responsible for generating of 25% of province electricity. In 2005, a cost effectiveness analysis was prepared for Ontario Ministry of Energy. Based on this study, an average annual of 660 premature death, 920 hospital admissions and 1090 emergency room visits, and 331000 minor illness cases could be avoided by phasing out coal-fired generation and replacing it by nuclear/gas generation. Many doctors and environmental activists started a campaign that led the government to issue a regulation requiring the closure of all five plants in a staged approach. As a result, all coal-fired GS was closed from 2005-2014.
In this study, I use coal plants closure in Ontario as an experimental setting to study the causal effect of environmental regulations on population health and migration decision.
The first stage is to investigate the effect of closures on air quality around the plants. I use National Air Pollution Surveillance Program (NAPS) which collects pollution data from pollution monitors in all provinces.
In the second stage, I analyze the effect of closures on infant and adult health. The independent variables are low birthweight and prematurity incidence, infant and adult death. To do so, I use four datasets. Power plants information, from Ontario Ministry of Energy, vital statistics birth database, vital statistics death dataset, and Postal Code Conversion File (PCCF) form Statistics Canada. Using PCCF, I link the plants and vital statistics based on longitude and latitude of the plants and mother’s, or the deceased person’s, residence. I use this geographical coordination to find the distance each birth or death observation has from the nearest power plants using ARCGIS software. I keep all the observations that are located within 40 miles of at least one power plant.
I adopt a difference in difference (DID) model to compare health outcome of individuals (infants) living in 0-20 miles of the power plants to the outcomes of observations living in 20-40 miles of the plants before and after closure. The reason for comparing these two groups is that they both have similar characteristics. However, the first group was more exposed to the pollution from power plants simply because they were closer to the power plants. Therefore, the second group can serve as counterfactual for the first one.
I also use Longitudinal Administrative Databank (LAD) and Census data to study the effect of closures on people’s migration decision from or to areas near coal plants.
This is the first comprehensive study that uses individual level data to investigate the effect of environmental policies on health and migration decision.


Mohammad Chowdhury

Affiliation: Carleton University

Project Title: The choice of job search methods and difference in the labour market outcomes among ethnic immigrants and refugees to Canada

Project Summary: There are three interrelated paper in this proposed research. In the first paper, I will present an empirical study of the determinants of the job-search strategy choice (e.g., informal and formal search) of immigrants and refugees in the Canadian labour market. Specifically, I want to examine how differently socio-demographic factors, human capital, social and ethnic networks, neighborhood economic characteristics, and duration of unemployment influence the search strategy choice. I will further assess how the choice of search methods varies across ethnic groups. The implications of findings of this paper are important because they will identify potential channels through which members of different ethnic groups look for a job in the Canadian labour market.
The second paper will empirically examine the role of the job-search methods, social networks, and ethnic concentration in determining the employment status of immigrants and refugees in Canada. Specifically, I want to examine how the search strategy choice, different forms of social networks, ethnic concentration and neighborhood economic opportunities produce different labour market outcomes. I will investigate the role of ethnic concentration by assessing the question of whether living and working in the coethnic enclave is more beneficial than trying to integrate into the wider economy and community. By augmenting the LSIC microdata with Canada’s 2001 and 2006 Census, this paper will estimate the labour market impacts of both the level of workplace segregation and coethnic concentration for various minority groups and refugees in Canada's large metropolitan areas. The major innovation of this paper is that it will demonstrate whether the choice of search methods, social networks, and enclave characteristics vary among different immigrant groups, resulting in a different labour market outcomes among the groups in Canada. The findings of this paper will help policymakers to learn about whether moving immigrants to areas where their skills are needed most, as opposed to allowing them to settle close to large communities of coethnics, could decrease the likelihood of underemployment amongst immigrants and refugees in the Canadian labour market.

The third paper will empirically assess the influence of social networks, ethnic concentration, post migration human capital, and the Internet-based job search on employment transitions. Specifically, I will study a dynamic process where labour market status, social networks, and change in job search strategy coevolve through time. The main innovation of this paper is that it will examine both the role of enrolment in government-assisted settlement programs and Internet-based job search on the likelihood of employment transitions. Additionally, I will empirically assess whether changes in job search behaviour and the structure of social networks facilitate the transition from low-skilled job to skill commensurate employment. Using the detailed information on employment trajectory and intended occupation provided by the LSIC, I will further assess the labour market outcomes of immigrants and refugees in terms of the time required to access their first job in the intended occupation. The findings of this paper will assist policymakers to modify current immigrant settlement policies with respect to the origin of immigrants.

