UNEMPLOYMENT BENEFIT LEVELS AND SEARCH ACTIVITY

AuthorJonathan Wadsworth,John Schmitt
Date01 February 1993
Published date01 February 1993
DOIhttp://doi.org/10.1111/j.1468-0084.1993.mp55001001.x
OXFORD BULLETIN
of
ECONOMICS and STATISTICS
Volume 55 February 1993 No. 1
OXFORD BULLETIN OF ECONOMICS AND STATISTICS. 55. 1(1993)
0305-9049 $3.00
UNEMPLOYMENT BENEFIT LEVELS AND
SEARCH ACTIVITY
John Schmitt and Jonathan Wadsworth *
I. INTRODUCTION
In standard search models of unemployment, the probability that an
unemployed worker enters a job equals the probability that the worker
receives a job offer times the probability that the job is acceptable to the
worker. An acceptable job pays a wage that exceeds the worker's 'reservation
wage', the wage where a worker is indifferent between working and remaining
unemployed.
Most theoretical analyses of the effect of state benefits on unemployment
have focused on the way that government assistance raises workers' reserva-
tion wages. These models assume that job offers either arive at some constant
rate or by some exogenously-determined stochastic process. Workers review
each offer and reject those which are below their cut-off wage. While it is
difficult to establish a worker's reservation wage, most empirical research has
concentrated on demonstrating the notion expounded by these early search
theories (e.g. Mortensen (1970)) that higher benefit levels prolong the
duration of unemployment spells by reducing the expected return from
employment.
Given the strength of the theoretical predictions concerning benefits and
reservation wages, the evidence on duration and transition effects of benefits
* We are grateful to seminar participants at the LSE, ANU, EALE, and UCL, and also to
Steve Nickell and an anonymous referee for helpful comments. Material from the General
Household Survey, made available through the Office of Population Censuses and Surveys and
the ESRC Data Archive, has been used by permission of the Controller of HM Stationery
Office. The Centre for Economic Performance is financed by the Economic and Social
Research Council.
1
2BULLETIN
is not strong. In Britain, for example, estimated benefit duration elasticities
range from 0.06 to 0.6.' In the United States, with a different unemployment
compensation system, many estimates of benefit duration elasticities range
around 0.4 (see Atkinson and Micklewright (1990) for a comprehensive
review). A notable outlying study by Clark and Summers (1983), however,
finds that in a large panel of male workers, higher unemployment benefit
levels were associated with a higher probability of moving into employment.2
For the Netherlands, van den Berg (1990) finds a very low, positive elasticity
of unemployment duration with respect to changes in the benefit level
(approximately 0.03).
Recent evidence suggests that the unemployed receive and reject very few
job offers. Using data on a cross-section of unemployed British workers in
1982, Jones (1989a) reports that over 85 percent have never received ajob
offer. Holzer (1988) presents a figure of 66 percent in similar results for male
youths in the United States in 1981. Van den Berg (1990), using a panel of
unemployed Dutch men, finds low job offer arrival rates (approximately a
one percent chance per week) and a very high proportion of acceptable offers
(approximately 97 percent for the whole sample). Survey evidence reported
in Jackman et aL (1991) suggests that unemployed men in Britain make a low,
but regular, number of job applications - around one or two a month -
which tend to diminish over the unemployment spell.
The experience of many workers therefore seems to be that they remain
unemployed primarily because they receive few job offers, not because they
reject many. If so, then the simple reservation wage effect of benefits is open
to question. Within the context of search models, it is conceivable that
unemployment benefits may still affect job matching by influencing the rate of
job offer arrival. This seems particularly pertinent, since recent work finding
that receipt of unemployment benefits increases search extensiveness among
the unemployed (see Blau and Robins (1990) for the United States and
Wadsworth (1991) for the United Kingdom), raises the possibility that
expanding benefit coverage could increase the search effectiveness of the
pool of unemployed.
Unemployment-related benefits might alter search extensiveness in two
ways. First, unemployed benefit recipients may use their greater resources to
finance search activity, other things equal. For example, the number of appli-
cations a worker makes may depend positively on the level of income
support. Conversely, search activity could decline if the relative returns to
employment fall as unemployment benefits rise. Second, receipt of state
benefit may increase ties to the labour force through information or incentive
'On this basis, given an average uncompleted duration of 12 months and an estimated
elasticity of 0.3, then a 10 percent increase in benefit levels, for example, would raise average
spell duration by around 11 days.
2effects were positive but not different from zero at standard levels of significance.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT