Content uploaded by Daniel J. B. Mitchell
Author content
All content in this area was uploaded by Daniel J. B. Mitchell
Content may be subject to copyright.
U
r.!
IV
SHELF
WORKING
PAPER
SERIES
-
66
Employer
Preferences
for
Long-Term
Union
Contracts
by
Sanford
M.
Jacoby
and
Daniel
J.B.
Mitchell
January
1984
Sanford
M.
Jacoby
Assistant
Professor
Graduate
School
of
Management
U.C.L.A.
and
Daniel
J.B.
Mitchell
Professor
Graduate
School
of
Management
U.C.L.A.
and
Director,
UCLA
Institute
of
Industrial
Relations
F
L
..B
:
K
`!F
'/
Employer
Preferences
for
Long-Term
Union
Contracts
Sanford
M.
Jacoby
Assistant
Professor
Graduate
School
of Management
U.C.L.A.
and
Daniel
J.B.
Mitchell
Professor
Graduate
School
of
Management
U.C.L.A.
and
Director,
UCLA
Institute
of
Industrial
Relations
January
1984
Address
for
both
authors:
Graduate
School
of
Management
U.C.L.A.
Los
Angeles,
California
90024
Telephone:
(213)
206-6550
or
(213)
825-2505
(Jacoby)
(213)
825-4339
(Mitchell)
Outline
I.
Recent
Views
on
Contracting
.................................
p.
1
II.
The
Association
Between
Contracting
and
Wage
Rigidity
..............................................
p.
3
III.
Problems
with
the
Linkage
...............
..................
p.
4
IV.
Need
for
a
Survey
of
Management
Viewpoints
.....
............
p.
5
V.
Research
Design
.............................................
p.
6
VI.
Survey
Results
.................
...........................
p.
7
i.
Negotiations
and
Strikes
...............
p.
8
ii.
Morale
and
Administration.........................p.
9
iii.
Labor
Costs
......................................
p.
9
iv.
Overall
Costs
and
Advantages
.....
.................
p.
10
VII.
Lessons
from
the
Survey
..............................
p.
11
Footnotes
.................................................
p.
Fl
In
recent years,
the
American
industrial
relations
system
has
undergone
considerable
stress.
One
byproduct
of
a
stressful
period
is
that
old
ways
of
conducting
industrial
relations
are
being
increasingly
questioned.
The
fact
that
questions
are
raised,
however,
does
not
necessarily
mean
that
the
cli-
mate
for
change
is
receptive
to
all
suggestions.
In
this
paper
we
provide
evidence
that
the
management
community
would
strongly
oppose
recent
sugges-
tions
for
the
abandonment
of
long-term
collective
bargaining
contracts.
I.
Recent
Views
on
Contracting
There
has
been
a
substantial
interest
among
economists
in
modeling
con-
tractual
arrangements.
This
interest
stems
from
various
causes
but
particu-
larly
from
dissatisfaction
about
macro-economic
models
and
macro-economic
performance.
Macro
models,
at
least
of
the
Keynesian
variety,
have
typically
included
some
assumption
of
wage
"rigidity"
in
their
labor-market
sectors.
Wage
determination
is
not
viewed
as
behaving
in
an
auction-market
fashion,
i.e.,
wages
are
not
seen
as
painlessly
falling
or
rising
in
response
to
minor
degrees
of
excess
supply
or
demand.
Econometric
models
incorporate
empirical
evidence
of
wage
stickiness,
but
-
because
they
are empirically
based
-
they
do
not
provide
an
explanation
of
why
wage
behavior
is
what
it
is.
The
difficulty
experienced
in
trying
to
reduce
the
rate
of
inflation
in
the
1970s
and
early
1980s
has
also
provoked
an
interest
in
contracting.
De-
mand
restrictions
have
tended
to
have
a
major
initial
impact
on
real
output
rather
than
on
wage
and
price
inflation,
thus
attaching
a
considerable
cost
to
anti-inflation
exercises.1
It
has
been
argued
that
if
wages,
in
particu-
lar,
could
be
made more
responsive
to
the
degree
of
demand
pressure
(as
they
would
be
under
an
auction
process)
reducing
inflation
would
be
a
much
less
painful
experience.2
2
Two
streams
of
literature
have
emerged
from
these
interrelated
con-
cerns.
Both
streams
assume
that
labor-market
contracts
inherently
limit
the
level
of
wage
responsiveness
to
demand
and
supply
pressures.
One
stream
es-
sentially
analyzes
the
behavior
of
wages
in
the
nonunion
sector
in
which
ex-
plicit
contracts
between
employer
and
employee
are
extremely
rare.
This
lit-
erature
attempts
to
explain
why
'implicit'
(unwritten)
contracts,
which
lim-
it
wage
flexibility,
might
exist
in
such
labor
markets.3
The
other
stream
looks
at
characteristics
of
explicit
contracts
in
the
union
sector
and
seeks
to
determine
their
effects.4
In
this
paper,
we
concentrate
on
the
union
sector
with
its
explicit
con-
tracts.
Various
authors
have
suggested
that
the
development
of
the
multiyear
union
contract
explains
why
wage
responsiveness
to
demand
and
supply
has
de-
creased
historically
and
why
the
American
labor
market
is
characterized
by
nominal
wage
rigidity
in
contrast
to
other
industrialized
countries.5
Some
authors
are
content
simply
to
make
the
association
between
long-
term
union
contracts
and
contemporary
U.S.
wage
behavior
without
overtly
drawing
policy
implications.
But
others
have
argued
that
long-term
contracts
should
be
discouraged,
perhaps
by
force
of
law.6
As
part
of
the
wave
of
un-
ion
contract
concessions
that
occurred
beginning
in
1979,
the
phenomenon
of
unscheduled
re-openings
of
contracts
became
more
common.
Examples
occurred
in
automobile
manufacturing,
steel,
meatpacking,
construction,
airlines,
and
other
industries.7
Thus,
it
might
appear
that
the
management
community
was
anxious
to
abandon
the
long-standing
system
of
multiyear
agreements.
In
fact,
evidence
is
presented
below
indicating
that
the
management
community
prefers
long-term
contracting
and
would
be
decidedly
opposed
to
any
public
policies
designed
to
discourage
the
system.
3
II.
The
Association
Between
Contracting
and
Wage
Rigidity.
It
is
not
surprising
that
a
linkage
has
been
drawn
between
the
develop-
ment
of
multiyear
union
agreements
and
wage
rigidity.
First,
contracts
typ-
ically
spell
out
wage
increases
over
an
extended
period
of
time
-
two
or
three
years
-
during
which
wages
cannot
deviate
from
the
specified
terms
ex-
cept
by
mutual
agreement
of
the
parties.
To
the
extent
that
contingency
clauses
have
been
written
into
union
agreements,
they
most
always
are
cost-
of-living
escalator
clauses
which
related
wages
to
the
Consumer
Price
Index,
not
to
a
measure
of
real
economic
conditions
or
to
labor
supply
and
demand.
Thus,
unless
it
is
believed
that
the
parties
can
accurately
forecast
the
state
of
the
economy
over
a
two
or
three
year
period,
it
is
difficult
to
argue
that
wages
under
multiyear
contracts
can
be
responsive
to
short-run
economic
fluctuations.
Second,
there
does
appear
to
have
been
a
lesser
degree
of
wage
respon-
siveness
to
demand
in
the
period
after
World
War
II
than
before.8
Multiyear
agreements
were
known
before
World
War
II,
especially
in
industries
which
had
a
long
history
of
collective
bargaining.
But
the
maturing
of
the
collective
bargaining
system
in
the
new
industries
which
became
unionized
in
the
1930s
and
1940s
was
accompanied
by
a
considerable
expansion
in
the
proportion
of
workers
covered
by such
arrangements.9
Thus,
there
appears
to
be
a
rough
association
between
the
spread
of
multiyear
agreements
and
the
decline
in
wage
responsiveness.
Third,
the
wage
equation
evidence
of
the
postwar
period
suggests
that
the
once-famous
Phillips
curve
deteriorated
in
the
1960s
and
1970s.
Those
who
estimated
Phillips
curves
using
data
up
through
the
mid-1960s
were
re-
warded
with
'good'
results,
i.e.,
wage
change
seemed
responsive
to
the
rate
of
unemployment.
But
the
situation
changed
thereafter.
Declarations
that
4
the
Phillips
curve
was
an
illusion
became
common
by
the
1970s,
as
observers
pointed
to
periods
of
simultaneous
rises
in
unemployment
and
inflation.10
Since
the
1960s
appeared
to
have been
a
period
in
which
the
long-term
con-
tract
became
cemented
into'
U.S.
collective
bargaining,
the
association
be-
tween
long-term
contracting
and
wage
rigidity
might
seem
all
that
more
precise.11
Fourth,
in
other
countries
-
even
where
unionization
is
strong
-
long-
term
contracts
between
unions
and
employers
are
rare.
Annual
bargaining
rounds
are
more
typically
the
norm.
And
as
noted
earlier,
various
authors
have
found
marked
contrasts
between
foreign
and
U.S.
wage
behavior.
In
par-
ticular,
it
has
been
noted
that
foreign
wages
tended
to
respond
more
quickly
than
U.S.
wages
to
the
sudden
boosting
of
oil
prices
in
1973-74.
Wage
slug-
gishness
in
the
U.S.
appeared
to
be
convincing
evidence
that
long-term
con-
tracts
were
the
explanatory
factor.
III.
Problems
with
the
Linkage.
There
are
some
difficulties
with
the
attribution
of
wage
rigidity
in
the
U.S.
to
multiyear
union
agreements.
First,
much
of
the
American
workforce
is
nonunion.
Almost
three-fourths
of
U.S.
wage
and
salary
workers
in
1980
were
not
represented
by
a
labor
organization.12
And
the
fraction
of
the
workforce
unionized
has
been
falling
since
the
mid-1950s.
To
the
extent
that
nonunion
workers
can
be
said
to
have
any
periodicity
in
wage
determination,
it
appears
to
be
annual.
When
combined
with
the
(small)
number
of
union
workers
who
are
under
one-year
agreements,
it
is
evident
that
the
bulk
of
the
American
work-
force
is
not
directly
touched
by
multiyear
agreements.13
only
if
one
is
pre-
pared
to
buy
an
argument
that
multiyear
contracts
have
very
substantial
rip-
ple
effects
on
nonunion
wages
is it
possible
to
attribute
U.S.
wage
inflexi-
bility
to
them.
