Ethnicity and men's unemployment dynamics
Lucinda Platt, University of Essex
[Project number 40011]
The data will be used to conduct a project on the comparative employment
situation of men from different ethnic groups in England and Wales. Specifically,
it will be used to:
a) describe patterns of unemployment duration between 1994 and 2004 by ethnic group and religious affiliation, including comparing survival functions for the different ethnic and religious groups
b) analyse the impact on hazards of both unemployment exit and re-entry of ethnic group and religious affiliation, controlling for a limited number of characteristics, using techniques of survival analysis that suited to the claimant count data (Jenkins, 2004).
NB the project has been designed specifically to enable the beta testing
of the claimant count data. It is thus planned as a self-contained project
that can be completed by October 2006. Potential future developments are
Understanding differences in unemployment durations has become a fundamental element of academic approaches to (un)employment analysis and to policy approaches to the unemployed. Moreover, given the striking differences in unemployment by ethnic group and the concern with equalising employment rates across ethnic groups (the subject of a DWP PSA target), applying such dynamic approaches to analysis of ethnic group differences clearly has an important policy-related role as well as being of intrinsic interest.
However, there is little extant work on the dynamics of unemployment or of benefit receipt in relation to how it varies by ethnic group in England and Wales. Exceptions are Thomas (1998a, b) who used the Survey of Incomes In and Out of Work to examine ethnic differences in unemployment relating those to attitudes and commuting differences; and a couple of studies exploiting the panel element of the Labour Force Survey to examine ethnic differences in men's unemployment durations in relation to job search methods (Battu et al, 2004; Frijters et al, 2005). However, in the first case, the data date from the late 1980s and thus can only provide limited information about the contemporary picture and in the latter case, the short duration of the LFS panel (five quarters), as well as the need for pooling to obtain sufficient sample sizes, restricts the detail and extent of our understanding of patterns of unemployment durations. Platt (2006) examined benefit dynamics for families with children in one British city, a study in which she also examined benefit re-entry, a rarely examined but important (in policy terms) aspect of benefit and unemployment dynamics. However, the observation window was again relatively short at 18 months and she considered the dynamics of benefit entitlement across both the low paid and the unemployed rather than a specific focus on unemployment.
There is therefore a clear need for a detailed, and in the first instance
predominantly descriptive account of unemployment durations and their
variation by ethnic group.