Working Paper Series

Horan D, Flannery D, Cullinan J (2024) What Explains Socioeconomic Inequality in Study Abroad Participation? New Evidence from Large-Scale Administrative Data (WP 2024/01)

While studying abroad as part of a degree programme is increasingly common, there are widespread concerns around socioeconomic inequalities in participation. Using large-scale high-quality administrative data from Ireland, we show that students from affluent backgrounds are 1.5 times (46%) more likely to study abroad than non-affluent students. Applying a Gelbach decomposition, we find that prior academic performance and field of study explain most of the observed difference. We also show, for the first time, considerable heterogeneity in the relationship between participation and socioeconomic status by field of study and that inequalities are much greater for high-performing students. Full Paper 2024/01.

Cullinan J, Flannery D (2023) The Commuting Paradox for Female College Students (WP 2023/02)

We study the relationship between living arrangements, commute time, and wellbeing for full-time undergraduate college students in Ireland. Exploiting geographic variation in system-wide accessibility to higher education as an instrumental variable, we use a partial identification approach to show that living at home reduces wellbeing by between 0.07 and 0.13 of a standard deviation. We find these effects are driven mainly by female students and show that long commutes are independently associated with very large increases in poor wellbeing for female students living at home. Our results challenge the theory that disutility from commuting is compensated by other factors relating to where an individual lives, providing new evidence on the so-called commuting paradox. Full Paper 2023/02.

Gillespie T, McHale P (2023) Wind Turbines and House Prices Along the West of Ireland: A Hedonic Pricing Approach (WP 2023/01)

Wind energy will become the dominant source of electricity in Ireland, with the most ideal sites for wind electricity generation located along the west and southern coast. Despite the benefits associated with wind energy, wind turbines impose undesirable externalities on residents mainly through aural and visual pollution. In this paper, I perform a preliminary evaluation on the effect of wind turbines on listed house prices in Ireland. I employ a unique dataset of exact turbine locations with housing and amenity data in seven counties along the west and southern coast of Ireland. With this I conduct a hedonic pricing analysis incorporating spatial and temporal fixed effects. The analysis finds a robust and significant reduction in property value of -14.7% within 1km of a turbine. The effect increases with turbine height, count, and level of urban influence. However, there is evidence that the price effect decays over time, becoming insignificant after 10 years. Furthermore, exhibited effects likely persist beyond 1km, although they are not significant in this analysis. In short, the results presented in this paper are consistent with existing European studies, enforcing the recommendation that turbines should be constructed in highly remote areas to minimise impacts on residents. Full Paper 2023/01‌

Gibney G, McDermott TKJ, Cullinan J (2022) Extreme Temperature, Morbidity and Behaviour: Evidence from A&E Attendances in England (WP 2022/01)

Climate change is expected to lead to increases in the prevalence of extreme temperatures with potentially significant consequences for human health. This paper investigates the relationship between temperature and morbidity in a high-income country with a relatively mild climate and considers the role of behavioural responses to extreme temperatures. Using weekly data on accident and emergency (A&E) attendances for 429 hospitals across England over the period 2010-2015, we find that while higher temperatures are in general associated with significant increases in hospital attendances, there are distinct effects evident across the temperature distribution. In particular, while cold weather is associated with lower contemporaneous A&E attendances, this effect appears to be entirely attributable to displacement of A&E visits to subsequent weeks. In contrast, for hotter temperatures, we find evidence of substantial contemporaneous increases in weekly A&E attendances that are not offset by subsequent reductions. In a setup that includes hospital, week, year, region-by-week, and region-by-year fixed effects, we estimate that for weeks with maximum temperatures exceeding 28^oC, A&E attendances increase by 7.8% relative to weeks with maximum temperatures of 10-13^oC. Over the subsequent four-week period, the estimated net increase in A&E attendances remains large at 7.5%. Overall our results are consistent with differences in individual-level behavioural responses to extreme cold and hot temperatures in England, which have important consequences for health outcomes and health system capacity, particularly in the context of increasingly frequent and intense extreme heat days as a result of climate change. Full Paper 2022/01.  

