Memorandum from WS Siebert and Kirsten
Daniel, University of Birmingham Business School
RECRUITING THE UNEMPLOYEDRESULTS FROM
RESEARCH INTO THE HIRING OF PRODUCTION WORKERS
1. To find ways of encouraging employers
to hire unemployed people, it is important to understand the hiring
process. In this submission, we report the results of a programme
of research into the hiring of production workers (see Siebert,
1999). The research uses personnel records from a sample of multinationals
with matched plants in the UK, the US and some continental European
countries including Italy, the Netherlands and Belgium. The attached
Appendix details our methods. While the sample is quite small11
manufacturing plants over 20 years 1975-94the results have
the advantage of being based on what firms have actually done.
The research complements studies based on countrywide data, and
should therefore be of interest to the Committee.
2. Our research considers both company "hiring
standards"new recruits' age, previous experience,
and educationand whether recruits' contracts are temporary
or permanent. Hiring standards and contract status are measured
from the recruits' personal records. (For the UK and US plants
we also have data on whether the recruit was unemployed prior
to recruitment, but have not analysed this aspect yet.) The lower
are a company's hiring standards, the easier it is for the less
qualified worker to obtain a job in the company. Put in another
waysince most unemployed workers are drawn from the less
qualified end of the spectrumthe lower are a company's
hiring standards, the easier it is for unemployed workers to find
a job in the company. And if the jobs are permanent, so much the
better. Our research thus provides a framework within which to
consider how Government can encourage employers to recruit unemployed
3. We here report briefly some broad results
of the approach that are likely to be of relevance to the Committee.
In the discussion, it is to be understood that many factors influence
hiring standards, so it is necessary to think of each of these
as being held constant while we conceptually vary one factor at
4. To begin with, the unemployed worker
is the classic "outsider". The "outsider"
category comprises workers who have few formal skills, and hence
tend to be difficult for the employer to assess, and so "risky"
to employ. Moreover, the outsiderunlike the incumbent "inside"
workerhas no ties to the firm in the form of firm-specific
skills, or union protection, or employment protection legislation.
The question is whether policies to reduce insider power via reduced
union power or weaker employment protection will make the firm
respond by being less choosy in recruitment, that is, by lowering
hiring standards to the advantage of unemployed workers. Similarly,
policies to educate unemployed workers and/or certify their skills
may also make the firm respond by being less choosy in recruitment,
again to the advantage of the unemployed. Let us take the issues
of union power, employment protection, and education/training
5. First, union power. Our findings are
that plant hiring standards tend to increase when national-level
union membership increases, other things equal. The link is not
always significant, admittedly, but it is consistent with the
findings of Nickell and Layard (1997) based on countrywide aggregates
(using a panel of OECD countries). Their results indicate that
increased union power is associated with increased unemployment
rates and longer unemployment duration, other things equalin
other words, union power has an adverse impact on recruitment
of the unemployed. Importantly, however, they find that this association
falls away if there is "co-ordinated" union-employer
bargaining, since co-ordination moderates wage pressure (1997,
48). This co-ordination variable is absent from our study. In
sum, data at plant and country level suggest that strong unions
have adverse effects for recruitment of the unemployedbut
responsible unionism might be able to control these effects.
6. For the UK, the new statutory union recognition
procedure of the Employment Relations Act is likely to bring about
a large increase in union membership. The increase could be of
the order of 1 million workers, with recognition eventually won
in about 1,000 extra firms (for discussion see Addison and Siebert,
1999). The simple policy implication is that this extra power
does not augur well for unskilled recruitment. The Committee needs
to think of ways to ensure the power is exercised responsibly.
7. Second, employment protection. Since
employment protection legislation is many-dimensioned it is difficult
to find a cardinal measure which enables us to say "how much"
protection there is, or how much it has changed over time. Moreover,
where employment protection is via case law rather than statute,
as is the case in the US, it is easy to be misled into thinking
that the US is less regulated than it is (see below). Our study
uses two methods. The first is a simple on-off dummy to track
changes in employment protection laws. For example, the 1985 change
in the UK lengthening the period of service to claim unfair dismissal
from one to two years is taken as a move from "on" to
"off" (see Annex). A separate dummy tracks restrictions
on temporary contracts. The second method is to compute the proportion
of short-service workers (less than two year's service) in a plant's
permanent workforce. The idea here is that the lower is the short-service
proportion the more settled are incumbent workers, and the greater
is insider power (Emerson, 1988, 780, first used this measure).
8. Our employment protection variables generally
behave in the expected way. In particular, a movement from "off"
to "on" in the employment protection dummy is associated
with a two-year increase (about 15 per cent) in the previous experience
required of new recruits, holding their education constant. Moreover,
the smaller the short-service proportion (ie, the stronger are
insiders) the greater is the proportion of workers hired on a
temporary basis. The effect here is that a 10 per cent increase
in the short-service proportion means a 2 per cent decrease in
the proportion of new hires that are temporary. The effect suggests
that when insiders are strong, employers are more cautious about
recruiting on a permanent basis. Our findings, therefore, indicate
that stricter employment protection is associated with increased
hiring standards and fewer permanent jobs, which is clearly bad
for the unemployed.
9. Returning to studies based on county-wide
aggregates, that by Nickell and Layard (1997, 43) also finds a
link between employment protection laws and heightened long-term
unemployment "as firms become more cautious about hiring".
