Essays on the economics of education and health

Abstract

Study I: This paper investigates the effect of the academic ordinal rank position of Swedish grade 9 students relative to their school peers on future educational achievement and adult earnings. The results show evidence of a positive impact of being more highly ranked in the class, and the effects are concentrated to the top and the bottom of the ordinal rank distribution. High-ability students from low-income families gained the most from having a higher ordinal rank in grade 9. The results contrast with US findings, which suggest a similar impact across the rank distribution.

Study II: This paper studies the effect of a reform that increased school-level autonomy in determining how to allocate time between different subjects in Sweden. It evaluates the impact of the reform using registry data in a Difference in-Differences framework. The results suggest that students' educational outcomes, including the subsequent choice of educational track, were not affected by the reform. However, there are some indications that students in large schools and students from low socioeconomic households may have benefited from the reform.

Study III: Research suggests that increases in gross domestic product (GDP) lead to increases in traffic deaths plausibly due to the increased road traffic induced by an expanding economy. However, there also seems to exist a long-term effect of economic growth that is manifested in improved traffic safety and reduced rates of traffic deaths. Previous studies focus on either the short-term, procyclical effect, or the long-term, protective effect. The aim of the present study is to estimate the short-term and long-term effects jointly in order to assess the net impact of GDP on traffic mortality. We performed error correction modelling to estimate the short-term and long-term effects of GDP on the traffic death rates. The estimates from the error correction modelling for the entire study period suggested that a one-unit increase (US$1000) in GDP/capita yields an instantaneous short-term increase in the traffic death rate by 0.58 (p<0.001), and a long-term decrease equal to −1.59 (p<0.001). However, period-specific analyses revealed a structural break implying that the procyclical effect outweighs the protective effect in the period prior to 1976, whereas the reverse is true for the period 1976–2011.

Study IV: Unemployment might affect several risk factors of cardiovascular disease (CVD), which is the leading cause of death globally. The characterization of the relation between these two phenomena is thus of great significance from a public-health perspective. The main aim of this study was to estimate the association between the unemployment rate and mortality from CVD and from coronary heart disease (CHD). We used time-series data for 32 countries spanning the period 1960–2015. We applied two alternative modelling strategies: (a) error correction modelling, provided that the data were cointegrated; and (b) first-difference modelling in the absence of co-integration. Separate models were estimated for each of five welfare state regimes with different levels of unemployment protection. We also performed country-specific ARIMAanalyses. Because the data did not prove to be co-integrated, we applied first-difference modelling. Our findings, based on data from predominantly affluent countries, suggest that heart-disease mortality does not respond to economic fluctuations.

Keywords: education, ordinal rank, decentralization, timetable, accident mortality, heart-disease mortality, unemployment, GDP