Liton Chakraborty

Afflilation: University of Waterloo

Project Title: Assessment of environmental justice implications for flood risk management across Canada

Project Summary: Flooding is a significant threat to cities and urban residential neighbourhoods in Canada as the number of extreme weather events and corresponding property damage costs have risen over the past two decades. In response, Canada has embraced a flood risk management (FRM) approach through initiating the 2015 National Disaster Mitigation Program (NDMP) to reduce, or even negate, the effects of flood events in Canada, which fundamentally aligns with the Sendai Framework for Disaster Risk Reduction 2015-2030. The NDMP initiative exclusively promotes individual-level flood risk reduction strategies by “focusing investments on significant, recurring flood risks and costs; and advancing work to facilitate private residential insurance for overland flooding”.

Social justice scholars, however, criticize that the non-governmental activities for FRM which deliberately overlook vulnerability-based justice principle. The Egalitarian’s principle focuses on ‘social vulnerability’ reduction and publicly-funded FRM strategies for the most vulnerable and socially disadvantaged group of communities by ensuring equal protection standards for all, and by enhancing the social capacity of individuals affected. Canadian existing FRM policy paradigm seems to prioritize ‘Utilitarianism’ (maximize utility) and ‘Libertarianism’ (individual responsibility) principles by ignoring the distributional justice consequences. To fulfill the need of the most flood-vulnerable group of communities, FRM policies should prioritize a socially-just approach in accordance with the Egalitarian’s principle.

The environmental justice (EJ) principle, as a form of distributive justice principle, focuses on identifying certain groups of communities (disadvantaged or marginalized or people with low socioeconomic status) who bear a disproportionate share of the environmental burden and have no, or limited, input regarding policy and legislation (Maantay & Maroko, 2009). Aligned with the EJ principle, my Ph.D. research investigates a series of relevant questions, including i) What demographic and socioeconomic factors contribute to social vulnerability to flood risk exposure across Canada? ii) Which geographic locations or groups of communities are most socially vulnerable to flood risk exposure across Canada? iii) Are socially vulnerable groups of communities disproportionately exposed to flood risk in Canada?

Data analysis consists of three phases. First, Principal Component Analysis will be exploited to identify a list of significant variables that contribute to the Social Vulnerability Indices (SOVI). Second, using the 2016 CT Digital Boundaries and the 2018 JBA flood hazard extent data, all CTs in a CMA will be identified whether they are geographically intersected with fluvial, pluvial, and storm surge (coastal) hazard exposure to the extent of the 100-year design flood. SOVI score for each CT will be integrated with the extents of JBA flood risk map data to find out which CT is mostly socially vulnerable to flood risk exposure. Third, a multivariate regression analysis will be used to examine whether racial/ethnic and socioeconomic disparities in flood risk exposure exist after controlling for all other contextual factors.

Consistent with environmental and social justice theory, this research supports the ‘Rawlsian Maximin rule’ - maximizing the opportunities and minimizing the inequalities, differences, and disadvantages in order to direct scarce public resources ‘to the greatest benefit of the least advantaged’

Laura Duncan

Affiliation: McMaster University

Project title:Measurement of child mental health & evaluation of child mental health service targeting and expenditure allocations

Project Summary: This thesis will draw on the 2014 Ontario Child Health Study (2014 OCHS) to address four contemporary and policy-relevant issues associated with measuring child and adolescent mental disorder and child mental health service use in the general population. The first paper focuses on the development of a simple, brief symptom checklist used to measure child mental disorder conceptualized as a dimensional phenomenon, a core concept in the 2014 OCHS. The second focuses on an even briefer version of this measure that could be used to measure child mental disorder dimensionally and categorically both in clinical settings and for population monitoring purposes. The third paper focuses on a substantive question about the appropriate targeting of child mental health services according to need based on a multilevel modelling analysis of 2014 OCHS data and government administrative data. The fourth paper determines the extent to which child mental health service government expenditures in 2014-15 were allocated according to need by developing and applying a funding formula based on child numbers from the Census, need and service use estimates from OCHS data and expenditure allocations from government administrative data. Findings from this thesis make unique contributions to the fields of: a) health measurement and assessment—through the development of psychometrically robust and relevant measures of child and adolescent mental disorder; and b) health policy—through the use of innovative analysis to explore important policy questions about whether children and youth with mental health need are receiving the services they need and whether past service expenditures were distributed according to mental health need.

