Does cultural diversity have an impact on wages?
Author: Simonetta Longhi, University of Essex
Date: 19 June 2014
Type of case study: Research
About the research
There's a large literature of research analysing the impact of cultural diversity on people's socio-economic outcomes. This study builds on that research by analysing how cultural diversity impacts the wages of natives (those who are working in their home country). In particular, it investigates whether people living in more diverse areas earn higher wages after controlling for other observable and unobservable characteristics, such as age and marital status.
The results show that cultural diversity is positively associated with wages, but only when cross-sectional data are used; panel data estimations show no impact of diversity. Since natives with comparatively higher skills – and wages – tend to self-select into more diverse areas, cross-section analyses may produce upwardly biased results.
The general conclusion is that even though we do see a correlation between wages and diversity in an area, it is not the level of diversity per se that has an impact on wages.
About the data
The British Household Panel Survey was used for collecting individual data on job characteristics such as occupation and wages. For each individual, the Local Authority District of residence was identified, to which a measure of cultural diversity was associated. This analysis focuses on interviews with working-age white British respondents living in 353 English districts in the period between 2002 and 2007.
The population estimates helped the resesarchers compute the share of ethnic groups at the district level.
Data from the 1991 Census for England were key for computing the historical share of ethnic groups at the district level.
The Labour Force Survey was used to compute the share of ethnic groups for some districts, and to double-check the validity of the measures computed using the population estimates. The survey was used in a sensitivity analysis, to compare its measure of diversity to the one from the population estimates.
Cultural diversity is measured by the number and size of ethnic groups living in each Local Authority District in England. The analysis is based on British Household Panel Survey data. The dependent variable is wages of people employed in England and the explanatory variables include individual characteristics such as gender, age, whether married, working part-time, qualification, occupation, and region of residence. Among the explanatory variables, the models also include a measure of diversity and of population density.
The analysis compares ordinary least squares, fixed effects, and first difference estimators. To validate the measure of diversity used, the researchers compared the size of ethnic groups as predicted by the population estimates to those computed using the Quarterly Labour Force Survey in the largest districts.
Publications and outputs
Longhi, S. (2013) ‘Impact of cultural diversity on wages, evidence from panel data’, Regional Science and Urban Economics, 43 (5), pp. 797–807. doi: 10.1016/j.regsciurbeco.2013.07.004