They find no link with the total unemployment rate, however, presumably
because employment protection laws also reduce firings and hence
reduce flows into unemployment. Nevertheless, they cannot gainsay
their result that the impact of stricter employment protection
laws is to worsen long-term unemployment. In short, employment
protection laws appear to concentrate unemployment, which is not
what we want at all. Such a finding is confirmed in the recent
OECD study (1999, Table 2.13), using a better employment protection
law index (one which allows for changes over time). Furthermore,
as regards youth unemployment specifically, results by Scarpetta
(1996) also indicate that stricter employment protection laws
are associated with higher youth unemployment.
10. Admittedly, as regards the temporary
employment share, the OECD study (1999, Table 2.10) does not confirm
our link with employment protection. On the other hand, for the
U.S. Autor (1999) is able to demonstrate a significant link between
increases in temporary worker hiring, and the onset of the doctrine
of "implied contractual right to continued employment"
in the States in which this occurred. Taken together, these results
broadly confirm our finding that employment protection laws make
the recruitment of the unemployed more difficult, and their jobs
11. For the UK, there has been a trend in
recent years towards stricter employment protection. In particular,
the Employment Relations Act raises the ceiling for unfair dismissal
compensation from £12,000 to £50,000, and also reverses
the 1985 change in unfair dismissal eligibilityreducing
the required period of service from two years to one. The implication
is that recruitment of the unemployed will become more difficultthough
how much more difficult is still an open question. After all,
average compensation in the UK for unfair dismissal is currently
only of the order of £3,000, compared to a figure 100 times
larger (Autor, 1999, 7) in "unregulated" California.
Evidently the UK still has some way to go. At the same time, as
Figure 1 shows, in most study plants on both sides of the Atlantic
the tendency to hire on a temporary basis is increasing, so something
is happening. It seems, therefore, that the Committee has to take
the employment protection obstacle seriously, and consider ways
of overcoming it.
12. Third, education. It is clear that extra
education would help unemployed workers, butequallythat
there are many difficulties. Such difficulties include the time
taken for education policies to take effect, or, where short-term
training programmes are contemplated, their low additionality.
(Gardiner (1997, Table 5) cites an estimate of only 3 per cent
of workers gaining a job as a result of the Training for Work
programme.) Indeed, the Committee's call for evidence does not
mention education policies. This might be a mistake, because our
research indicates that recruits' education is important for employers.
13. The importance of education is shown
by the trade-off between recruits' starting age or previous experience
and their education. Other things equal, a recruit with a year's
extra education (10 per cent more) has about two years less previous
experience (15 per cent less). We would expect education and experience
to be substitutes, but the size of the association comes as a
surprise. Over the last two decades, recruits' education has in
fact increased by about a year on average across our UK plants,
so enabling many less experienced workers to have job opportunities
from which they would previously have been ruled out. We are still
analysing movements in the education variable, to see, in particular,
what role the raising of the school-leaving age has played. In
any case, our results underline the importance of extra education
in the recruitment process, and the Committee may wish to consider
methods of improving education opportunities for the unemployed.
14. In conclusion, we have provided results
on three factors underlying the recruitment environment of unemployed
people; union power, employment protection laws, and worker education.
Using 20 years of data from plants in several countries we find
these factors to be important in determining employers' willingness
to hire unqualified workers, and by extension, unemployed workers.
According to our results, limiting trade union power and employment
protection, and improving worker education would provide the best
foundation for the recruitment of unemployed people. Of course,
such a course of action might have disbenefits for employed workers,
and in any case would run counter to Government policy. Nevertheless,
the Committee could lend its support to the voice of caution with
regard to further extension of collective power, or employment
protection (eg, via the proposed directives for informing and
consulting employees, and for regulating employment agencies).
Research into methods of improving the education qualifications
of unemployed workers would also be fruitful.
W S Siebert and Kirsten
University of Birmingham Business School
Addison John and W S Siebert, 1999, "Labour
Market Reform in the United Kingdom: From Thatcher to Blair",
unpublished paper, Birmingham: Birmingham University Department
Autor David, 1999. "Outsourcing at Will:
Unjust Dismissal Doctrine and the Growth of Temporary Help Employment",
unpublished paper, Harvard: Harvard University Department of Economics.
Emerson Michael, 1998 "Regulation or Deregulation
of the Labour Market", European Economic Review, 32:
Gardiner Karen, 1997. Bridges from Benefit
to Work: A Review, York: Joseph Rowntree Foundation.
Nickell Stephen and Richard Layard, 1997. "Labour
Market Institutions and Economic Performance", Discussion
Paper Series on the Labour Market Consequences of Technical and
Structural Change No 23, London: Centre for Economic Performance,
London School of Economics.
OECD, 1999. Employment Outlook 1999,
Paris: Organisation for Economic Co-operation and Development.
Scarpetta Stephano, 1996. "Assessing the
Role of Labour Market Policies and Institutional Settings on Unemployment:
A Cross-Country Study", OECD Economic Studies, 26:
Siebert W S, 1999. Company Recruitment Policies:
Implications for Unskilled Workers, York: Joseph Rowntree
3 We are grateful to the Joseph Rowntree Foundation
for financial support for the research and to the members of the
Foundation's Advisory Group including Gerry Makepeace, Pamela
Meadows, Paul Teasdale, Derek Williams, and Stephen Wood for their
advice. However, we are solely responsible for interpretation
of the results. Back