 Eugena Kwon

 Affiliation: Western


 Project Title: Examining the structural challenges and barriers to immigrants’ post-migration eating habits in Canada 

 Project Summary: I draw on intersectional approaches to explore how various dimensions of inequalities (e.g., gender, race/ethnicity, and class) are intertwined, and may exacerbate disadvantages and reproduce inequalities for immigrant populations. My research interest spans across the following four inter-related research areas: (1) immigrant integration and race/ethnic relations; (2) sociology of work and occupations; (3) gender and professions; and (4) population health and well-being.
My first line of research examines the gender inequalities that women in the professions experienced at the intersection of race/ethnicity and immigrant status. This research explored whether women’s anticipated work-family conflict and gendered experiences in medical school shaped their career specialty choices and family/life considerations. Professional work, and the organizations in which it takes place, are not only gendered, but also racialized, and further shaped by other dimensions of inequality such as immigrant status. Taking an intersectional approach, this research revealed that the challenges experienced by women in professions may be further exacerbated for women of racial/ethnic minority and immigrant backgrounds. This research has been published in peer-reviewed journals such as Gender Issues and Canadian Ethnic Studies, and I also have a forthcoming book chapter which will appear in a volume published by Routledge for the Series on Gender and Organizations. Before being published in Gender Issues, this particular paper received the best student paper award from the Work, Professions, and Occupations (WPO) Cluster of the Canadian Sociological Association.
My current SSHRC-funded doctoral dissertation continues to expand on my research interest in inequalities experienced by those of racial/ethnic minority and immigrant backgrounds, but in a new direction. This research focuses on examining the structural challenges and barriers that shape immigrants’ post-migration eating habits in Canada. A mixed-method approach is employed as the quantitative results are effective at providing generalizable statistical evidence, while qualitative results offer richer data for a better understanding of the problem and the identification of a potential solution. I use the 2015 Canadian Community Health Survey (CCHS) and the 2016 General Social Survey (GSS) for my quantitative analyses. For the qualitative component, I collected a total of 45 semi-structured, in-depth interviews with recent immigrants (those who arrived in Canada within the last ten years) in Toronto and London.

The findings from my dissertation revealed that structural inequalities related to gender and race/ethnicity during the economic integration process (e.g., discrimination in the labour market) shaped immigrants’ post-migration lifestyles and well-being. Labour market inequalities, such as the prevalence of unemployment and underemployment – contributed to not only financial constraints, but also feelings of time scarcity among immigrants. Transferring their foreign credentials and upgrading their qualifications – combined with other structural barriers – created competing demands that de-prioritized healthy lifestyles. These experiences also varied significantly by the admission class (e.g., skilled, family, refugee). This research provides new insight on how the challenges of immigrant integration are exacerbated by gender, race/ethnicity, and class – and its implications for their post-migration eating habits and the inequalities that shape the way they engage in a healthy lifestyle.  


Michela Planatscher

Affiliation: University of Ottawa, COOL RDC

Project title: The effect of charitable activity on Aboriginal communities

Project summary: The purpose of this research is to advance the existing literature in the area of Aboriginal economics, by studying empirically the interaction between charitable organizations and the welfare status of Aboriginal peoples. Hence the key question of the study is to investigate whether charitable organizations exert a causal effect on communities’ well-being and other socio-economic outcomes. 

 Aboriginal peoples represent 4.9 % of the Canadian population. There is historical evidence of disadvantage between non-Aboriginal and First Nations, Métis and Inuit populations. Despite improvements in the quality of life in the past decades, severe discrepancies between the two groups still exist. Data from the Aboriginal Affairs and Northern Development Canada (AANDC) show, for instance, that 94.3% of First Nations communities have a below-average community well-being score compared to other Canadian communities. As a consequence, the notion of “closing the gap” has become of major concern for all levels of governmental authorities and policy makers in Canada. Moreover, the philanthropic sector is playing an increasingly important role in helping provide much-needed services to communities. 

The first step of my research consists of combining administrative data on charities from the Canada Revenue Agency (CRA) with Census and Aboriginal Peoples Survey (APS) data through a geographic linkage, that will allow me to have fine details on the characteristics of individuals living in different communities. I will then study empirically the impact of charitable organizations on communities’ well-being using appropriate statistical (econometric) procedures. I expect that endogeneity will be an issue that will need to be addressed. For instance, it is possible that the access to charities variable is endogenous and an Instrumental Variable (IV) approach is the most likely procedure to use in this case.

