Content uploaded by Sanford M Jacoby
Author content
All content in this area was uploaded by Sanford M Jacoby on Apr 29, 2017
Content may be subject to copyright.
American Economic Association
Sticky Stories: Economic Explanations of Employment and Wage Rigidity
Author(s): Sanford M. Jacoby and Daniel J. B. Mitchell
Source:
The American Economic Review,
Vol. 80, No. 2, Papers and Proceedings of the
Hundred and Second Annual Meeting of the American Economic Association (May, 1990),
pp. 33-37
Published by: American Economic Association
Stable URL: http://www.jstor.org/stable/2006538
Accessed: 29-04-2017 17:01 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
American Economic Association
is collaborating with JSTOR to digitize, preserve and extend access to
The
American Economic Review
This content downloaded from 128.97.27.20 on Sat, 29 Apr 2017 17:01:48 UTC
All use subject to http://about.jstor.org/terms
Sticky Stories: Economic Explanations
of Employment and Wage Rigidity
By SANFORD M. JACOBY AND DANIEL J. B. MITCHELL*
There was a time when "institutional" la-
bor economists and economic theorists liter-
ally parted company over methodology.
Specifically, in the late 1940s labor econo-
mists formed their own Industrial Relations
Research Association (IRRA) as a rump
group of the AEA. A notable feature of the
IRRA was that its membership was drawn
from both academics and labor-management
practitioners. The idea was that practition-
ers, who knew the facts of the labor market,
would inform academic researchers.
Knowledge of the stylized facts of the
labor market may once have been disdained
by mainstream economists, especially if these
observations seemed to contradict theoreti-
cal expectations. At the same time, rigorous
analytical and empirical approaches were of-
ten omitted from the institutionalists' tool
kit. But in recent years there has developed a
"new economics of personnel" (NEP) that
bases its analysis on certain potentially un-
comfortable facts, yet brings to bear a high
level of rigor. Some (not all!) of the older
doyens of labor market institutionalism have
even taken pleasure in the "new generation
of exceedingly competent labor economists"
(Clark Kerr, 1983, p. 21). An important
question, therefore, is whether the gap be-
tween institutional knowledge and theoreti-
cal analysis has now been bridged. We are
encouraged by the bridging attempt, but ar-
gue that a gap remains due in part to a
tendency for the stylized facts to shift.
I. Time and Space
The NEP offers efficiency explanations
for modern American employment practices
such as career labor markets, wage rigidity,
and seniority-based layoffs. Its reasoning is
functional; the practices must be efficient or
else firms pursuing alternative policies would
become dominant. The explanations prof-
fered are derived from analysis of rational
actors in what often seems to be a timeless,
placeless steady state. Unfortunately, that
mode of analysis cannot tell us what specif-
ically is modern or American about current
U.S. employment practices. This void would
not be a problem were there but negligible
national or historical variation in employer
policy. However, the evidence shows such
variation to be substantial.
American employment practices have
rarely matched the textbook model of a spot
labor market. But, over the course of this
century, those practices went through a tran-
sition. Employment relationships became
markedly more stable, while internal labor
markets and seniority-layoff systems prolif-
erated (see Jacoby, 1983). Similarly, nominal
wages in the United States never were com-
pletely flexible. But, during the 1910s and
1920s, nominal wage cuts were more preva-
lent (and wage sensitivity to labor market
conditions was greater) than in the post-
World War II period (see Mitchell, 1985).
Europe and Japan did not go through
the same transitions. The relative reliance
on layoffs, nominal wage change, and real
wage change varies from country to country.
These time-and-space departures from U.S.
arrangements suggest that rationalizations
based on current American practices cannot
be the basis of universal theory.
Confronted with the specificity of their
stylized facts, NEP theorists have three
responses. One is to argue that observed
variations are transitory disequilibria; even-
tually employers will shift from less to more
efficient employment practices. Oliver
Williamson (1985) makes precisely this argu-
ment for the American historical case, infer-
*Anderson Graduate School of Management, UCLA,
Los Angeles, CA 90024.
33
This content downloaded from 128.97.27.20 on Sat, 29 Apr 2017 17:01:48 UTC
All use subject to http://about.jstor.org/terms
34 AEA PAPERS AND PROCEEDINGS MA Y 1990
ring that present practices displaced earlier
ones because of superior efficiency. But the
inference depends on other assumptions. One
is that market forces are always strong
enough to weed out inefficiency. A second is
that there is one right way to manage, that
is, that multiple efficiency points do not ex-
ist.