5
A
second
problem
with
the
asserted
linkage
between
long-term
union
con-
tracts
and
aggregate
inflexibility
of
wages
is
that
even
in
the
union
sec-
tor,
the
actual
pattern
of
contract
expirations
would
permit
a
much
more
ra-
pid
reduction
in
inflation
than
the
wage-equation
evidence
suggests.
As
Taylor
has
shown,
the
pattern
of
multiyear
agreements
would
retard
inf
lation
reduction
for
a
period
of
a
little
over
one
year.
Thereafter,
fairly
rapid
inflation
reductions
could
occur
as
new
negotiations
took
place.14
Taylor's
finding
that
contracts
do
not
appear
to
have
been
the
binding
constraint
in
the
union
sector
suggests
that
some
other
force
would
retard
wage
responsive-
ness
even
in
the
absence
of
multiyear
accords.
Third,
as
the
experience
since
1979
has
demonstrated,
it
is
quite
feasi-
ble
for
the
parties
to
rip
up
existing
contracts,
if
they
feel
it
is
in
their
mutual
interest
to
do
so.
Since
such
actions
are
unusual,
it
might
be
con-
cluded
that
wage
rigidity
is
generally
preferred;
the
rigidity
need
not
be
simply
the
results
of
"mistaken"
economic
forecasts.
IV.
Need
for
a
Survey
of
Management
Viewpoints.
It
is
evident
that
alternative
views
are
possible
about the
economic
ef-
fects
of
multiyear
union
contracts.
Even
so,
the
fact
that
suggestions
to
abandon
the
multiyear
contract
have
gained
currency
indicates
the
need
to
de-
termine
the
reactions
to
such
development
within
the
collective
bargaining
sector
itself.
It
seems
most
useful
to
establish
the
reactions
of
the
man-
agement
community
because
of
its
historical
role
in
pressing
for
multiyear
agreements.
The
impetus
for
a
lengthening
of
contract
duration
primarily
came
from
management
during
the
postwar
period.15
Although
the
usual
explanation
for
this
development
is
that
employers
sought
to
reduce
their
exposure
to
strike
6
risk,
there
is
no
hard
evidence
that
they
actually
obtained
any
reduction
in
strike
duration
or
frequency
between
1953
and
1980.16
However,
there
are
other
explanations
of
management's
preferences
for
multiyear
agreements.
First,
long-term
contracts
reduce
uncertainty
and
facilitate
the
planning
and
implementation
of
multiyear
projects.
Second,
there
are
fixed
costs
that
can
be
amortized
over
a
longer
period
if
contracts
expire
less
frequently.
These
include
strike
preparation
and
start-up
costs,
which
are
unrelated
to
a
strike's
duration,
as
well
as
fixed
costs
attached
to
negotiations.
Unions
also
may
save
organizational
resources
by
having
to
negotiate
less
frequent-
ly.
But
union
officials
were
reluctant
to
give
up
the
appearance
of
an
an-
nual
'delivery"
of
benefits
and
they
demanded
concessions
such
as
union
se-
curity
clauses
in
return
for
signing
longer
contracts.17
V.
Research
Design
A
national
survey
to
determine
management's
current
beliefs
and
atti-
tudes
concerning
contract
duration
issues
was
conducted
by
the
authors
during
the
first
half
of
1983.
Questionnaires
were
mailed
to
persons
with
U.S.
ad-
dresses
who
listed
their
occupation
as
'business"
in
the
Industrial
Relations
Research
Association's
1979
Membership
Directory.
Unfortunately,
the
IRRA
sample
contained
no
members
from
the
construction
industry,
a
sector
of
par-
ticular
interest
because
of
its
contracting
practices.
A
listing
of
local
directors
of
the
Associated
General
Contractors
(a
national
employers'
as-
sociation
in
the
construction
industry)
was
obtained
and
a
questionnaire
was
sent
to
each
of
them.18
Persons
who
did
not
reply
to
initial
questionnaires
were
then
sent
a
second,
identical
questionnaire.19
Data
related
to
the
sam-
ple
are
presented
in
Table
1.
6a
Table
1
Responses
to
Questionnaire
Non-construction
Construction
(IRRA
list)
(AGC
list)
Total
(1)
(2)
(3)
(a)
Number
of
Questionnaires
mailed
out
636
138
774
Questionnaires
returned
(b)
First
Wave
180
62
242
(c)
Second
Wave
66
18
84
(d)
Total
246
80
326
(d)/(a)
38.7%
58.0%
42.1%
(e)
Number who
bargained
206
78
284
(f)
Note:
Returned
as
undeliverable
131
2
133
(g)
Mean
Age
(years)
46.4
48.5
47.0
(h)
Mean
size
of
firm's
workforce
20,827
n.a.
n.a.
Ci)
Percent
with
graduate education
60.1%
25.3%
50.8%
7
The
questionnaires
asked
for
background
data
on
the
respondent
(age,
ed-
ucation,
title)
and,
for
the
IRRA
sample,
the
respondent's
organization
(size,
industry).
Except
for
those
in
the
construction
sample,
each
person
was
asked
if
his
or
her
organization
engaged
in
collective
bargaining.
Only
those
individuals
answering
in
the
affirmative
were
requested
to
fill
out
the
remainder
of
the
questionnaire.
The
respondents
then
were
asked
to
consider
the
impact
on
various
aspects
of
their
bargaining
outcomes
over
the
next
ten
years
of
a
hypothetical
law
banning
collective
bargaining
contracts
greater
than
one
year
in
duration.