I expect that I will be able to say how the charitable sector is currently able to help improve community prosperity, especially for Aboriginal communities. Moreover, my results have the potential of identifying another policy lever to help eliminate the existing Aboriginal social problems in the context of the aforementioned notion of “closing the gap”.

  Audrey Minta Appiah

Affiliation: University of Ottawa, COOL RDC

Project title: Essays in Health Economics

Project summary: Rising income inequalities continue to be of major concern for policy makers. Due to the concerns about income inequality, measurements have been developed to monitor not only income inequalities but the socioeconomic inequalities in health over time. Popular indices of socioeconomic inequalities in health, like the concentration indices, are used for analyzing ratio scale variables such as weight and life expectancy. However, when population health is measured using life expectancy, the quality of life is often ignored. Because researchers are interested in quality of life, the use of self-assessed health status which is ordinal in nature as a measure of health has become popular in this literature. Researchers also interested in other well-being measures also use categorical variables such as happiness, life satisfaction etc. However, using the popular socioeconomic inequality measures for a categorical variable raises another issue since these indices are not readily applicable to ordinal variables (Erreygers, 2006; Makdissi and Yazbeck, 2017). Also, these inequality indices are "mean based", in that, they measure deviations from the mean. 

Allison and Foster (2004) argue that using the mean as the reference point for the measurement of health inequality in the presence of categorical variable is not robust because of the sensitivity of the mean to any chosen numerical scale for the ordinal variable. To illustrate, the Canadian community health survey has 5 categories to the question, "Would you describe yourself as being usually": happy in life, somewhat happy, somewhat unhappy, very unhappy with little interest in life and so unhappy that life is not worthwhile. Consider two numerical scales [1, 2, 3, 4, 5] and [1, 2, 3, 4, 10]. Assume that the distribution of health is [2, 2, 2, 2, 2], then the mean under the first and second scales are 3 and 4 that is "somewhat unhappy" and "somewhat unhappy" respectively but the median is "somewhat unhappy" in both cases. The authors suggest using the median as the reference in ordering inequality in health (univariate measure). Following this approach, this research aims at extending the idea of robust ordering of socioeconomic inequalities in health for categorical variables using the median as the reference point and empirically estimating median based socioeconomic inequalities in wellbeing (happiness and life satisfaction) using the indices that will be developed in the theoretical section and data mainly from the Canadian Community health Survey and the General Social Survey. This can be achieved by considering a socially weighted average deviation about the median.

The main contributions of this research are, first, we intend to derive properties for the social weighted function so that it follows the principle of income-related transfers. Second, our aim is to provide dominance test for robust ranking of median based inequalities and finally, to suggest statistical inference.


Dylan Simone

Affiliation: University of Toronto, Toronto RDC

Project title: The Financial Inclusion Illusion: Immigrant Financial Literacy, Wealth, and Debt Trends in Canadian Cities

Project summary: My dissertation research is focused on analyzing the relationships between immigration policy, financial literacy programs, theories and practices of financial inclusion, and processes of financial innovation. I explore three over-arching topics in my dissertation. First, I analyze the socio-spatial determinants and implications of wealth and debt levels among immigrant households and neighbourhoods in Canadian cities. Second, I assess the tenets and assumptions undergirding the promise and practice of immigrant financial inclusion as it relates to homeownership, financial literacy, and asset-based welfare. Third, I probe the place of migrant and capital flows in financialization processes as they pertain to financial innovation.

My dissertation involves both quantitative and qualitative research, and I draw on these to intervene in key theoretical debates surrounding financial innovation, literacy, and inclusion. Quantitatively, I analyze the master files of two Statistics Canada surveys – the Canadian Financial Capability Survey (CFCS: 2008, 2014) and the Survey of Financial Security (SFS: 1999, 2005, 2012, 2016), within the Toronto Region Statistics Canada Research Data Centre (RDC). From my analysis of this data I will be submitting two articles for publication in March 2019; the first targets the journal Economy and Society, while the second will be submitted to Economic Geography. Qualitatively, I provide a textual analysis of transcripts from a House of Commons investigation into the sales practices of Canada’s ‘Big Six’ banks. I demonstrate that Canadian financial policy is guided not by linear, ‘rational,’ scientific epistemologies, but rather, by resourceful improvisation, with policy innovations evolving in non-linear, path-dependent trajectories, using the tools at hand to deal with unanticipated events. My third dissertation article stems from this work and will be submitted in January 2019 to Environment and Planning A. Theoretically, I advance debates on the role of finance in political economic value theory vis-à-vis regressive redistributions of financial risk. Using the Financial Consumer Agency of Canada as example, I argue that financial literacy programs, like those designed and enacted in Canadian policy, enable the production and offloading of financial risk from the state and industry to households and individuals. The fourth and final article of my dissertation, which relates to this work, will be submitted in January 2019 to Antipode.