Vila D, McDermott TKJ (2021) On the Frontlines: An exploratory analysis of unequal exposure to air pollution and COVID-19 in the United States (WP 2021/07)

Recent literature has suggested a link between poor air quality and worse COVID-19 outcomes. In the United States, this link is particularly noteworthy because of residential sorting along ethnic lines within the US population; minorities are disproportionately exposed to health hazards, including air pollution. The impacts of the COVID-19 pandemic have also been disproportionately concentrated amongst minorities. We explore the association between air quality and COVID-19 outcomes, using county level data for the United States from the first wave of the pandemic in 2020, and test whether exposure to more polluted air can account for some of the observed disparities in COVID-19 outcomes among minorities. ‌Full Paper 2021/07.

Castells-Quintana D, del Pilar Lopez Uribe M, McDermott TKJ (2021) Population Displacement and Urban Conflict: Global Evidence Fom More Than 3300 Flood Events (WP 2021/06)

In this paper, we study the effect of displacement of population into cities on urban conflict in developing countries. To do so, we construct a novel measure of exposure to floods, using data on more than 3,300 flood events worldwide, as an exogenous source of population displacement. We combine this with city level observations of more than 9,000 urban social disorder events. Exposure to floods is found to be associated with higher likelihood and frequency of urban social disorder. Our evidence suggests that the effects of floods on urban disorder occur mainly through the displacement of population into large cities. Exploring the information on urban disorder events in more detail, we find that the association between city growth and urban disorder is strongest for events related to public service provision, wages and food prices. Full Paper 2021/06. 

Lin Y, McDermott TKJ, Michaels G (2021) Cities and the Sea Level (WP 2021/05)

Construction on low elevation coastal zones is risky for both residents and taxpayers who bail them out, especially when sea levels are rising. We study this construction using spatially disaggregated data on the US Atlantic and Gulf coasts. We document nine stylized facts, including a sizeable rise in the share of coastal housing built on flood-prone land from 1990-2010, which concentrated particularly in densely populated areas. To explain our findings, we develop a model of a monocentric coastal city, which we then use to explore the consequences of sea level rise and government policies. Full Paper 2021/05

Henry E, Cullinan J (2021) Mental Health Spillovers from Serious Family Illness: Doubly Robust Estimation Using EQ-5D-5L Population Normative Data (WP 2021/04)

People are interconnected and ill-health is rarely experienced in isolation. However, while there has been extensive research on health spillovers related to informal caregiving, there is comparatively little evidence on how ill-health may impact upon non-caregiving family members. This paper analyses EQ-5D-5L normative data from a nationally representative sample of adult residents of Ireland to estimate the independent relationship between serious family illness and five distinct dimensions of health. The empirical strategy combines inverse probability weighting and multivariate ordered probit regression in a doubly robust estimation. We find that experience of serious family illness is associated with large mental health decrements that are independent of caring responsibilities, while similar results are not evident for the four other health dimensions. Furthermore, stratified sub-sample analyses indicate considerable heterogeneity by sex and by income. In particular, we find evidence consistent with larger mental health spillovers for females than for males, as well as for low and medium income households relative to high income households. The latter suggests that such spillovers may be substantially worse for those with fewer resources. Overall, the findings have a range of potential implications, including for the provision of mental health supports and services, for equity of health outcomes, as well as for health economic evaluation. For example, we calculate that our estimates of health spillovers are consistent with a 1.3% reduction in health utility for non-caregiving family members.‌ Full Paper 2021/04

Cullinan J, Flannery D (2021) Distance Matters: Geographic Accessibility and Higher Education Participation Decisions (WP 2021/03)

