"The idea here is to redistribute the available work to more people. But
once again, the available work is not given - that is the 'lump of labour
fallacy'. The equilibrium unemployment rate is independent of hours of
work. Thus if hours are reduced and employment rises for a while, wage
pressure will soon increase and the amount of work available will have to be
reduced. Employment will revert to its former level. Thus cuts in hours
provide a poor antidote to unemployment. But they certainly provide a lower
standard of living." (page 73).
with the following quotation from the posting by Jonathan Portes:
>But David Chapman's basic point is clearly correct. If it is
>currently economic to employ three workers for 40 hours a week each,
>then a work-spreading tax - by, for example, exempting the first 10
>hours wages per employee from labour taxation - would make it relatively
>more attractive for employers to hire 4 workers at 30 hours a week
>(ignoring transitional costs,etc). Unemployment would fall. This is
>not the lump-of-labour fallacy; indeed, it is the basic neoclassical
>analysis of tax incidence, and does not rely on any assumptions about
>labour market structures. There might, under certain circumstances, be
>efficiency losses, but this would depend on other factors.
At first sight it might appear to be difficult to reconcile these two points
of view. However, Jonathan Portes is only considering the immediate effects
of the work-sharing tax (the 'partial equilibrium' to use some economic
jargon), while Layard, Nickell and Jackman are considering both the
immediate effect and the wider process of adjustment in the economy (the
'general equilibrium'). If the work-sharing tax were to be limited to a
particular industry or to a particular group of workers, Jonathan's argument
might hold. But only if the number of workers covered was small relative to
the size of the economy, so that wage bargaining was not affected.
However, when applied to all workers, the increase in wage pressure from a
decrease in working hours must be inflationary. Unless the government is
willing to accept the inflationary consequences, the unemployment rate must
return to its original level. For a given inflation rate, we therefore end
up with fewer jobs and lower output. The cut in working hours has reduced
national income.
Far from being a canard or a second order effect, this is a clear
implication of reasonable economic theory based on imperfect competition in
the labour market and is supported by a wide range of empirical evidence.
The labour market is a market like any other, although unusually subject to
institutional distortions and imperfect competition. In general,
unemployment will be stable when it is high enough to prevent leapfrogging
of wage claims, and to reconcile the desired mark-up of wages over prices by
unions with the planned mark-up of prices over wages by firms. Lower
unemployment therefore does not neccessarily require lower take-home pay,
but a lower level of wage pressure for any given level of unemployment.
How can wage pressure be reduced? First, there are many unemployed people
who do not exert any downward pressure on wages because they have been
unemployed for a long time or because they lack skills. Second,
unemployment is also affected by the search behaviour of the unemployed, and
is higher when the unemployed search less (whether because of benefits or
because of demoralization).
This suggests that any strategy for reducing unemployment has to focus on
the marginalized groups amongst the unemployed. A first step is to ensure
that all of unemployed continue to search for jobs. Any who do not find
jobs after a certain period should be offered retraining. Employers could
also be offered a subsidy to employ the long-term unemployed. As I
understand it, some of this is already government policy.
Any subsidy has to be carefully designed, since employers may choose to
reduce their recruitment of unsubsidised workers, and instead take on the
long-term unemployed. If no net jobs are created, at least some of the
subsidy is a pure deadweight loss.
However, when the unemployed search actively and have appropriate skills,
they will exercise more downward pressure on wage claims than before, and so
there will be lower unemployment in the long-run.
As mentioned earlier, while the unemployed exert downward pressure on wage
claims, real wages need not fall in order to raise unemployment.
In some sense we need to change the relationship between unemployment and
vacancies. In Britain in the 1950s and 1960s, inflation emerged when there
were roughly twice as many vacancies as unemployed. In contrast, in the
1980s boom, inflation emerged when there were only half as many vacancies as
unemployed. This means either that employers now find it much more
difficult to find workers from among the unemployed than before, or that the
unemployed are less keen on, or eligible for, the jobs offered.
Any unemployment policy has to confront this problem in some way.
Gavin Cameron
-------------------------------------------------------------
Posted to uk-policy, a service of Nexus. http://www.netnexus.org/
Hosting and email provided by new media consultants On-Line Publishing