My dissertation research contributes to a growing body of international literature on financialization, housing and mortgage markets, and immigrant financial inclusion, literacy, wealth, debt, and integration. Immigrant wealth accumulation is increasingly seen to be more important than immigrant income in the scholarly literature as an accurate proxy for integration, and much of the published work has focused on the United States, in context of the history of racialized financial exclusion. Investigations of debt and wealth are dominated by neoclassical economic approaches, and spatial perspectives in the scholarly literature on financial inclusion and financial literacy are scarce. I aim to challenge these perspectives, drawing on the Canadian context.


Khrisha B Alphonsus

Affiliation: University of Saskatchewan, SKY RDC

Project title: Multiple sclerosis: Risk perception, symptoms and treatment

Project summary: Multiple sclerosis (MS) is a neurological disease of the central nervous system which causes the nerves to deteriorate over time. MS affects individuals between the ages of 20 to 45 years of age and both biological and environmental factors are associated with the disease. Canada has one of the highest rates of MS along with several European countries. The definitive cause and cure of MS had not been determined and because of this the quality of life of people living with MS is poor. Two projects with RDC will be used in my PhD thesis. 

Disease modifying medications (DMTs) can help to slow down the progression of the disease but they cannot reverse the lesions that have already taken place. Up to now there hasn’t been many studies that have been conducted on whether using complementary/alternative treatments (CAMS), rehabilitation therapy or other conditions impact medication adherence in positive or negative ways. Due to adverse side effects of medication use people often turn to CAMS such as natural health products, herbs, homeopathic medicine, vitamins, acupuncture or exercise. The first objective of the study was to examine whether use of CAM as well as other types of therapies (rehabilitation and counseling) were associated with MS medication use. This research will help practitioners be aware of the reasons behind why patients don’t adhere to medications and whether using social behavioral models could be effective in changing patient’s behaviors. The Survey on Living with Neurological Conditions (SLNCC) 2011-2012 linked to the Canadian Community Health Survey (CCHS) 2010-2011 will be used to perform the analysis. A logistic regression model will be used to examine these associations. 

The second objective was to examine the psychosocial factors associated with pain in people living with MS. Pain is a common problem in people living with MS. The prevalence of pain in MS patients is 50%, and 75% of patients report having pain within one month of their assessment. The presence of pain among patients with MS is associated with increasing age, a longer duration of their illness, depression and increased functional impairment and fatigue. Social support helps in three areas: the first is it provides emotional support such as love and affection, the second is instrumental support such as lending a helping hand to someone and the third is informational support such as from a physician or nurse. Learning to cope with the illness is also important in reducing disease burden. Social support is an important aspect of coping and whether lacking in one type of social support could lead to pain is an understudied area. The main objective of this study is to determine the association between amount of social support and it’s association with odds of pain among individuals with MS. The SLNCC data set (2011-2012) will be used to carry out a logistic regression model which will examine the predictors associated with pain among individuals with MS. 


Megan (Meng) Yu

Affiliation: Memorial University of Newfoundland, Newfoundland branch, NB RDC

Project title: The Emotional Well-being of Immigrants in Canada

Project summary: My dissertation research focuses on whether and how religious involvement, maintaining transnational ties, and acculturation are related to the future emotional well-being of immigrants. Using the three waves of microdata from the Longitudinal Survey of Immigrants to Canada (LSIC), this manuscript-based dissertation devotes each of its three papers to one of the above-mentioned topics. 

Research questions:
1) Does religious participation predict better emotional well-being for immigrants in the future? If so, how does frequency of participation, gender, and religious affiliation affect this relationship? 
2) Does maintaining transnational ties predict better emotional well-being for immigrants in the future? Transnational ties are conceptualized as keeping contact with family and friends in the home country, sending remittance outside Canada to relatives or friends; receiving income from outside Canada; and immigrants’ settlement plan. 
3) Does acculturation predict improved emotional well-being for immigrants? Acculturation is conceptualized as host country education and language training.