There is extensive evidence that substantial inequalities persist in relation to higher education participation and outcomes in many countries. One potential barrier to participation is geographic accessibility, which can lead to a wide range of direct and indirect costs. This paper examines the theoretical and empirical economics literature on the role and importance of geographic accessibility for a range of decisions relating to higher education participation. In doing so, it reviews three main methodological approaches used to estimate the impact of distance, namely those that model individual student choices, those that consider aggregate student migration flows, as well as approaches based on stated preference techniques. In addition, the chapter presents an illustrative example of the importance of travel distance based on recent data from Ireland. In particular, it uses Sankey diagrams to illustrate the highly localised nature of school to higher education transitions and gravity models of student migrations to measure the elasticity of flows with respect to distance. The paper’s overall main conclusion is that distance matters for students, higher education institutions, and policymakers, and it finishes with a discussion of a range of proposed policy responses that aim to address various equity and efficiency concerns related to disparities in geographic accessibility to higher education. Full Paper 2021/03

Cullinan J, Flannery D, Palcic D (2021) Study Abroad Programme Participation and Subsequent Academic Performance: Evidence from Administrative Data (WP 2021/02)

There is increasing policy attention on international student mobility programmes in higher education. However, there is little evidence on how participation might impact students’ subsequent academic performance. This paper uses administrative data from Ireland to examine the relationship between spending a semester studying abroad and academic outcomes on return. The results suggest that while study abroad is not related to improved academic performance on average, there are differences across student groups. In particular, male students from poorer backgrounds do worse after studying abroad than their peers, higher-achieving students do better, while language students perform better on return in language subjects. Full Paper 2021/02

Gillespie T, Lyons R, McDermott TKJ (2021) Information Matters: Evidence from Flood Risk in the Irish Housing Market (WP 2021/01)

In this paper, we test the effect of new information about flood risk on housing prices in Ireland. Our core finding is that information matters. Sales prices responded dramatically to the release of flood risk maps in 2011, with the emergence of a 3.1% price discount for dwellings at risk of flooding. This flood discount is not observed prior to the release of the new risk information, for dwellings defended by flood relief schemes, or for rental dwellings. We also document public attitudes to flood risk through survey findings that suggest a continuing information deficit in relation to flood risk.‌ Full Paper 2021/01

Cullinan J, Connolly S, Whyte R (2020) The Sustainability of Ireland’s Health Care System (WP 2020/02)

This paper provides an assessment of the sustainability of Ireland’s health care system. It starts by describing the historical development of the Irish system and identifying key features of the current system that raise potential challenges for sustainability. It then provides an analysis of recently compiled and up-to-date data on trends in health care expenditures. A number of specific demand and supply side challenges to sustainability are then described and discussed. This is followed by an examination of recent and current reforms to the health care system, focussing on their likely impact on sustainability, as well as a discussion of how health economics has and can inform policy, practice and debate. We also discuss the potential implications of the COVID-19 pandemic for the Irish system. Full Paper 2020/02

Cullinan J, Flannery D, Harold J, Lyons S, Palcic D (2020) The Disconnected: Covid-19 and Disparities in Access to Quality Broadband for Higher Education Students (WP 2020/01)

The Covid-19 pandemic forced many higher education institutions (HEIs) across the world to cancel face-to-face teaching, close campus facilities, and displace staff and students to work and learn from home. Given the persistent nature of the pandemic, many HEIs are now delivering more courses online and/or using a blended learning approach. However, there are concerns around differences in student access to digital learning resources while at home, particularly high quality broadband connectivity. This is important, since variation in connectivity may impact the type of online/blended model that faculty can deliver or constrain student engagement with online content. In this context, we combine national data on the domiciles of students enrolled in Irish HEIs with detailed spatial data on broadband coverage to estimate the number of higher education students ‘at risk’ of poor accessibility to high quality internet connectivity. Overall we find that one-in-six students come from areas with poor broadband coverage, with large disparities by geography and by HEI. We also find that students with the poorest broadband access are more likely to be socioeconomically disadvantaged. As a result, we recommend that HEIs use their detailed registration data to help identify and support at-risk students. In particular, our results suggest that some HEIs may need to prioritise access to campus facilities and services to less well-off students living in poor broadband coverage areas. Given the global nature of the issues raised in this paper, our findings and recommendations are relevant to HEIs and policymakers across the world. ‌Full Paper 2020/01