Finally,
respondents
were
asked
whether
their
or-
ganizations
compiled
strike
cost
estimates
(IRRA
sample
only)
and
what
they
thought
were
the
relative
costs
of
various
contractual
arrangements.20
As
shown
on
Table
2,
the
industrial
distribution
of
the
survey
respon-
dents
was
consistent
with
the
distribution
of
major
union
contracts
in
the
private
sector.
As
compared
to
the
industrial
distribution
of
all
unionized
workers,
the
sample
underrepresents
non-manufacturing,
and
overrepresents
the
construction
sector.
However,
as
is
true
of
the
unionized
sector
generally,
most
of
the
respondents
who
engaged
in
collective
bargaining
indicated
that
their
organizations
signed
multiyear
contracts.
(See
Table
3.)
The
con-
struction
sector
had
a
slightly
higher
proportion
of
short-duration
agree-
ments,
although
this
too
is
consistent
with
national
patterns.
VI.
Survey
Results
The
chief
finding
of
the
survey
was
that
managers
were
strongly
opposed
to
a
law
that
would
ban
multiyear
contracts.
Nearly
all
agreed
that
a
ban
would
hurt
management
over
a
ten-year
period
in
various
aspects
of
labor
re-
lations.
No
statistically-significant
differences
of
opinion
were
found
on
this
issue:
Respondents
viewed
the
ban
as
harmful
regardless
of
their
firm's
7a
Table
2
Distribution
of
Private
Unionized
Workers
as
Compared
with
Respondents
All
Unionized
Major
Respondents
Workersa/
Contractsb/
to
Survey
(1)
(2)
(3)
Total
Private
Economy
100.0%
100.0%
100.0%
Non-construction
92.7
73.6
72.5
Manufacturing
32.5
45.1
46.5
Non-manufacturing
60.2
28.5
26.1
Construction
7.3
26.4
27.5
a/
Private-sector
workers
represented
by
labor
organizations,
May
1980.
b/
Agreements
covering
1,000
or
more
workers,
late
1982,
private
sector.
Source:
Column
(1)
from
U.S.
Bureau
of
Labor
Statistics,
Earnings
and
Other
Characteristics
of
Organized
Workers,
May
1980,
bulletin
2105
(Washington:
GPO,
1981),
pp.
14-15;
Column
(2)
from
William
M.
Davis,
Collective
Bargaining
in
1983:
A
Crowded
Agenda,"
Monthly
Labor
Review,
vol.
106
(January
1983),
p.
5.
Table
3
Use
of
Multiyear
Contracts
Proportion Negotiating
Number
of
Contracts
of
Duration
Respondents
Greater
than
1
Year
(1)
(2)
Total
Sample
96.8%
284
Non-construction
98.5
206
Manufacturing
99.2
132
Non-manufacturing
97.3
74
Construction
92.3
78
Larger
firmsa/
100.0
118
a/
Firms
with
5,000
or
more
employees
(non-construction).
8
size
and
industry,
or
their
own
age
and
education.
In
fact,
respondents
seemed
compelled
to
say
that
a
ban
would
hurt
management
even
in
areas
where
such
an
effect
was
not
intuitively
obvious.
i.
Negotiations
and
strikes:
There
was
nearly
unanimous
agreement
that
having
to
negotiate
on
a
yearly
basis
would
increase
the
amount
of
time
spent
on
negotiations
(Table
4).
As
one
respondent commented,
'in
bargaining
a
contract
as
complex
as
they
are
today,
you
just
could
not
ever
complete
bar-
gaining
if
you
had
only
one
year
contracts."
Another
noted
that
"We
typic-
ally
take
4+
months
to
settle
...
I
doubt
that
an
annual
contract
would
take
1/3
the
triennial
cycle.
More
likely
it
would
take
50
percent
of
the
time."
Other
comments
suggested
a
similar
view
that
bargaining
entails
a
high
fixed
negotiating
cost.
Fixed
costs,
in
this
case
associated
with
strikes,
may
also
explain
why
most
respondents
thought
that
annual
bargaining
would
increase
their
strike
costs
over
a
ten-year
period.
In
their
written
comments,
managers
pointed
to
various
costs
independent
of
strike
duration,
such
as
storing
or
moving
struck
inventory,
product
perishability
(retail
food
industry),
and
disrup-
tions
of
customer
relations.21
Note
that
about
two-thirds
of
the
non-con-
struction
firms
reported
that
they
compiled
strike
cost
estimates;
of
these,
80
percent
said
that
they
made
these
estimates
both
before
and
after
a
strike.
A
high
proportion
of
respondents
thought
that
annual
bargaining
would
increase
the
number
of
strikes.
To
support
this
view,
several
respondents
noted
that
there
was
a
greater
likelihood
of
a
strike
under
annual
bargaining
because
labor
relations
would
be
in
constant
turmoil:
"All
energy
would
be
spent
being
an
adversary,"
said
one
manager.
Another
felt
it
would
be
impos-
sible
to
establish
"a
meaningful
relationship
with
the
union"
because
'both
8a
Table
4
Expected
Impact
of
Ban
on
Multiyear
Contracts
On
Strikes
and
Negotiating
Time
Proportion
Expecting
Increase
(1)
Proportion
Expecting
Decrease
(2)
Proportion
Compiling
Strike
Cost
Estimates
Negotiations
Time
Total
Sample
94.4%
1.8%
Non-construction
95.1
1.0
Manufacturing
96.2
0
Non-manufacturing
93.2
.8
Construction
92.3
3.8
Larger
firmsa/
95.3
1.4
Cost
of
Strikes
Total
Sample
71.1%
2.5%
n.a.