My research employs quantitative methods. For each of my three papers, data are analysed with descriptive statistics and a series of longitudinal logistic regression models. Random effects models are used because I am also interested in predicting how differences between groups of people identified via such time-invariant characteristics as gender, visible minority status, and immigration class relate to the dependent variable. Interaction terms are used to examine the moderating effects of certain variables on the key independent variables. All-item-interacted models with each of the independent variables interacted with the wave variable were conducted, allowing for a comparison of regression coefficients obtained from different waves of data.


Jeffrey Hicks

Affiliation: University of British Columbia, BCI RDC

Project title: Essays in Taxation and Social Insurance

Project summary: In Canada, and many other countries, unemployment insurance beneficiaries can receive part-time earnings without a full claw-back of their unemployment insurance benefits. Such earnings allowances promote part-time workforce participation and, consistent with the core mandate of Employment Insurance (EI), offer an additional avenue for the liquidity-constrained unemployed to smooth their consumption through income shocks. Aside from promoting part-time work, earnings allowances may also affect how long beneficiaries continue to draw benefits. Simultaneous receipt of earnings and EI benefit increases the payoff of part-time work relative to searching for full-time employment which may cause longer overall claim durations. Selection into unemployment insurance also becomes more attractive. Alternatively, if part-time jobs serve as stepping-stones to full-time employment, earnings allowances could decrease claim durations by putting claimants on track for full-time work. The built-in incentives for part-time work have undergone multiple changes over the past 15 years, offering an opportunity to examine the causal effects of such incentives. In this project, I examine i) the effects of part-time work incentives on work behavior while on claim, and ii) the relationship between work behavior and overall claim duration. I exploit a novel administrative dataset on Employment Insurance claims in Canada and a series of natural policy experiments. 

First, I examine how the pre-2012 non-linear claw-back schedule shapes the distribution of reported earnings. Prior to 2012, claimants could earn up to some threshold without any reduction in their benefits. Additional earnings reduced benefits dollar-for-dollar. This very large discontinuity in the marginal tax rate creates a large incentive for claimants to alter their real work behavior and to mis-report their earnings. I document substantial “`bunching" around these thresholds, suggesting that individuals have substantial flexibility to adjust their earnings levels or that mis-reporting is a non-trivial issue. The potential for mis-reporting raises the practical consideration of how well part-time work incentives can be enforced. Since EI benefits are calculated on a weekly basis, part-time earnings are required to be self-reported at the same frequency, providing ample opportunity to mis-report. The presence of such evasion has real economic consequences. Verification is costly to the government and imposes a compliance burden on firms. Evasion also increases the fiscal cost to the government by effectively increasing net-payments to EI recipients without a corresponding increase in labour supply. 

Second, I examine how increasing the thresholds affects reported work behavior. In 2005 and 2008, the earnings thresholds were increased by 60 percent in subsets of EI regions. These changes altered incentives on both the intensive (how much to work) and extensive (whether to work) margins. Exploiting these policy changes, I estimate the response on both. 

Third, I examine the relationship between work behavior and claim duration. Descriptive evidence suggests that part-time work does not reduce claim durations by improving transitions to full-time employment. Rather, part-time work may increase overall claim length. However, causal inference in this setting is very difficult, and correlation-based evidence should be interpreted with caution. Ongoing work aims to identify the causal effects.


Said Yousef Abdelrazeq

Affiliation: University of Ottawa, COOL RDC

 Project title: Association of Vitamin D Deficiency and Risk of Health Deterioration Among Canadian Immigrants

Project summary: A growing body of literature indicates that, worldwide, immigrants experience health deterioration after their arrival into their adopted country, and moreover, they have lower vitamin D compared to the native-born population. 

The overall objective of this thesis is to evaluate the global vitamin D status of immigrants in a systematic review and meta-analysis. Moreover, the CHMS data will be used to compare the concentration levels of vitamin D among immigrants to native-born population, to investigate the major risk factors for low vitamin D and possible association with disease status (health deteriorations). To our knowledge, no previous studies used the CHMS data, or any other national data to find a possible association between immigration/disease status and vitamin D levels.

This work may partially help clarify vitamin D-related health deterioration in migrants; moreover, to develop a global guideline that specifies sub-populations, in which the evidence and vitamin D-related recommendations might differ from the overall immigrant population.