[II. Involuntary Unemployment]
I once had an economist friend who said that he
could not sell his house, a complaint that I reiterated
sympathetically to one of his colleagues.
The colleague responded that there was only one problem:
the house was unreasonably priced. At a
lower price the house would sell, perhaps instantly.
New Cla**ical economics views involuntary
unemployment as a logical impossibility, like
my friend's inability to sell his house. Could not
an unemployed worker obtain a job if only she
were willing to reduce her reservation wage?
The New Cla**ical answer is yes: unemployed
workers are those searching for work (hence
unemployed, rather than out of the labor force)
but rejecting jobs that are available because they
had expected better pay. The unemployed may
be unhappy that they cannot sell their labor at
the wage or salary that they would ideally like,
but except for those affected by the minimum
wage or union bargaining, they are voluntarily,
not involuntarily, unemployed. Everyone can get a
job at the market-clearing wage. In New Cla**ical
theory, periods of declining employment—businesscycle
downturns—may be caused by an unexpected
decline in aggregate demand, which
leaves workers mistakenly holding out for nominal
wages that exceed the new market-clearing
level. Alternatively, declining employment
may be due to negative supply shocks, which
cause workers to withdraw from the labor force
and eschew the jobs which are available. Any
account of the business cycle based on voluntary
variations in job-taking faces a significant
empirical difficulty—to explain why quits decline
in cyclical downturns. If higher unemployment
results from workers' rejection of the poor
returns from work, quits should rise along with
unemployment.
But there are fewer quits, not more, when unemployment rises. The procyclic behavior of quits is indisputable.
Instead of denying the very existence of involuntary unemployment, behavioral macroeconomists have provided coherent explanations.
Efficiency wage theories, which first
appeared in the 1970's and 1980's, make the
concept of involuntary unemployment meaningful.
These models posit that, for reasons
such as morale, fairness, insider power, or
asymmetric information, employers have strong
motives to pay workers more than the minimum
necessary to attract them.
Such “efficiency wages” are above market clearing, so that jobs
are rationed and some workers cannot obtain
them. These workers are involuntarily unemployed.
In the next section I will extend this reasoning to explain why involuntary unemployment varies cyclically.
The pervasive empirical finding of a wide spread of earnings for seemingly similar workers is strongly suggestive of the near ubiquity of efficiency wages. Long before the efficiency
wage was a gleam in the eye of macroeconomists,
labor economists had documented wide dispersion in earnings across seemingly similar jobs and among workers with apparently iden-tical characteristics.
an*lysis of panel data indicates that workers of the same quality receive different wages depending upon their
place of work. Moreover, data show that workers
who switch industries receive wage changes
that are correlated with the respective wage
differentials between the industries.
Industries with higher pay (conditional on characteristics)
also have lower quit rates, suggesting that pay
differences are not simply compensating differentials
due to different working conditions or benefits.
It thus appears that there are “good jobs” and “bad jobs.”
The existence of good jobs and bad jobs
makes the concept of involuntary unemployment
meaningful: unemployed workers are willing to accept, but cannot obtain, jobs identical to those currently held by workers with identical ability. At the same time, involuntarily
unemployed workers may eschew the lowerpaying
or lower-sk**ed jobs that are available.
The definition of involuntary unemployment
implicit in efficiency wage theory accords with
the facts and agrees with commonly held perceptions.
A meaningful concept of involuntary
unemployment constitutes an important first
step forward in rebuilding the foundations of
Keynesian economics.
But why do firms pay wages above rock
bottom? In my view, psychological and sociological
explanations for efficiency wages are
empirically most convincing.20 Three important
considerations are: reciprocity (gift exchange
theory from anthropology), fairness (equity theory
from psychology), and adherence to group
norms (reference group theory in sociology and
theory of group formation in psychology). In
the earliest “sociological” version of efficiency
wage theory based on gift exchange, firms give
workers above market-clearing wages and
workers reciprocate in their commitment to the
firm. The payment of above-market-clearing
wages may also be motivated by considerations
of fairness: in accordance with the psychological
theory of equity, workers may exert less
effort insofar as their wage falls short of what is
considered fair. Group norms typically determine
the conceptions workers form about how
gifts should be reciprocated and what constitutes
a fair wage. In the laboratory, Ernst Fehr
and his coauthors have established the importance
both of reciprocal behavior and social
norms for worker effort in experimental settings.
My favorite version of efficiency wages
is the insider-outsider model, whereby insider
workers prevent the firm from hiring outsiders
at a market-clearing wage lower than what the
insiders are currently receiving. This theory
implicitly a**umes that insiders have the ability
to sabotage the inclusion of new workers into a
firm. A detailed study by Donald Roy of an
Illinois machine shop reveals the dynamics by
which this may occur: In Roy's machine shop,
insiders established group norms concerning effort
and colluded to prevent the hiring of ratebusting
outside workers. Workers who produced
more than the level of output considered “fair”
were ostracized by others. Collusion by insiders
against outsiders is a compelling motive for
many firms to pay wages that are above market
clearing.
An alternative version of efficiency wage theory,
grounded in asymmetric information,
views above-market-clearing wages as a disciplinary
device. In the Shapiro-Stiglitz model,
firms pay “high” wages to reduce the incentive
of workers to shirk. The attempt of all firms to pay “above-average” wages, however, pushes
the average level of wages above market clearing,
creating unemployment. Unemployment
serves as a disciplinary device, because workers
who are caught shirking and fired for lack of
effort can become reemployed only after a period
of unemployment.
The worker-discipline model fits the standard
logic of economics more comfortably than approaches
grounded in sociology and psychology.
But sociological and psychological models, including
the insider-outsider model, that rely on
elements outside the standard economic box,
probably yield a better overall explanation for
involuntary unemployment. These behavioral
models capture Keynes' emphasis, in the initial
chapters of the General Theory, on equity and
relative wage comparisons.