Non-construction
71.4
1.0
63.1
Manufacturing
78.8
0
74.2
Non-manufacturing
58.1
2.7
43.2
Construction
70.5
6.4
n.a.
Larger
firmsa/
72.9
2.4
65.9
Number
of
Strikes
Total
Sample
73.2%
3.2%
-
Non-construction
73.3
1.5
-
Manufacturing
79.5
.8
-
Non-manufacturing
62.2
2.7
-
Construction
73.1
7.7
-
Larger
firmsa/
72.9
2.4
-
a/
Firms
with
5,000
or
nore
employees
(non-construction).
9
sides
(are)
constantly
preparing
for
contract
negotiations."
It
is
nice
to
think
that
the
"quiet
years
of
multiyear
bargaining
help
to
create
trust
be-
tween
labor
and
management.
But,
as
already
noted,
the
record
of
actual
strike
incidence
does
not
support
the
view
that
a
shift
to
shorter-duration
agreements
would
necessarily
increase
management's
"downtime"
due
to
strikes.
ii.
Morale
and
Administration:
Given
the
previous
finding,
it
is
not
surprising
that
most
respondents
believed
that
the
level
of
employee
morale
would
decrease
under
annual
bargaining
(Table
5).
One
manager
attributed
this
to
the
fact
that
during
the
period
surrounding
negotiations,
employees
"are
always
more
unsettled
and
sensitive
than
at
other
times."
Another
thought
that
its
effect
on
morale
made
the
one-year
contract
"a
productivity
'deflator'."
The
belief
that
employee
morale
is
adversely
affected
may
ex-
plain
why
most
respondents
thought
that
the
amount
of
time
spent
on
contract
administration
(e.g.,
handling
grievances)
would
increase
with
one-year
agreements.
Despite
the
assertions
of
the
respondents,
it
is
not
obvious
that
this
adverse
morale
effect
would
actually
occur;
there
could
be
fewer
grievances
filed
because
bargaining might
supplant
the
grievance
procedure
as
an
arena
for
resolving
worker
disputes.
Some
respondents
were
aware
of
this
possibil-
ity
-
a
relatively
large
minority
(as
compared
with
the
responses
to
the
morale
question)
said
that
contract
administration
time
would
decrease.
Nevertheless,
most
thought
that
in
this
area
as
in
others,
mandatory
annual
bargaining
would
produce
detrimental
effects.
iii.
Labor
Costs:
A
majority
of
managers
said
that
a
ban
on
multiyear
contracts
would
raise
wage
and
benefit
costs
(Table
6),
although
none
ex-
plained
why
they
thought
that
this
would
happen.
Perhaps
they
thought
that
management
would
offer
less
resistance
to
union
demands
under
annual
bar-
gaining
because
it
would
wish
to
avoid
incurring
the
high
fixed
costs
of
a
9a
Table
5
Expected
Effect
of
Ban
on
Multiyear
Contracts
on
Time
Spent
on
Contract
Administration
and
Employee
Morale
Proportion
Expecting
Increase
(1)
Proportion
Expecting
Decrease
(2)
Employee
Morale
Total
Sample
5.3%
68.3%
Non-construction
4.9
71.4
Manufacturing
6.8
72.0
Non-manufacturing
1.4
70.3
Construction
6.4
60.3
Larger
firmsa/
1.2
77.6
Contract
Administration
Time
Total
Sample
54.9%
10.2%
Non-construction
54.9
11.2
Manufacturing
53.8
9.8
Non-manufacturing
56.8
13.5
Construction
55.1
7.7
Larger
firmsa/
55.3
14.1
a/
Firms
with
5,000
or
mre
employees
(non-construction).
9b
Table
6
Expected
Impact
on
Labor
Costs
of
Ban
on
Multiyear
Contracts
Proportion
Proportion
Expecting
Expectng
Increase
Decrease
(1)
(2)
Wage
and
Benefit
Costs
Total
Sample
70.8%
3.5%
Non-cons
truct
ion
73.3
2.4
Manuf
acturing
73.5
3.0
Non-manufacturing
73.0
1.4
Construction
64.1
6.4
Larger
firmsa/
72.9
2.4
Predictability
of
Labor
Costs
Total
Sample
23.6%
62.3
Non-construction
22.8
61.2
Manufacturing
19.7
65.9
Non-manufacturing
28.4
52.7
Construction
25.6
65.4
Larger
firmsa/
21.2
67.1
a/
Firms
with
more
than
5,000
employees
(non-construction).
10
strike.
There
would
be
less
time
to
amortize
these
costs
than
with
multiyear
contracts.
Yet
it
is
certainly
not
self-evident
that
wages
and
benefits
would
be
higher
over
a
ten-year
period;
strikes
are
costly
to
workers
too.
Finally,
a
majority
of
respondents
expected
that
labor
costs would
be
less
predictable
under
annual
bargaining
(Table
6).
Several
respondents
noted
that
labor-cost
predictability
facilitated
long-run
planning
and
pricing:
'The
major
problem
with
one-year
contracts
is
that
the
company
can-
not
accurately
plan
or
budget
product
costs
for
product
lines
having
a
life
expectancy
exceeding
one
year.
