ISER Working Paper Series 2007-06
On-the-job search and job competition: relevance and wage impact in the UK
23 Mar 2007
In the literature job competition is often measured by the unemployment rate. The unemployment rate, however, neglects the possibility that a number of workers might be dissatisfied with their job or hold a temporary position which is close to an end, and might therefore look for a (new) job while employed.
Various studies have found that on-the-job search varies over time and across groups of people, and might have a relevant impact on the outflow from unemployment. For these reasons the unemployment rate is likely to be a biased measure of job competition; the bias might vary in a complex way over the business cycle and across groups of people.
In the UK on-the-job search is a relevant phenomenon: less than half of people who are actively looking for a job are unemployed; the other half engages in on-the-job search.
This paper estimates the direct impact of job competition on wages using data from the UK Labour Force Survey over the period 1993-2005. Measures of job competition based only on unemployment are compared to measures taking into account on-the-job search as well as inter-regional commuting.
The wage impact of job competition is estimated for the whole population and by groups of workers. The results suggest that job competition has a negative impact on wages, and that failing to correctly measure labour supply and demand might lead to an overestimation of the wage impact of job competition. Measuring job competition by the total unemployment rate often generates an overestimation of the negative impact that job competition has on wages, but the overestimation is reduced when the unemployment rate is computed by skill groups. Including on-the-job search by a 'search rate' - computed by skills by adding on-the-job search to unemployment - generates estimates that are very similar to the ones computed using the skill-specific unemployment rate. When job competition is estimated by measuring both labour supply and demand, however, the impact on wages is consistently smaller.
The estimations of the wage impact of job competition are fairly robust across the different measures only for some sub-groups of the population (e.g. men, workers with no qualification); for the other subgroups (e.g. women, workers with high education), changes in the way of measuring job competition lead to rather unstable estimated impacts on wages. The results also seem to suggest that job competition might have a slightly bigger impact on some groups (e.g. married workers) than others (e.g. unmarried workers).