Browsing by Subject "Decomposition"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Publication
    Obesity inequality and the changing shape of the bodyweight distribution in China
    (2018) Nie, Peng; Ding, Lanlin; Sousa-Poza, Alfonso
    Using data from the China Health and Nutrition Survey (CHNS), this study analyses changes in bodyweight (BMI and waist circumference) distributions between 1991 and 2011 among adults aged 20+ in China. To do so, we quantify the source and extent of temporal changes in bodyweight and then decompose the increase in obesity prevalence into two components: a rightward shift of the bodyweight distribution (mean growth) and a (re)distributional skewing. Our analysis reveals a clear rightward distributional shift combined with a leftward skewing. Although the relatively large size of this skewing in the first decade analysed reflects an increase in obesity inequality, this inequality growth subsides in the second decade. Nevertheless, over the entire 20-year period, obesity inequality increases significantly, especially among females, younger age groups, rural residents and individuals with low socioeconomic status.
  • Thumbnail Image
    Publication
    Recent developments in gender differences in pay
    (2017) Töpfer, Marina; Beißinger, Thomas
    Gender differences in pay continue to persist, despite decades of equal-pay legislation and the promotion of equal opportunities. This thesis examines differences in pay between men and women in Italy during the period 2005-2014 and puts special emphasis on the effects of sample selection. It decomposes the gender pay gap in different subsamples and identifies drivers of the gap that remained unobserved so far. In particular, it shows the empirical disappearance of the gender pay gap in Italy for public-contest recruited employees. It further reveals that the wage gap between men and women for overeducated workers is mainly explained by generally unobservable characteristics. From the methodological perspective, this work provides two novelties. First, it adds to the literature on quantile-regression approaches by adjusting the wage model based on unconditional quantile regression for sample selection. Second, an alternative estimation approach that builds on the omitted variable bias formula is proposed, in order to directly estimate the change of the gender pay gap and its components over time. The empirical part of this thesis is based on a large Italian data set (ISFOL PLUS 2005-2014). The case of Italy is particularly interesting for the study of gender differences in pay and gender-specific selection into wage work given low levels of the aggregate gender pay gap (approximately 6.0%) on the one hand, and high employment gaps between men and women (more than 20.0%) on the other hand.