How well individuals from ethnic minorities do in the labour market is heavily influenced by
the structure of the particular city in which they live and work. Because of discrimination, if
they live far away from where the jobs are and/or in residentially segregated
neighbourhoods, they are going to have worse outcomes than if they lived closer to jobs or
in more mixed neighbourhoods.
That is the conclusion of a new study of city structure, job search and labour market
discrimination by Harris Selod and Yves Zenou, published in the October 2006 Economic
Journal.
Should anti-discrimination policies differ in London or Birmingham? The surprising answer
is probably yes according to Selod and Zenou’s theory of urban unemployment. Given that
in many countries anti-discrimination policies, if any, are designed at a national level, this
finding calls for a metropolitan approach to labour market policies.
In their theory, the labour market outcomes of ethnic minorities are tightly linked to the
spatial organisation of cities. This is because the degree of residential segregation between
population groups and the extent of the disconnection between places of work and places
of residence significantly affect the different channels through which unemployed minority
workers can find a job.
The theory runs as follows. Residential segregation spontaneously forms in a city when at
least two population groups, for example a majority ‘white’ group and a minority ‘black’
group (or any other relevant minority group), value living with neighbours of the same racial
background. The extent of these racial preferences is such that families are ready to live
very far away from central places where jobs are located just in order to benefit from a
racially homogenous residential neighbourhood.
In this context, the unemployed workers of communities that live far away from jobs tend to
experience difficulties finding a job because of ‘informational frictions’ in their job search.
Their unemployment spells and the community’s unemployment rate are in turn
mechanically increased.
But since blacks are also discriminated against by employers (who may prefer to hire
whites), blacks and whites are not equally affected by distance to jobs. In city structures
where whites reside far away from jobs, physical distance and the labour market
preferences of employers in favour of whites tend to offset one another, so that whites are
little affected by their location within the city.
In contrast, in city structures where blacks reside far away from jobs, distance to jobs and
labour market discrimination against blacks combine to aggravate unemployment.
But this is not the end of the story since finding a job also involves resorting to one’s social
network, which is built on local connections, and thus depends on the extent of residential
segregation. The idea is that unemployed workers who live among other unemployed
workers are less likely to have indirect connections to employers who could hire them. This
further aggravates the unemployment problem of blacks when they are segregated from
whites.
But there is a third group of blacks that Selod and Zenou label ‘status-seekers’ and who do
not choose to isolate themselves residentially from the majority white group. For
unemployed workers in the status-seeking group, the harmful effects of labour market
discrimination and job isolation are partially offset by the possibility to exchange information
with some of their white neighbours.
Selod and Zenou provide a deep analysis of their theory’s implications for labour market
analysts and policy makers. They reach three main conclusions: