Maria Abreu, University of Cambridge
Özge Öner, University of Cambridge
Yolanda Pena-Boquete, AYeconomics Research Centre S.L.
Maria Abreu, University of Cambridge – https://www.landecon.cam.ac.uk/directory/dr-maria-abreu
Brexit neighbourhoods: composition and contextual effects in the EU Referendum vote
A number of studies have recently analysed the role of individual and regional characteristics in explaining the extent of the Leave vote across UK districts and wards, but several unresolved questions remain. An important issue is the extent to which people with very similar characteristics in terms of age, education, ethnicity, employment circumstances, and household income voted differently in Remain and Leave areas. This observed geographical variation in voting patterns could be due to contextual effects, where neighbourhood characteristics or social networks affected the decision to vote Leave, or but they may originate from self-selection of individuals into certain neighbourhoods. We investigate the interaction between individual- and constituency-level effects in the propensity to vote Leave in the EU referendum, using data from the British Election Study. Using a Coarse Exact Matching (CEM) method, we find that nearly identical individuals voted differently in Leave- and Remain-voting areas. We analyse whether those differences are due to differences in relevant area-level variables (such as immigration, low growth, and cultural factors), and find that in some cases the differences are persistent, while in others the differences disappear entirely, suggesting that some of the observed effects are entirely driven by selection mechanisms.
Özge Öner, University of Cambridge – https://www.landecon.cam.ac.uk/directory/ozge-oner-dr
When Weak Ties are Strong: Ethnic Enclaves, Information Frictions, and Labor Market Sorting of Immigrants
In this paper, we investigate the effects of ethnic enclaves on the economic prospects of newly arrived immigrants. Using a geo-coded full population longitudinal matched employer-employee data for Stockholm metropolitan region, we create dyads between residential neighborhoods and potential workplace neighborhoods for newly arrived unemployed immigrants from the Balkans and the Middle East, who arrived to Sweden during two distinct time periods following wars. By estimating a conditional logit model on the first workplace location, we control for unobserved individual heterogeneity, as well as general enclave effects stemming from the residential neighborhood. We find that immigrants are more likely to find their first jobs in locations where many ethnic peers from their neighborhoods are employed. Our analysis provides evidence for causal effects from ethnic networks on reducing labor market frictions associated with information on jobs and job locations.
Yolanda Pena, AYeconomics Research Centre S.L. – https://sites.google.com/site/yolandapenaboquete/research
A Different Approach to Quantile Wage Decompositions with an Application to Gender Wage Gaps in the U.K. and the U.S.
Previous approaches on quantile wage decompositions do not identify the workers associated to each quantile. We introduce a quantile decomposition strategy that more closely follows the spirit of conventional decompositions with linear wage models and we associate the workers around the quantile that minimizes the residuals of the decomposition. Additionally, we generalize the methodology to incorporate the central feature of the generalized decompositions, i.e. we identify the separate contributions of favouritism and pure discrimination to the unexplained wage gap. We illustrate the application of our quantile regression decomposition methodology with two labour market samples drawn from the UK and the US.