The
tendency
would
be to
build
a
contingency
factor
into
the
pricing
of
the
product
to
insure
against
future
negotia-
tions."
This
comment
provides
a
plausible
reason
to
favor
multiyear
agree-
ments.
Thus,
it
is
unclear
why
roughly
one-fourth
of
managers
expected
their
labor
costs
to
be
more
predictable
with
one-year
agreements.
Perhaps
they
believed
that
the
uncertainty
associated
with
escalator
clauses
would
be
removed.
iv.
Overall
Costs
and
Advantages:
When
asked
which
alternative
would
be
more
costly
-
a
three-month
strike
once
every
three
years
or
a
one-month
strike
every
year
-
a
majority
of
respondents
chose
the
latter
alternative
(Table
7).
Similarly,
most
managers
said
that
unions
would
be
the
party
to
gain
most
from
a
ban
on
multiyear
contracts
(Table
7).22
Both
responses
are
consistent
with
previous
findings:
management
opinion
is
strongly
opposed
to
a
ban
on
multiyear
contracts.
In
written
comments,
this
viewpoint
came
out
as
a
repeated
charge
that
the
proposal
would
represent
an
unwarranted
govern-
ment
intrusion
into
private
affairs.
As
one
industrial
relations
executive
put
it,
"If
the
parties
want
to
slit
their
throats,
they
should
be
free
to
do
so."
lOa
Table
7
Management
Perception
of
Relative
Strike
Costs
and
Party
Likely
to
Gain
Most
From
Multiyear
Contract
Ban
Perceived
Most
Costly
Perceived
Party
Gaining
Most
Alternative:
from
Multiyear
Contract
Ban:
One-month
3-month strike
strike
every
every
3
Management
Union
year
years
(1) (2)
(3)
(4)
Total
Sample
56.3%
7.0%
4.2%
62.7%
Non-construction
59.2
7.3 1.5
64.6
Manufacturing
61.4
5.3
1.5
65.9
Non-manufacturing
55.4
10.8
1.4
62.2
Construction
48.7
6.4
11.5
57.7
Larger
firmsa/
61.2
7.6
0
68.2
a/
Firms
with
5,000
or
more
employees
(non-construction).
I1
VII.
Lessons
from
the
Survey
Despite
recent
suggestions
to
do
away
with
long-term
union
contracts,
and
despite
the
rash
of
contract
interruptions
which
erupted
after
1979,
the
management
community
would
strongly
oppose
any
attempt
to
force
a
shortening
of
contract
duration.
The
opposition
is
so
vehement,
if
the
survey
results
are
any
guide,
that
as
a
practical
matter
further
debate
over
the
issue
-
at
least
from
a
public
policy
viewpoint
-
would
be
futile.
Except
under
the
most
extraordinary
circumstances,
say
a
wartime
emergency,
no
bill
proposing
a
ban
on
long-term
contracts
could
be
enacted.
It
is
interesting
to
note
that
the
management
respondents
to
the
survey
are,
if
anything,
"oversold"
on
the
merits
of
long-term
contracts
from
the
employer
viewpoint.
Many
appear
to
believe
that
more
frequent
negotiations
under
contracts
would
produce
more
expensive
wage
and
benefit
packages;
many
also
believe
that
there
would
be
more
strikes
with
short
contracts.
But
it
is
certainly
not
evident
that
this
would
be
the
case.
It
may
be,
for
exam-
ple,
that
the
probability
of
a
strike
at
the
end
of
a
long
contract
is
higher
than
after
a
short
contract,
precisely
because
a
lengthy
period
has
elapsed
during
which
grievances
could
accumulate.
Nevertheless,
management
does
ob-
tain
a
more
lengthy
period
of
industrial
peace
with
long-term
contracts
and
has
the
opportunity
to
"plan
around"
the
potential
strike
that
may
occur
upon
expiration.
Written
comments
on
the
questionnaires
referred repeatedly
to
the
"stability"
provided
by
long-term
contracts.
Thus,
long-term
contracts
appear
to
provide
an
important
reduction
in
uncertainty
to
employers.
Given
the
strong
opposition
to
their
position,
proponents
of
bans
on
long-term
contracts
would
do
well
to
reconsider
the
proposition
that
long-
term
contracts
cause
wage
rigidity.
It
is
certainly
the
case
that
those
who
negotiate
long-term
contracts
exhibit
less
wage
responsiveness
to
business
12
cycle
pressures
than
those
who
negotiate
short
contracts.
And
it
is
also
the
case
that
nonunion
wages
are
more
responsive
to
supply
and
demand
pressures
than
union
wages.
But
there
is
no
sector
of
wage
determination
that
even
ap-
proaches
the
auction-style
wage
setting
of
the
micro-economic
textbooks.
Although
this
paper
has
focused
on
employers'
attitudes,
it
is
important
to
note
that
employees
may
well
have
preferences
for
stable
wage-determina-
tion
systems.
Where
unions
exist,
these
preferences
are
especially
reflected
in
wage
outcomes.
Within
the
union
sector,
those
best
able
to
negotiate
un-
responsive
wage
systems
find
it
most
useful
to
incorporate
them
into
long-
duration
contracts.
