HOW EFFECTIVE ARE STATE EMPLOYMENT AGENCIES? JOBCENTRE USE AND JOB MATCHING IN BRITAIN
| Author | Jonathan Wadsworth,Paul Gregg |
| Date | 01 August 1996 |
| DOI | http://doi.org/10.1111/j.1468-0084.1996.mp58003002.x |
| Published date | 01 August 1996 |
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 58,3 (1996)
0305-9049
HOW EFFECTIVE ARE STATE EMPLOYMENT
AGENCIES? JOBCENTRE USE AND JOB
MATCHING IN BRITAIN
Paul Gregg and Jonathan Wadsworth*
I. INTRODUCTION
The state employment agency and its network of services, hereafter
referred to as the Jobcentre, is an integral part of the job matching
process in Britain. Around 70 percent of the unemployed, 30 percent of
employed job seekers and 50 percent of employers utilize the Jobcentre
network as part of their search efforts. Use of the service is free to both
employers and job seekers, though not compulsory unlike in Sweden, for
example. Vacancies are communicated to the nearest Jobcentre, but
information on vacancies in other areas of the country can be obtained by
job seekers. Continued state funding of the Jobcentre network is based
on the number of job seekers using its services and its beneficial impact
on the labour market. This may arise for example, if the Jobcentre did
one, or all, of the following: improved the chances of work for those using
the service; by raising the number of workers using its services (market
share), increased search extensiveness and the aggregate flow into
employment; improved the quality and length of job matches and with it
lifetime earnings. In addition, the service may be particularly helpful
toward those, for example the long-term unemployed, for whom the costs
of alternative methods of job search may exceed the potential benefits.
This paper utilizes Labour Force Survey (LFS) data from 1984 to 1992
to outline the determinants of the decision to use the Jobcentre as a
potential means of locating work and 1992 quarterly LFS flow data to
analyse the effectiveness of Jobcentre use in securing a job match,
controlling for individual characteristics and any sample selection bias
implicit in the choice of Jobcentre. Since most searchers use multiple
*National Institute of Economic and Social Research and Centre for Economic Perform-
ance, London School of Economics. This study was undertaken on behalf of the Employment
Service. The views expressed are those of the authors and do not necessarily represent those
of the Employment Service, though the paper has benefited from their comments and those
of seminar participants at Trinity College Dublin, Birkbeck College EMRU, Andrew Britton
and the referees. 443
© Blackwell Publishers 1996. Published by Blackwell Publishers, 108 Cowley Road, Oxford 0X4 1JF,
UK & 238 Main Street, Cambridge, MA 02142, USA.
444 BULLETIN
methods of search, we also analyse the actual method which secured a
transition.
Section III details the pattern of Jobcentre use between 1984 and 1992
by user characteristics in order to identify the type of job seeker most
reliant on this means of search. Jobcentre usage is highest amongst the
less skilled and long-term unemployed and, significantly, higher during
recessions. Section IV investigates the role of Jobcentres in the transition
to work. The results indicate that the greatest beneficial effect is again
amongst those, the less skilled and the long-term unemployed, who are
least likely to find work. Section V concludes.
II. THEORETICAL FRAMEWORK
According to Layard, Nickell and Jackman (1991) and Narandranathan
and Stewart (1992) among others, workers apply for jobs which offer
wages above their reservation wage. Hence refusals of offers are rare and
due to information asymmetries or the receipt of a better offer.' The
search process involves (i) the collection of details about vacancies, (ii) a
decision to apply, (iii) attending an interview, (iv) taking the offer if made
or resuming search if not. An individual is likely to be at different points
on this cycle for a number of vacancies at the same time. We focus on
stages (i) and (iv) of the process since the intermediate steps are
unobserved in our dataset. The transition equations of Section IV are
therefore best thought of as reduced form.
The probability of securing employment through a Jobcentre is given
by
P(Move and Jobcentre) =P(Move/Jobcentre) *P(Jobcentre) (1)
where P(Move) is the probability of moving from unemployment into
employment and P(Jobcentre) is the probability that the individual uses a
Jobcentre as part of their search strategy. The aggregate outflow prob-
ability is the sum of the job-finding probabilities for each possible search
method. Other things equal, the higher the level of search activity, as
measured by a rise in the number of search methods used, the greater the
chances of locating an acceptable job.2
Choice of search method reflects an individual's notion of the associ-
Erens and Hedges (1980) show that in the recovery of 1987, only 7 percent of their
unemployed sample had received and turned down a job offer.
2In Pissarides (1984), the optimal level of search effort equates marginal cost with the
marginal benefit multiplied by the effect of higher intensity on the hiring probability. This
creates an externality and a rationale for third party intervention in the matching process,
since both sides ignore the effect of a job match on reducing the search costs of the other
side.
© Blackwell Publishers 1996.
To continue reading
Request your trial