Comparative observation suggests that
multiple equilibria within a market setting
are possible. On the one hand, critics of
"Eurosclerosis" in the 1980s pointed to the
American model of labor market " flexibility"
(easy layoffs) as proof that European long-
term job security was undesirable. At the
same time, American management experts
journeyed to Japan to marvel at long-term
employment commitment and the ability it
gave management to move workers to needed
tasks.
A third response is to argue that any de-
parture from current U.S. practices such as
wage rigidity or career ladders must be due
to variations in the incentives (for example,
risk aversion, monitoring costs) that make
those practices efficient. But historical re-
search shows that, in the past, these incen-
tives at least needed the push of social forces
and historical accident (unions, wars, the
Great Depression) to put now-existing prac-
tices into place. At the international level,
economic rationality and technology (from
which incentive structures should spring)
are similar in modern industrial nations. So
economic incentives can only be part
of the story of historical and national varia-
tion in employer policies. The NEP either
abstracts from such observations or, less of-
ten, stretches economic theory to endogenize
them. A more fruitful approach would be to
examine the interplay between institutions,
incentives, and market outcomes, even if it
means building partial models and acknowl-
edging a limited ability to forecast.
II. Nonns of Fairness
The institutionalists of the 1940s put
great emphasis on notions of fairness in the
labor market. For example, pattern bargain-
ing in the union sector was seen as based on
worker views of comparative equity. In con-
TABLE 1-FORMAT OF REPORTED MAJOR
PRIVATE UNION WAGE ADJUSTMENTS
1959 1969 1979 1989
Nominal 69 83 80 68
Percent 12 12 19 25
Mixeda 15 2 1 1
Zero 2 0 0 6
Other 3 3 b 0
Source: Current Wage Developments, June issues of year
shown.
Note: Percent in each category. Details need not sum to
100 due to rounding.
a Combined nominal and percentage description.
bLess than 0.5 percent.
trast, NEP analysts would probably prefer to
focus on reducing transactions costs in bar-
gaining as a rational explanation for pat-
terns. Yet there is other evidence that norms
of fairness are important in labor markets.
Moreover, these norms often seem to involve
nominal-not real-measures.
Consider, for example, the annoying (to
economists) tendency for union wage settle-
ments to be reported in cents-per-hour terms
rather than as percentage adjustments. Table
1 shows the proportion of private wage ad-
justments reported in the June issues of Cur-
rent Wage Developments over the past three
decades. Despite intervening episodes of sig-
nificant inflation and wage guidelines and
controls that emphasized percentages, the
tenacity of wage setters in clinging to nomi-
nal expression of their settlements is note-
worthy. The use of nominal expression sug-
gests that when intersettlement comparisons
are made for equity purposes, base wages
matter little. That is, if two units receive 30
cent increases, workers in both feel that eq-
uity has been achieved even if one unit has a
higher average wage (and therefore receives
a smaller percentage increase). The tendency
of escalator formulas to be specified in nom-
inal terms (typically a cents-per-hour in-
crease per index point increase) is a symp-
tom of the nominal orientation. Employees
apparently view wage relationships differ-
ently than do economists.
Survey data suggest a public aversion to
nominal wage cutting. (Daniel Kahneman
This content downloaded from 128.97.27.20 on Sat, 29 Apr 2017 17:01:48 UTC
All use subject to http://about.jstor.org/terms
VOL. 80 NO. 2 NEW ECONOMICS OF PERSONNEL 35
TABLE 2-DISTRIBUTION OF EMPLOYEES UNDER
MAJOR UNION WAGE SErrLEMENTS IN
THE CONSTRUCTION INDUSTRY
1980-81 1982 1983-84 1985-88
Decrease 0 5 13 3
Zero 0 10 41 26
0.1-1.9 3 16 16
2.0-3.9 7 10 8 27
4.0-5.9 16 7 20
6.0-7.9 J 17 6
8.0-9.9 15 19 5
10.0-11.9 21 11 7
12.0-13.9 17 4 3
14.0 &up 39 5 _
Mean
Adjustment 13.4 6.5 1.1 2.3
Source: Current Wage Developments, various issues.
Note: Multiyear figures are weighted by annual number
of employees covered. Percent in each category. Details
need not add to 100 due to rounding.
et al., 1986.) While such aversion might be
explained as a reflection of a rationally based
,implicit contract by NEP theorists, the fact
that the concern is focused on the nominal
wage (not the real wage) creates obvious
analytical problems. Table 2 presents the
distribution of union wage settlements in the
construction industry during the 1980s, a
period when the industry was especially
prone to "concession bargaining."' The dis-
tribution shows a decided downward shift in
mean adjustments beginning in 1982. Yet
the distribution bunches at exactly the
(nominal) zero level. While some workers
did receive wage cuts, zero was clearly a
resistance point for pay adjustments.
J. M. Keynes (1936, p. 14) was quite will-
ing to accept fairness as a basis for apparent
money illusion in the labor market; it formed
a major element in the General Theory. In
his view, decentralized wage setting led to a
worker focus on comparative (nominal)
wages, even if price movements eroded pur-
chasing power. Would today's NEP analysts
be willing to buy such an explanation? And
even if he or she did, what would be done
with the observation that the degree of real
vs. nominal wage rigidity seems to vary
across countries? The Keynesian vs. classical
debate over European unemployment in the
1980s suggests that worker norms of fairness
can change over time and vary across coun-
tries.
III. Changing Times
Although NEP provides a rationale for
various anomalous facts first pointed out by
labor economists of the 1940s and 1950s, it
appears that this theoretical success has come
just at the time when the facts themselves
are changing rapidly. Labor markets are
moving away from the rigidities captured by
those facts. Having rationalized a difference
between labor and product markets, NEP
analysts must now consider the possibility of
greater similarity across these markets than
existed in the past.
Impelling the move away from rigid forms
of employment are cost pressures arising
from a more open and competitive econ-
omy. One response to those pressures
was the wave of permanent layoffs that swept
American industry during the early to mid-
1980s, eliminating many previously secure
jobs. Disproportionately affected were older,
blue-collar, unionized, manufacturing work-
ers. Managers have not been exempt from
such cost-cutting moves.
A related development was the shift of
employment during the last decade to smaller
firms and establishments, where compensa-
tion is lower and personnel practices are less
formalized. To compete, large establish-
ments are using part-time employees or are
contracting work out to the burgeoning busi-
ness service industries that supply temporary
workers and other business support services.
While not all of the work being contracted
out is done at lower wage and benefit costs,
the bulk of business services are provided by
lower-status workers whose pay is less than
that of regular, in-house employees. Gen-
rally, full-timers are paid more than compa-
rable part-timers, with the latter predomi-
'Construction is presented because most contracts in
the industry do not feature escalator clauses or lump
sum bonuses that obscure actually experienced pay ad-
justments. However, distributions similar to those of
Table 2 characterized the entire union sector in the
1980s.
This content downloaded from 128.97.27.20 on Sat, 29 Apr 2017 17:01:48 UTC
All use subject to http://about.jstor.org/terms
36 A EA PAPERS AND PROCEEDINGS MA Y 1990
nantly made up of female clerical fand ser-
vice workers.
Part-time, temporary, and business service
workers comprise the contingent labor force,
each segment of which has grown more
rapidly since 1980 than the labor force as a
whole. In addition to cost factors, use of
contingents has been spurred by an unstable
and less predictable business environment,
which results from more variable product
demand, shorter product life cycles, and
flexible exchange rates. Employers are trying
to keep a lean core of employees with job
security and to smooth out demand fluctua-
tions by using contingent workers who lack
such security.
Tenure data offer a way of tracing these
shifts in employment stickiness. Table 3
shows that between 1963 and 1981, job
tenure for men declined but there was little
or no change for women or mature workers.
Because workers can change jobs without
changing employers, a more accurate gauge
of employment stickiness is tenure with a
given employer. This measure has been used
by the Bureau of Labor Statistics since 1983,
unfortunately breaking the older series. As
shown in Table 4, between 1983 and 1987,
groups experiencing sharp tenure declines
included mature men (previously a stable
group); male managers and machine opera-
tors; and female clerical and service workers.
Those outcomes reflect an expansion in con-
tingent labor and shifts of employment away
from stable jobs in large firms.
Wages also are becoming less rigid and
more sensitive to firm performance, despite
the pressures emanating from norms of fair-
ness. In 1986, profit sharing in medium-
to-large firms stood at 22 percent of full-time
employees. Lump sum bonuses, that hardly
existed before the 1980s, were in the con-
tracts of 51 percent of private, unionized,
nonconstruction workers under major agree-
ments in late 1988. There is some evidence
(not definitive as yet) that these bonuses will
prove more variable than wage rates and
that they are a response (like contingent
employment) to demand uncertainty (Chris-
topher Erickson and Andrea Ichino, 1989).
As variable pay methods spread, the kind of
TABLE 3-MEDIAN TENURE OF U.S. EMPLOYEES
ON CURRENT JOB (YEARS)
All Males All Females
Males 45-54 Females 45-54
1963 5.7 11.4 3.0 6.1
1966 5.2 11.5 2.8 5.7
1968 4.8 11.3 2.4 5.1
1973 4.6 11.5 2.8 5.9
1978 4.5 11.0 2.6 5.9
1981 4.0 11.0 2.5 5.9
Source: U.S. Bureau of Labor Statistics, Special Labor
Force Reports, various issues, and Bulletin no. 2162
(1983). Figures are for January.
TABLE 4-MEDIAN TENURE OF U.S. EMPLOYEES
WITH CURRENT EMPLOYER (YEARS)
Females Males
1983 1987 1983 1987
All 3.7 3.6 5.1 5.0
25-34 3.2 3.1 3.8 3.7
45-54 6.9 7.3 13.2 12.3
Managers 4.6 4.9 7.5 6.9
Profil. 4.7 5.0 6.1 6.6
Sales 2.5 2.5 4.7 4.4
Clerical 3.9 3.7 b b
Servicea 2.8 2.5 b b
Operators b b 5.6 5.1
Craft b b 5.5 5.5
Source: U.S. Bureau of Labor Statistics, "Job Tenure:
1983, 1987," unpublished tables. Figures are for Jan-
uary.
aExcluding protective service workers.
bNot one of the five largest occupations for that sex.
wage stickiness associated with long-term
employment is likely to be reduced.
Why has the NEP literature been slow to
recognize and acknowledge these changes?
As in other fields, scholars tend to learn a
given set of facts and stick with them. This
tendency was characteristic of experts on
Eastern Europe (who until recently insisted
that nothing was changing in that part of the
world) as well as U.S. industrial relations
scholars (who were slow to recognize the
pace of union sector shrinkage during the
1980s).
While slow response is not unique to NEP,
it is nonetheless true that the NEP's static
This content downloaded from 128.97.27.20 on Sat, 29 Apr 2017 17:01:48 UTC
All use subject to http://about.jstor.org/terms
VOL. 80 NO. 2 NEW ECONOMICS OF PERSONNEL 37
approach to theory is not well-suited for
predicting outcomes of dynamic relation-
ships, such as those between firms and their
environments. As in other branches of eco-
nomics, the NEP tends to give more empha-
sis to internal theoretical consistency than to
external consistency with a welter of change-
able facts.
IV. Where Do We Go from Here?
The NEP analysts have demonstrated
something important; almost any behavior
can be rationalized within the neoclassical
framework. While the usual failback is to use
empirical testing to sort out the true expla-
nation, such testing in practice rarely sorts
out competing views definitely. Starting from
stylized facts, if nothing else, prevents one
from saying silly things about nonexistent
phenomena. But neither institutionalists nor
NEP theorists can base their observations on
a set of stylized facts, and ignore trends
which are changing those facts, without risk-
ing obsolescence.
Some physicists reportedly feel they are
on the verge of a Theory of Everything, a
unified theory that will encompass all known
forces. But it is unlikely that such a theory
will arrive for labor market analysts anytime
soon. What we can hope for are informed
but modest partial theories and explana-
tions.
REFERENCES
Erickson, Christopher and Ichino, Andrea, " Lump
Sum Bonuses in Union Contracts," work-
ing paper, MIT, 1989.
Jacoby, Sanford M., "Industrial Labor Mobil-
ity in Historical Perspective," Industrial
Relations, Spring 1983, 22, 261-82.
Kahneman, Daniel, Knetch, Jack L. and Thaler,
Richard, " Fairness as a Constraint on Profit
Seeking: Entitlements in the Market,"
American Economic Review, September
1986, 76, 728-41.
Kerr, Clark, "A Perspective on Industrial-
Relations Research-Thirty-Six Years
Later," in Barbara D. Dennis, ed., Pro-
ceedings of the Thirty-Sixth Annual Meet-
ing, Madison: IRRA, 1984.
Keynes, John Maynard, The General Theory of
Employment, Interest, and Money, New
York: Harcourt, Brace, 1936.
Mitchell, Daniel J. B., " Wage Flexibility: Then
and Now," Industrial Relations, Spring
1985, 24, 266-79.
Williamson, Oliver E., The Economic Institu-
tions of Capitalism: Firms, Markets, Rela-
tional Contracting, New York: Free Press,
1985.
This content downloaded from 128.97.27.20 on Sat, 29 Apr 2017 17:01:48 UTC
All use subject to http://about.jstor.org/terms