The
preferences
cause
the
outcomes;
the
contracts
are
merely
the
forms
by
which
preferences
are
expressed.
None
of
this
means
that
concerns
about
the
macro-economic
effects
of
wage
inflexibility
are
unfounded.
But
the
best
way
to
address
the
wage
in-
flexibility
issue
is
by
changing
preferences
themselves.
Tax
incentives
for
gain-sharing
plans
(such
as
profit
sharing,
bonus
arrangements,
etc.)
are
a
more
appropriate
means
of
inducing
wage
responsiveness
to
economic
condi-
tions.23
Such
arrangements
are
not
inherently
incompatible
with
long-term
contracts;
they
are
basically
contingency
clauses,
just
as
escalator
clauses
are,
but
they
are
linked
to
indicators
other
than
the
Consumer
Price
Index.
-Fl-
Footnotes
1.
Arthur
M.
Okun,
'Efficient
Disinflationary
Policies,"
American
Economic
Review,
vol.
68
(May
1978),
pp.
348-352.
2.
Daniel
J.B.
Mitchell,
Gain-Sharing:
An
Anti-Inflation
Reform,"
Chal-
lenge,
vol.
25
(July-August
1982),
pp.
18-25.
3.
There
are,
in
turn,
two
streams
within
the
implicit
contracting
litera-
ture.
Some
writers
emphasize
turnover
and
transactions
costs.
See,
for
example,
Arthur
M.
Okun,
Prices
and
Quantities:
A
Macroeconomic
Analysis
(Washington:
Brookings
Institution,
1981),
especially
chapters
2
and
3;
and
Michael
H.
Riordan
and
Michael
L.
Wachter,
"What
Do
Implicit
Con-
tracts
Do?"
in
Barbara
D.
Dennis,
ed.,
Proceedings
of
the
Thirty-Fifth
Annual
Meeting,
Industrial
Relations
Research
Association,
December
28-30,
1982
(Madison:
IRRA,
1983),
pp.
291-298.
Others
emphasize
risk
shifting
from
employees
to
employer.
See
Robert
E.
Hall
and
David
M.
Lilien,
'Efficient
Wage
Bargains
Under
Uncertain
Supply
and
Demand,"
American
Economic
Review,
vol.
69
(December
1979),
pp.
868-879.
4.
Daniel
J.B.
Mitchell,
"Union
Wage
Determination:
Policy
Implications
and
Outlook,"
Brookings
Papers
on
Economic
Activity
(3:1978),
pp.
537-591;
Wayne
Vroman,
"Union
Contracts
and
Money
Wage
Changes
in
U.S.
Manufac-
turing
Industries,"
Quarterly
Journal
of
Economics,
vol.
XCVII
(November
1982),
pp.
571-594.
5.
Robert
J.
Gordon,
"Why
U.S.
Wage
and
Employment
Behavior
Differs
from
that
in
Britain
and
Japan,"
Economic
Journal,
vol.
92
(March
1982),
pp.
13-44;
Jeffrey
D.
Sachs,
"Wages,
Profits,
and
Macroeconomic
Adjustment:
A
Comparative
Study,"
Brookings
Papers
on
Economic
Activity
(2:1979),
pp.
269-319;
Robert
J.
Gordon,
"A
Century
of
Evidence
on
Wage
and
Price
Stickiness
in
the
United
States,
the
United
Kingdom,
and
Japan"
in
James
Tobin,
ed.,
Macroeconomics,
Prices,
and
Quantities
(Washington:
Brook-
ings
Institution,
1983),
pp.
85-133.
Some
authors
have
simply
noted
signs
that
the
U.S.
is
characterized
by
greater
nominal
wage
rigidity
than
other
countries
without
explicitly
connecting
this
phenomenon
to
union
contracts.
See
Dennis
Grubb,
Richard
Jackman,
and
Richard
Layard,
"Wage
Rigidity
and
Unemployment
in
OECD
Countries,"
European
Economic
Review,
vol.
21
(March/April
1983),
pp.
11-39;
William
H.
Branson
and
Julio
J.
Rotemberg,
"International
Adjustment
with
Wage
Rigidity,"
Euro-
pean
Economic
Review,
vol.
13
(May
1980),
pp.
309-332.
6.
Barry
Bosworth,
"Policy
Choices
for
Controlling
Inflation,"
Alternatives
for
the
1980s
(1:1981),
a
publication
of
the
Center
for
Democratic
Poli-
cy
(now
the
Center
for
National
Policy),
p.
21;
Lester
Thurow,
"Thurow's
Third
Way"
Economist,
January
23-29,
1982,
p.
32;
Felix
Rohatyn,
"Time
for
a
Change,"
New
York
Review
of
Books,
vol.
30
(August
18,
1983),
p.
48.
See
also
the
remarks
of
Joel
Popkin
reported
in
"In-
flation
Deceleration
Seen
Leaving
CPI
Rate
at
8
Percent
for
1982,
1983,"
Daily
Labor
Report
(May
11,
1982),
p.
A-10.
Some
of
the
criticism
has
been
aimed
at
cost-of-living
escalator
clauses
which
are
linked
to
long-term
contracts.
See
Amitai
Etzioni,
"One
Way
to
Keep
Inflation
Down,"
Newsweek,
May
2,
1983,
p.
17.
Indeed,
Thurow
appears
to
believe
(incorrectly)
that
all
long-term
contracts
have
escalators.
