J ob Search Behavior among the Employed and Unemployed
Abstract:
This paper uses a
new and exhaustive questionnaire on the labor market outcomes of roughly 1,
300 individuals surveyed through the New York Fed’s Survey of Consumer Expe
ctations. We use the data to examine the job search behavior of both employ
ed and non-employed individuals. The data have detailed information on seve
ral aspects of job search, including the duration and intensity of search,
the offers received, the reservation wage of the jobseeker, and the relativ
e importance of various non-wage job benefits. The survey is unique in that
it includes information on the job search outcomes of those employed and n
on?employed, including those who report not searching for work (e.g., we ha
ve information on job referrals and unsolicited offers). It is also unique
in that it has the same information on the job search outcomes that led to
the job of those currently employed. We use the data to examine the relatio
nships between search behavior and search outcomes as a function of labor f
orce status. The survey allows us to examine one’s search effort regardless
of their self-reported labor force status (i.e., regardless of whether the
y are employed, unemployed or out of the labor force). We are able to exami
ne an individual’s current job search, as well as the job search that led t
he employed to their current job. We use extensive information on the types
of methods used, the types of jobs offered, and the reservation wages job
seekers report while seeking work. We find that a large fraction of the emp
loyed (24 percent) is actively searching for work, with about 40 percent of
employed job seekers seeking an additional job rather than a new job. A sm
all but notable fraction of those who report to be out of the labor force a
ctively look for work as well. Both those who seek an additional job and th
ose seeking work while reporting themselves as out of the labor force tend
to look for, and attract offers from, part-time jobs with few benefits. The
unemployed put by far the most effort into job search, especially the shor
t-term unemployed. Nevertheless, the unemployed have the lowest yields in t
erms of employer contacts and offers, of all job seekers. Even employed ind
ividuals not seeking work tend to generate more offers than the unemployed.
In fact, the unemployed account for 38.5 percent of all applications sent
in the month leading up to our survey, but only account for 9.5 percent of
the offers received. In contrast, the 2 employed receive about 67 percent o
f all offers, with over one-third of those going to those that did not look
for work. We also find that job seekers employ a variety of different meth
ods in the search and that offers through referrals and through unsolicited
contacts play a large role in the hiring process.
We further find that, c
onditional on receiving an offer, the unemployed continue to fare far worse
than employed job seekers. Offers to the unemployed pay lower wages, offer
fewer hours, and are much less likely to offer benefits (about two thirds
of offers to non-employed individuals did not offer any employer-provided b
enefits).The offers are relatively lower than their last wage, while the of
fers of the employed are about equal to their current wage. The offers are
also relatively lower than the unemployed’s reported reservation wages and
desired hours. Our data on reservation wages and acceptance decisions revea
l that unemployed have lower reservation wages relative to their previous j
ob, whereas employed job seekers’ reservation wage on average is higher tha
n their current job’s wage. However, a surprisingly high fraction of employ
ed job seekers report a reservation wage that is lower than the wage on the
current job. This finding is reminiscent of the literature that found that
many job-to-job transitions are associated with wage declines, though it i
s unclear whether these wage declines are due to selection or differences i
n nonwage amenities. The unique feature of our data allows us to control fo
r a number of factors, such as anticipation of layoffs or quits, and isolat
e the importance of non-wage amenities for job-to-job transitions. Finally
, we analyze the search process that led to the current job, as a function
of labor force status at the time of search. We find that those who started
working on their current job immediately after the previous job ended – ei
ther through quit or layoff – fared significantly better than workers who w
ere hired after a spell of non-employment in terms of wages and benefits. M
oreover, those hired from non-employment were also more likely to be search
ing for work again at the time of the survey. We find evidence that workers
who quit their previous jobs have somewhat better search outcomes relative
to workers who were laid off from their previous jobs. Overall, our result
s suggest that job search is pervasive among both the employed and unemploy
ed, but the unemployed face far worse prospects, despite putting in the gre
atest effort among all job seekers.