(See
his
remarks
in
the
article
cited
above:
"without
indexed
contracts
nobody
would
sign
a
three-year
contract").
-F2-
7.
On
concession
bargaining,
see
Daniel
J.B.
Mitchell,
"Recent
Union
Con-
tract
Concessions,"
Brookings
Papers
on
Economic
Activity
(1:1982),
pp.
165-201.
8.
For
a
review
of
the
literature
and
some
independent
evidence,
see
Daniel
J.B.
Mitchell,
"Wage
Flexibility:
Then
and
Now,"
unpublished
working
paper
number
65,
UCLA
Institute
of
Industrial
Relations, December
1983.
9.
Sanford
M.
Jacoby
and
Daniel
J.B.
Mitchell,
'Development
of
Contractual
Features
of
the
Union-Management
Relationship,"
Labor
Law
Journal,
vol.
33
(August
1982),
pp.
512-518.
10.
As
Herschel
I.
Grossman
put
it,
"The
optimism
prevalent
in
the
mid-six-
ties,
associated
most
vividly
with
the
idea
of
fine
tuning
the
macro-
economy,
has
soured
in
the
face
of
recession
and
inflation
in
the
seven-
ties.'
See
his
chapter
on
"Rational
Expectations,
Business
Cycles
and
Government
Behavior"
in
Stanley
Fischer,
ed.,
Rational
Expectations
and
Economic
Policy
(Chicago:
University
of
Chicago
Press,
1980),
p.
6.
11.
Sanford
M.
Jacoby
and
Daniel
J.B.
Mitchell,
"Does
Implicit
Contracting
Explain
Explicit
Contracting?"
in
Barbara
D.
Dennis,
ed.,
Proceedings
of
the
Thirty-Fifty
Annual
Meeting,
Industrial
Relations
Research
Associa-
tion,
December
28-30, 1982
(Madison:
IRRA,
1983),
pp.
319-328.
12.
U.S.
Bureau
of
Labor
Statistics,
Earnings
and
Other
Characteristics
of
Organized
Workers,
May
1980,
bulletin
2105
(Washington:
GPO,
1981),
p.6.
13.
Less
than
1
percent
of
workers
under
"major"
private
union
contracts
were
covered
by
contracts
of
12
months
or
less
in
1980.
See
U.S.
Bureau
of
Labor
Statistics,
Characteristics
of
Major
Collective
Bargaining
Agreements,
January
1,
1980,
bulletin
2095
(Washington:
GPO,
1981),
p.
14.
However,
the
proportion
would
probably
be
somewhat
larger
were
com-
plete
data
available
on
smaller
units
and
on
the
public
sector.
14.
John
B.
Taylor,
"Union
Wage
Settlements
During
a
Disinflation,"
American
Economic
Review,
vol.
73
(December
1983),
pp.
981-993.
15.
W.S.
Woytinsky,
Labor
and
Management
Look
at
Collective
Bargaining
(New
York:
Twentieth
Century
Fund,
1949),
37-52;
Frederick
H.
Harbison,
"The
General
Motors-United
Auto
Workers
Agreement
of
1950,"
Journal
of
Poli-
tical
Economy,
vol.
58
(October
1950),
397-411;
Joseph
Garbarino,
Wage-
Policy
and
Long-Term
Contracts
(Washington,
D.C.:
Brookings
Institution,
1962),
15-19.
16.
Jacoby
and
Mitchell,
"Does
Implicit
Contracting
Explain
Explicit
Con-
tracting?",
op.
cit.,
pp.
324-325.
17.
Jacoby
and
Mitchell,
"Development
of
Contractual
Features
of
the
Un-
ion-Management
Relationship,"
op.
cit.,
p.
515.
18.
The
construction
industry
questionnaire
was
similar
to
the
IRRA
questionnaire
except
for
certain
omissions
noted
below
in
the
text.
19.
There
were
no
statistically-significant
differences
in
the
responses
to
the
initial
and
follow-up
questionnaires.
-F3-
20.
Since
those
in
the
construction
sample
did
not
bargain
on
behalf
of
their
immediate
employing
organization
(i.e.,
the
AGCs),
the
two
ques-
tions
noted
in
the
text
were
omitted.
21.
For
a
full
discussion
of
these
issues,
see
Charles
R.
Perry,
Andrew
M.
Kramer,
and
Thomas
J.
Schneider,
Operating
During
Strikes
(Philadelphia:
University
of
Pennsylvania
Industrial
Research
Unit,
1982),
pp.
101-109.
22.
Note
that
the
construction
responses
were
somewhat
different
than
manu-
facturing
on
this
question.
A
significant
minority
of
construction-
industry
respondents
thought
that
management
would
benefit
from
a
ban
on
multiyear
contracts.
As
one
construction
executive
explained,
'The
main
benefit
I
see
to
short-term
labor
agreements
is
that
an
employer
could
shed
himself
of
his
union
agreements
more
quickly
than
waiting
until
the
expiration
of
a
long-term
agreement.
While
this
might
incur
the
union's
wrath,
he
would
not
be
in
violation
of
his
agreement
by
going
open
shop."
Thus,
the
well-publicized
growth
in
nonunion
activity
in
con-
struction
appears
to
have
influenced
the
pattern
of
responses.
23.
Mitchell,
"Gain-Sharing:
An
Anti-Inflation
Reform,"
op.
cit.,
pp.
24-25.
